Downloads 2026
Number of events: 7114
- $1+1<1$? Breaking the Standalone Barrier in Federated Fine-Tuning of Multimodal Large Language Models under Non-IID Data
- $\alpha$-UCAVI: Spectral Risk Decomposition for Nonlinear State Space Models via Fractional Variational Inference
- $\chi$-Bench: Can AI Agents Automate End-to-End, Long-Horizon, Policy-Rich Healthcare Workflows?
- $\delta$-Mem: : Efficient Online Memory For Large Language Models
- $h$-control: Training-Free Camera Control via Block-Conditional Gibbs Refinement
- $\mathbf{\mathtt{MAD\text{-}Bench}}$: How Do Multimodal Agents Deceive You?
- $\mathrm{LL}(k)$ Decision Processes
- $MemSysBench: A Reproducible Evaluation Framework for LLM Memory Systems$
- $PAS^2$: Physics-Anchored Spectral Reasoning for Air Quality Forecasting
- $\Psi$-Resilience: Model-Free Feature Importance from 1D Topological Signals
- $\textbf{MARC}$: Multi-Agent Reflective Correction for Sparse-View Articulated Object Code Generation
- 123D: Unifying Multi-Modal Autonomous Driving Data at Scale
- 1-Step Kac Flow via Koopman Linearization
- 2nd Embodied Spatial Reasoning (ESR) Workshop
- 2nd Workshop on Advances in Representation Learning for Earth Observation (REO-2)
- 2nd Workshop on Principles of Generative Modeling (PriGM)
- 3DABSeg: Adaptive 3D Ankle Bone Segmentation with Multiscale Feature Fusion Mixture-of-Experts
- 3D-Belief: Embodied Belief Inference via Generative 3D World Modeling
- 3D Gaussian Splatting for Large Indoor Environments
- 3D-MAP: Dual-Manifold Constrained Diffusion for Single-Stack Fetal 3D MRI Reconstruction
- 3D MRI Image Pretraining via Controllable 2D Slice Navigation Task
- 3D-PLOT-LLM: Part-Level Object Tokens for 3D Large Language Models
- 3D-VITAL: Visibility-aware Identity Training from Artificial Liftings of 3D Reconstruction model for Cross-View Object Re-Identification
- 3DVLA: Enhancing Vision-Language-Action Models via 3D Spatial and Instance Understanding
- 4D-GSW: 4D Gaussian Splatting Watermarking with Spatio-Temporal Consistent Learning
- 4D Parametric Future World Modeling for Dexterous Robot Planning
- 8th Robot Learning Workshop: Is Physical AI Going Zero-Shot?
- A$^2$R: Training-Free Sparse Attention for Autoregressive Video Diffusion
- A2FPO: Agent-Adaptive Fisher Preconditioned Policy Optimization for Cooperative Multi-Agent Reinforcement Learning
- AAA Feature Economy: A Sparse Autoencoder Study of Availability, Access, and Allocation in DINO and I-JEPA Representations
- A BAF-Based Multi-Agent System for Diverse and Convergent Debate
- A Benchmark for Object Physical Relationship Perception in Humans and Machines
- A Benchmark for Omni-Modal Reasoning in Long Videos
- A Benchmark for Strategic Auditee Gaming Under Continuous Compliance Monitoring
- A Bias-Variance Tradeoff Perspective for Improving Test-Time Scaling in LLM Reasoning
- A Birational Geometry of Regularization in Deep Learning
- A Black-box Optimization Framework for Evaluating Worst-case Robustness in Monocular Depth Estimation
- A Bottleneck Theory for Zero-Shot Cross-Modal Transfer
- ABRAHAM: A Scalable Generative Modeling and Tokenization Method for Self-Supervised Multi-Source Sequence-to-Sequence Remote Sensing
- A Bundle-based Augmented Lagrangian Framework: Algorithm, Convergence, and Primal-dual Principles
- AcademiClaw: When Students Set Challenges for AI Agents
- AcceleGrad#: Adaptive Geometry-Aware Acceleration
- Accelerated Image Editing via Consistency-Aware Source Token Pruning
- Accelerating Multi-Property Molecular Design via Entropic-Risk-Based Counterfactual Explanations
- Accelerating Newton-Schulz Iteration for Orthogonalization via Chebyshev-type Polynomials
- Accelerating Power Method with Fast Sketching for Stronger Low-Rank Approximation
- Accelerating Saddle Avoidance in Decentralized Optimization on Directed Graphs via Gradient Clipping
- AccentMatch: Adaptive Competence Control via External Network Teaching
- Accessing Emotion Within: Tracing Latent Affective Cues in Multimodal Large Language Models
- Accord: Validated Query Contracts for Reliable Text-to-SQL
- Accuracy vs. Accuracy: Computational Tradeoffs Between Classification Rates and Utility
- ACES: Accent Subspaces for Coupling, Explanations, and Stress-Testing in Automatic Speech Recognition
- Achieve Latency-Efficient Temporal-Coding Spiking LLMs via Discretization-Aware Conversion
- Achieve Performatively Optimal Policy for Performative Reinforcement Learning
- Achieving $\epsilon^{-2}$ Sample Complexity for Single-Loop Actor-Critic under Minimal Assumptions
- Achieving Better Local Regret Bound for Online Non-Convex Bilevel Optimization
- Achieving the Lower Bound in Decentralized Optimization via Chebyshev-Optimistic Tracking
- A Closer Look at Dynamic Scene Graph Generation in the Era of Multimodal Large Language Models
- A-CODE: Fully Atomic Protein Co-Design with Unified Multimodal Diffusion
- A Compass for Useful Data: Online Data Selection via Alignment-Gated Fisher Geometry
- A Complexity-Aware Benchmark for Opioid Use Disorder Treatment
- A Complexity Theory for Detection: Detection Order as a Fundamental Limit of Learning
- A Comprehensive View of Fairness through Distributional Stability
- A Concentration Inequality for the Covariance Matrix of an Arbitrary Subset of Random Vectors
- A Concept-Level Vision-Language Alignment for 3D CT contrastive learning
- A Conditional U-Net Pipeline with Pre- and Post-Processing for Aerial RGB-to-Thermal Image Translation
- A Connectome-Constrained Generative Model of C. elegans Brain Activity and Behavior
- A Continual-XAI Benchmark for Temporal Provenance Graphs
- A Continuous Energy Ising Solver Leveraging Difference-of-Convex Programming
- A Controlled Audit of Fixed-Grid Pre-Encoding Allocation Under Shared Token Prices
- A Controlled Audit of Planner-Visible Latent-Resolution Allocation in Fixed-Budget World-Model Planning
- A Control-Theoretic Approximation to Predictive Coding Dynamics
- AcquisitionSynthesis: Optimizing Data Generation using Acquisition Functions
- Acting Appropriately: A Methodological Framework for Evaluating Contextual Privacy in Embodied AI
- Acting without Knowing: Planning, Prediction, and Transfer Dissociate in Interactive Visual Physics
- Action at a Distance: A Universal Reproducing Kernel Hilbert Space from Polynomial Alignment and IMQ Distance
- Action Chunking Proximal Policy Optimization with Feedback Correction
- Action-Driven World Modeling: Eliciting Latent Dynamics in VLA via Next-Chunk Prediction
- Action-Graph Policies: Learning Action Co-dependencies in Multi-Agent Reinforcement Learning
- Action Language Engineering: Improving Reliability of Tool-Augmented Reasoning Through Task-Specific Grammars
- Action-Value Geometry Under Hidden Confounding: GeoSCOPE for Reliability-Oriented Off-Policy Evaluation: A Closer Look
- Activation-Aware Weight Tensorization: A Calibration-Time Preconditioner for Tensor-Network LLM Compression
- Activation Functions Shape Token Synchronization in Stochastic Transformer Dynamics
- Active Flow Expansion for Out-of-Distribution Discovery: from Theory to Molecules
- Active Learning for Gaussian Process Regression Under Self-Induced Boltzmann Weights
- Active Memory Feedback Loop for Fast–Slow Dynamics in Liquid Neural Networks
- Active Multiple-Prediction-Powered Inference
- Active-Set Projection Layers for Hard-Constrained Neural Networks
- Active-Subspace Corrections for Low-NFE Diffusion Sampling
- Active Zero: Self-Evolving Vision-Language Models through Active Visual Space Exploration
- Activity-Driven Intrinsic Plasticity Promotes Efficient Neural Representations
- Actor-Accelerated Policy Dual Averaging for Reinforcement Learning in Continuous Action Spaces
- Actor-Critic Algorithm for Dynamic Expectile and CVaR
- ActRight: Counterfactual Training for Causally-Grounded Vision-Language-Action Models
- ActWorld: From Explorable to Interactive World Model via Action-Aware Memory
- A Curvature Phase Transition Governs Coherence Penalty Efficiency Against Feature Absorption in SAEs
- ACWM-Phys: Investigating Generalized Physical Interaction in Action-Conditioned Video World Models
- AdaCom: Adaptive Heterogeneous Compression for Resource-Constrained LLM Fine-Tuning
- AdaKVQ: Adaptive Mixed-Precision KV Cache Quantization For Efficient Reasoning Models
- AdaLN-Based Distance-Augmented Link Prediction in Graph Neural Networks
- AdaMAP: Learning Adaptive Multi-Action Prediction with Grounded Dreaming Guidance
- AdaMerge: Multi-Granularity KV Cache Compression via Retain-Quantize-Evict
- Adam or Gauss-Newton? A Comparative Study In Terms of Basis Alignment and SGD Noise
- AdaMuon: Adaptive Muon Optimizer
- AdapMatch: Adaptive Bias Decoupling for Semi-Supervised Partial Label Learning under Unknown Class Distributions
- AdaPreLoRA: Adafactor Preconditioned Low-Rank Adaptation
- Adaptation-Free Heterogeneous Collaborative Perception with Unseen Agent Configurations
- ADAPT: Hybrid Prompt Optimization for LLM Feature Visualization
- Adaptive Accelerated Mirror Descent in Primal and Dual Spaces
- Adaptive Agent Selection for Multi-Stage LLM Workflows: An Online Learning Approach with Strategic Experience Sharing
- Adaptive Bayesian Partner Selection for Federated Clinical Centers
- Adaptive Communication Range for Scalable Cooperative Multi-Agent Reinforcement Learning
- Adaptive Correction for Ensuring Conservation Laws in Neural Operators
- Adaptive Covariance and Multi-Layer Alignment for Out-of-Distribution Detection
- Adaptive Distillation via Online Drift Projection for Exemplar-Free Class-Incremental Learning
- Adaptive Experimentation for Censored Survival Outcomes
- Adaptive Fine-Tuning Scheduler for Multi-Tenant Edge LLM via Convergence-Aware Bandits
- Adaptive Information Filtering for Quality-Aware Time-Series Forecasting
- Adaptive Internal Readout for Native Multimodal Models
- Adaptive Joint Testing of Policies in Discounted Markov Decision Processes
- Adaptive Kernel Patching for Irregular Time Series Forecasting
- Adaptive LLM Routing for Multi-Turn Conversations with Continuously Evolving User Queries
- Adaptive Multi-Objective Alignment for Pretraining EEG Foundation Models
- Adaptive Multi-View Ordinal Learning for EEG-Based Dementia Diagnosis in Low-Data Regimes (Extended)
- Adaptive Non-Euclidean Algorithm For Deep Learning Distributionally Robust Optimization
- Adaptive Order Policies for Masked Diffusion
- Adaptive Prior Selection in Gaussian Process Bandits with Thompson Sampling
- Adaptive Regret-Optimal Distributed Control via System-Level Synthesis
- Adaptive Routing for Quantized Mixture-of-Experts Serving with Theoretical Guarantee
- Adaptive SAE-Locked LoRA for Low-Interference Knowledge Editing
- Adaptive Scheduling Pipeline For Multi-Instance Asynchronous Reinforcement Learning
- Adaptive Selection of LoRA Components in Privacy-Preserving Federated Learning (Extended)
- Adaptive Semantic Evidence Correction for Robust Medical Image Segmentation under Weak Visual Cues
- Adaptive Stepsizes for Eligibility Traces in Deep Reinforcement Learning
- Adaptive Test Case Discovery for LLM-Assisted Decision Making in High-Stakes Domains
- Adaptive Visual Evidence-Guided Decoding for Hallucination Reduction in LVLMs
- Adapt or Forget: Provable Tradeoffs Between Adam and SGD in Nonstationary Optimization
- AdaptSplat: Adapting Vision Foundation Models for Feed-Forward 3D Gaussian Splatting
- ADA: Resolving Attribution Ambiguity in End-to-End Power System Dispatch via Two-Time-Scale Stochastic Approximation
- AdaRollout: Accelerating Synchronous RLVR with Adaptive Parallelism Switching
- AdaRS: Adaptive Reward Shaping for Robust Large Language Model Alignment
- AdaST: Adaptive Coupling for Spatial-Temporal Forecasting
- A Data-Centric Approach to Improving LLM Generalization
- A Data-Parallel Additively Preconditioned Trust Region Strategy for Physics Informed Neural Networks
- Adding Thermal Awareness to Visual Systems in Real-Time via Distilled Diffusion Models
- Add One, Take One: Count Editing with Intra-Category Coherence
- Addressing Exogenous Variability in Cooperative Multi-Agent Reinforcement Learning
- A Decision Audit of Same-Defect Salvage Under Partial Verification
- ADePT: Adaptive Depth Retrieval in PDE Foundation Models
- AdERA: Adaptive Exponent Reuse for Lossless Allgather in Sharded MoE Training
- A Detection-Theoretic Explanation of Sparse Autoencoder Dark Matter
- AD-GRPO: Associative Diversity-Aware GRPO for Multimodal Reasoning
- A Differentiable Interior-Point Method in Single Precision
- ADIS-Law: Unified Scaling Laws for Annealing-Phase Domain Injection in Large Language Models
- A Distribution Mapping Approach to Counterfactually Fair Reinforcement Learning
- ADKV: A Low-Overhead Adaptive Delta Quantization for KV Cache in LLM Inference
- ADM: Adaptive Discrete World Model for Graph-Guided Long-Horizon Reinforcement Learning
- Admissible Evidence for Fixed-Retrieval Groundedness Repair: Claim-Reversal Diagnostics with CDLI as a Case Study
- ADPAS: Adaptive Dynamic Partition Scheduling for Heterogeneous Parallel and Distributed Computing
- AdShot: Benchmarking Multimodal Large Language Models for Video Advertisement Clipping
- A Dual-Head Time-Series Generation Model for Simultaneous Language and Action Production
- AdvADD: Fast Adversarial Example Generation via Adversarial Diffusion Distillation
- Advances in Approximate Inference for Chain-Graph Logical Credal Networks
- Advancing Affordance-Grounded Creative Tool Use in Large Multimodal Models
- Advancing Machine Unlearning Evaluation Requires Rethinking Retraining
- Advantage Matching: A General Reinforcement Learning Framework for LLM Reasoning
- Adversarial Agent Collaboration for Correctness Improvements of C to Safe Rust Translation
- Adversarial Attack and Defense for Machine Learning in Statistical Physics
- Adversarial Corpus Selection to Attack Subgraph Matching based Graph Retrieval
- Adversarial Deletion Attacks on Approximate Graph Unlearning: A Cross-Family Vulnerability Audit
- Adversarial Effects on Expressibility and Trainability in Distributed Variational Quantum Algorithms
- Adversarial Training for Group-Invariant Missing Data Imputation
- AdvJudge-Zero: Binary Decision Flips in LLM-as-a-Judge via Adversarial Control Tokens
- AegisFlow: Training-free Non-myopic Path-safe Guided Flow Matching
- AEGIS: Visual Evidence Grounding and Inference Selection for Multimodal Reasoning
- AEPF: Agentic Execution Privacy Framework for Runtime Privacy Enforcement in AI Agent Graphs
- AeroChem: Closed-loop Physics-Informed State Space Modeling for Long-term Chemically-Reactive Air Quality Forecasting
- AeroTrack HardCases: A Stress Benchmark for Low-SNR Biological Weather-Radar Inference
- AF-Arena: A Multi-Dimensional Evaluation Suite for Alignment Faking
- A fast, explainable and highly accurate identification of chemical elements in molecules and materials
- A Favorable Regime Between ODE and SDE for Few-Step Diffusion Sampling
- AffectAtlas: Structured and Neuro-Symbolic Affect Modeling for Emotionally Consistent Talking Face Generation
- AffectGPT-RL: Revealing Roles of Reinforcement Learning in Open-Vocabulary Emotion Recognition
- Affective Adaptive Interaction Strategy for ASD Companion Social Robots via LLM-Driven Simulation
- Affective AI Safety: The Missing Piece in LLM Safety
- Affine-invariant Cubic Newton with Weak Learners
- Affordances Enable Partial World Modeling with LLMs
- AffordMotionCom: Affordance-Grounded Motion-Aware Image Composition
- A Finite-Iteration Theory for Asynchronous Categorical Distributional Temporal-Difference Learning
- A First Guess is Rarely the Final Answer: Learning to Search in the Traveling Salesperson Problem
- A First-Order-Logic Graph as Critic for Deep Reinforcement Learning in Systems
- A Fixed-Objective Audit of Monotone Recency in Timestamped Multimodal Replay
- A Formal Kinetic Theory for Zeroth-Order Newton Dynamics: Stein-Corrected Hessian Estimation and Curvature--Variance Trade-offs
- A Framework for Analyzing DNN Representation Quality: Dynamics of Interaction Generalizability
- A Framework for Studying Adaptation Across Tasks in LLM Agents
- AFUN: Towards an Affordance Foundation Model for Functional Understanding
- A Gauge-Fixed Sinkhorn Algorithm for Semi-Relaxed Optimal Transport
- AGC: Adaptive Geodesic Correction for Adversarial Robustness on Vision-Language Models
- A Generative Model of Contextual Integrity: Appropriate vs. Inappropriate Information Sharing
- Agent$^2$ RL-Bench: Can LLM Agents Engineer Agentic RL Post-Training?
- AgentEscapeBench: Evaluating Out-of-Domain Tool-Grounded Reasoning in LLM Agents
- Agent Explorative Policy Optimization for Agentic Multimodal Reasoning
- AgentFugue: Agent Scaling for Long-Horizon Tasks through Collective Reasoning
- AgentGrad: Intervention-guided Prompt Optimization for Multi Agent Systems
- Agent Harnesses as Optimization Algorithms: Manifold-Aligned Decomposition for Constrained Generation
- Agentic Abstention: Do Agents Know When to Stop Instead of Act?
- Agentic AI-Empowered Dynamic Survey Framework
- Agentic AI for Biological Discovery: Toward Closed-Loop Life-Science Intelligence
- Agentic AIs Are the Missing Paradigm for Out-of-Distribution Generalization in Foundation Models
- Agentic Bayesian Data Analysis: From Scientific Questions to Validated Probabilistic Models
- Agentic-DPO: From Imitation to Agentic Policy Optimization on Expert Trajectories
- Agentick: A Unified Benchmark for General Sequential Decision-Making Agents
- Agentic Memory Engineering: Learning Programmatic Memory for Web Agents
- AgenticOCR: Parsing Only What You Need for Efficient Retrieval-Augmented Generation
- Agentic ODE Discovery and Parameter Distribution Inference from Summary Statistics for Rare Diseases
- AgenticOS: Co-designing Systems and ML Foundations of an OS Layer for Agentic AI
- Agentic PCG: Procedural Content Generation via Tool-using LLMs
- Agentic Planning with Reasoning for Image Styling via Offline RL
- Agentic Procurement: Active Learning and Surplus Split under a Probing Fee
- Agentic Systems for Molecular Sciences
- Agentic Trajectory Reasoning in Closed-Loop Text-to-Image Generation
- Agentic Transformers Provably Learn to Search via Reinforcement Learning
- AgenticVBench: Can AI Agents Complete Real-World Video Production Tasks?
- AgentJudge: Conformal Evaluation of General-Purpose AI Agents via Multi-Modal Evidence Fusion, Six-Dimensional Scoring, and Dual-Track Adaptive Routing
- Agent MechSuits : Mechanistic Subspace Safety Steering for Multi-Turn CLI Agents
- Agent Meltdowns: The Road to Hell Is Paved with Helpful Agents
- AgentPSO: Evolving Agent Reasoning Skill via Multi-agent Particle Swarm Optimization
- AgentPulse: A Continuous Multi-Signal Framework for Evaluating AI Agents in Deployment
- AgenTracer-v2: Agentic Failure Tracer for LLM Agentic Systems
- Agent-R: Adaptive Multi-Agent Coordination for Recommender Systems
- AgentReID: Lightweight Agentic Routing for Multi-Modal Video Person Re-Identification
- AgentSkiller: Scaling Generalist Agent Intelligence through Semantically Integrated Cross-Domain Data Synthesis
- AgentSSL: Can MLE Agents Leverage Unlabeled Data?
- Agents That Cannot Undo: Agentic AI Needs Null Results from Irreversible Physical Decision Domains
- Agent-ToM: Learning to Monitor Autonomous LLM Agents via Theory-of-Mind Reasoning
- AgentVista: Evaluating Multimodal Agent in Ultra-Challenging Realistic Visual Scenarios
- Age of Learning: Modeling Temporal Persistence of Errors in Machine Learning
- A Geometric Regularization Strategy for Efficient Training of Diffusion Models without Visual Encoders
- A Geometric Theory of Self-supervised Contrastive Learning on Manifolds
- A Geometric Unification of Concept Learning with Concept Cones
- Aggregate-Use Consensus--Disagreement Coding for Federated Update Compression
- AGILE: Adaptive Interaction and Guided Upsampling for Efficient Multi-Task Dense Scene Understanding
- AGI-Occam: Towards a Parsimonious Self-Growing Autonomous Agent
- AGL‑KT: Aligning Graphs and Language for Cognitive Scaffolding in Explainable Knowledge Tracing
- Agnostic Language Identification and Generation
- A Graph Foundation Model with Spectral Parsing and Prototype-Guided Spatial Propagation
- Agreement without Coverage: Conditional Agreement is Non-Identifying in Variable-Support Structured Extraction
- Agree to Disagree: Multimodal Autonomous Negotiation and Calibration for Entity Representation Learning
- AgroCoT: A Chain-of-Thought Benchmark for Evaluating Reasoning in Vision-Language Models for Agriculture
- A Hebbian Recurrent Neural Network Explains the Hierarchical Geometry of Sequence Memory
- A High-Frequency Multi-Station Surface Water Quality Dataset and Mask-View Augmentation Benchmark for Time-Series Imputation
- AHPA: Adaptive Hierarchical Prior Alignment for Diffusion Transformers
- A Hyperbolic Reinforcement Learning Formulation for Large-Scale Traffic Signal Control
- AI4Mat-NeurIPS-2026: NeurIPS-2026 Workshop on AI for Accelerated Materials Design
- AI Agents Are Making Knowledge Workers Busier, Not Freer
- AI Agents Are Not Ready to be Agents
- AI Agents for Biomedical Imaging and Multimodal Clinical Data
- AI Agents May Always Fall for Prompt Injections
- AI Agents Need Market Competence, Not Just Task Competence
- AI Alignment Can Build Moral Autonomy
- AI Alignment via Incentives and Correction
- AI and Science: Evolution or Extinction?
- AI and the Self: Human Identity, Authenticity, and Agency in the Age of AI
- AI at Scale for Clinical Impact (ASCI): Cancer Pathology Foundation Models
- AI Companies Should Disclose Failure Rates
- AIDaR: AI Data Readiness for Scientific Discovery
- AIdoc: A Framework for Generating and Identifying AI-Edited Documents.
- AI evaluation should measure verification cost, not correctness alone: A Closer Look
- AI Evaluation Should Require Standardized Item-Level Data Releases
- AIFF: Adaptive Continuous Flow Fields via Spectral Convolutions for 2D Field Modeling
- AI-Figures: A Fine-grained Task-oriented Dataset for Multimodal Scientific Literature Understanding
- AI for Chip Design
- AI for Drug Discovery: Bridging the Translation Gap
- AI for Meta-Science: Scaling and Organizing Science in the Age of AI Scientists
- AI for Peace
- AI for science needs measurement stability, not causality
- AI for Science: Verification in the Age of AI Scientists
- AI for Stochastic Dynamics: From Theoretical Foundations to Scientific Applications
- AI for Verifiable Coding: Human-Aligned Collaborative Agents for Autoformalization, Proofs, and Heuristics
- AI Foundations for Power Grids: From Models to Deployment at Scale
- AI-Generated Content Should Be Evaluated by Its Substance, Not Its Source
- AI Labor Rubric: Evaluating Multi-Agent Research Systems by Verified Marginal Value
- Aim Before You Generate: Token-Aware Attention Correction for Prompt Following
- AIRA-Compose: Agentic Discovery of Neural Architectures
- AirTrace: Weather-Proxy Sequence Modeling and Wind-Conditioned Message Passing for Regional PM2.5 Classification
- AI Security Pipelines Need Reports That Say How to Exploit
- AI Swarm Intelligence Must be Studied as a Sociotechnical Phenomenon
- AI World Modeling: scientific principles and primitives
- A Kernel Nonconformity Score for Multivariate Conformal Prediction
- A Large-Scale Multi-Source Dataset Linking Hacker Community Discourse to the CVE Vulnerability Lifecycle
- A Latent-Variable Estimator for Individualized Counterfactual Distributions under Sequential Treatments
- A Latent World-Action Model with Jointly Aligned Reasoning
- A Layout-Aware Grounding Model with Decoupled Instruction Parsing for GUI Agents
- A Learnable Regularized Nonlocal Functional Minimization Method for NLOS Imaging
- ALER: Adaptive Learnable Experience Rewriting for Reinforcement Learning
- Algebraic Machine Learning: learning as computing subsets of the subdirect decomposition from Abstract Algebra
- Aligned but Fragile: Enhancing LLM Safety Robustness via Zeroth-Order Optimization
- Aligned Cut-Shift Consistency for Seam-Local Robustness in Cylindrical Unfolding
- Aligning Flow Map Policies with Optimal $Q$-Guidance
- Aligning Forest and Trees in Images & Long Captions for Visually Grounded Understanding
- Aligning MLLMs with the Latent Structure of Human Cognition via Behavior-Derived Semantic Dimensions
- Alignment Is Not Enough for Safe Medical LLM Evaluation
- A Living In-the-Wild Benchmark for Measuring the Static-Benchmark Gap in AI-Generated Video Detection
- All Counterfactuals Are Wrong, and Few Are Useful
- All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- AlloGen: Conformation-Selective Binder Design with Differential State Scoring
- “Allow” to Achieve, Over-Privileged Inadvertently: The Unintended Cost of Task-Completion-Driven Pop-up Decisions in Mobile GUI Agents
- Alloy Agents Can Be More Dangerous Than Either Model Alone
- All-Parameter Inverse Optimization for Diffusion Curves
- A Local Geometric Analysis of Maximal Coding Rate Reduction via Error Bounds
- A Locally Tokenized Generative Model for Robust Time-Series Watermarking
- Alphabet of the Deep: From Zipfian Sperm Whale Click Counts to Convergent Learned Tokens
- AlphaMind: A Closed-Loop LLM Pipeline for Automated Alpha Factor Discovery
- AlphaQ: Calibration-Free Bit Allocation for Mixture-of-Experts Quantization
- AlphaSkill: Agentic Reinforcement Learning with Self-evolving Skills
- Alt-Mirage: An AI Director for Multi-Agent Social Deduction with Bounded Soft Control
- A Margin Perspective on LoRA: Robustness to Catastrophic Forgetting and Adapter Merging (MaLoRA): A Closer Look
- AMARIS: Merging Generalist and Specialist LLMs via Adaptive Subspace Inheritance
- A Matched-Control Study of Controller-Target Mismatch for Low-Entropy Late Failures in Exact-Verifier LM Pipelines
- A Matter of Interest: Understanding Interestingness Judgments of Math Problems in Humans and Language Models
- A Matter of TASTE: Improving Coverage and Difficulty of Agent Benchmarks
- AmbientEye: A Dataset for Pupil Segmentation under Natural Ambient Infrared Illumination
- AmbientFM: A Foundation Model for Ambient Sensing
- AmbiguityBench: A Decision-Theoretic Diagnostic Suite for Evaluating LLM Behavior Under Uncertainty
- AMDBench: A Progressive Benchmark for MLLMs across Analog, Digital, and Mixed-Signal Circuits
- A Mechanistic Investigation of Theory of Mind in a Large Language Model
- A Mechanistic Study of Tabular Foundation Models
- AMIGO: Adaptive Multilingual Instruction Tuning via Gradient Similarity and Budgeted Optimization (Extended)
- A Mixed-Signal Chua Optimizer for Edge Adapter Learning
- A Modality Balance Mechanism-Empowered Large Language Model Framework for Multivariate Time Series Forecasting
- AmongUs-X: Benchmarking Strategic Deception in LLMs via Theory-of-Mind Metrics
- A Monte-Carlo HJ Reachability Sampling Scheme
- Amortized Guidance for Image Inpainting with Pretrained Diffusion Models
- Amortized Molecular Optimization via Group Relative Policy Optimization
- Amortized Physics-Informed Learning via Generative Initialization of Radial Basis Functions
- Amortized-Precision Quantization for Early-Exit Vision Transformers
- Amortized Weak Gradient Estimation from Scalar Function Evaluations
- Amortizing Causal Sensitivity Analysis via Prior Data-Fitted Networks
- Amortizing Federated Adaptation: Hypernetwork Driven LoRA for Personalized Foundation Models
- A Motor Imagery Classification Method Based on Dynamic Channel Correlation Refinement and Confident Denoising Domain Adaptation
- Amplification Atlas: Why Ablation-Sensitive Layers Are Not the Best Targets for Constructive Intervention in LLMs
- Amplitude-Phase Consistent Frequency Interaction for Low-Light Image Enhancement
- A Multi-agent Framework Enabling End-to-end Finite Element Analysis for Solid Mechanics Problems
- A Multi-Institutional Multimodal EEG Benchmark for Foundation Model Generalization and Early Neurological Diagnosis
- A Multimodal Dataset of Naturalistic Emotional Expression Across Neurodiverse Populations
- AMUSE: Anytime Muon with Stable Gradient Evaluation
- An $\alpha$-Law for Scale-Optimizing Post-Training Quantization:\\Methodological Characterization and 70B Validation
- An AGM Theorem for Offline Safe RL with Stochastic Action Delay and Mode Switching: A Conditional Decomposition and Assumption-Level Diagnostics
- Analogical Trajectory Transfer
- Analysis of the Required Number of Measurements for Matrix Recovery with Prior Information Based on Degrees of Freedom
- Analytical Correction for Subsampling Bias in Drifting Models
- Analyzing Creative Capabilities in RLVR Models: Overconfidence and High-Entropy Segments
- Analyzing Mask-Breaking Mechanisms: A Methodology for Deep Learning Side-channel Analysis
- Analyzing the Impact of Data Heterogeneity and Data Repair on Fairness in Federated Learning
- An Asset Foundation Model for Industrial Asset Performance Management
- An Asynchronous Multi-Agent Framework for Adaptive Tool-calling Data Synthesis
- Anatomical Consistency Enables Efficient Diffusion Inference for MRI Reconstruction
- An Automatically Constructed Dataset of Editable Indoor Scenes with High-Quality Material and Structured Lighting
- AnchorDiff: Training-Free Concept Grounding for MM-DiTs via Anchor-Based Graph Propagation
- Anchored Confabulation: Partial Evidence Non-Monotonically Amplifies Confident Hallucination in Large Language Models
- Anchored, Not Merely Stored: A Predictive Score for In-Context Control
- Anchored Transfer for Matrix Estimation under Expanding Ambient and Representation Spaces
- Anchoring In Replay: Towards Consistent Online Class-Incremental Learning with Pretrained Models
- Anchoring Reasoning Distillation via Syntactic Constraints
- Anchoring Refusal Direction: Mitigating Safety Risks in Tuning via Projection Constraint
- Anchoring the Eigengap: Cross-Modal Spectral Stabilization for Sample-Efficient Representation Learning
- Anchoring the Real, Adapting the Fake: Asymmetric Adaptive Modeling for Face Forgery Detection
- AnchorSIPS: A Synthetic Dataset and Evaluation Resource for Evidence-Supported Psychosis-Risk Symptom Measurement
- ANCORA: Learning to Question via Manifold-Anchored Self-Play for Verifiable Reasoning
- An Economic Framework for Generative Engines: Advertising or Subscription?
- An Elastic Shape Variational Autoencoder for Skeleton Pose Trajectories
- An Empirical Comparison of ECG Foundation Models Across Modalities
- An Empirical Study on Noisy Data and LLM Pretraining Loss Divergence
- An Equivariance Principle for Optimizer Design: Symmetry-Compatible Updates for Embeddings, LM Heads, and MoE Routers
- A Neuropsychologically Grounded Evaluation of LLM Cognitive Abilities
- An Evaluation Framework for Attention-Based Agents
- A New Subspace-Based Trust Region Policy Optimization Method
- An generative-AI framework for target-Specific MicroRNAs towards RNAi-based drug design
- Angular Networks: Low-Bit Learning from Randomized Similarity Estimators
- An HTS-derived AI evaluation dataset with realistic virtual screening space and non-trivial class separation
- Animation2Code: Evaluating Temporal Visual Reasoning in Video-to-Code Generation
- AnimationBench: Are Video Models Good at Character-Centric Animation?
- An Individual-Level Stability Framework for Evaluating Clinical Risk Prediction Models
- An Informational Curse of Horizon in Offline GCRL
- An Information-theoretic Framework for Auditing Unfairness in Training Data
- An Interpretable Latency Model for Speculative Decoding in LLM Serving
- An Investigation of Robustness of LLMs in Mathematical Reasoning: Benchmarking with Mathematically-Equivalent Transformation of Advanced Mathematical Problems
- An LLM-Native Instrument for Self-Description: A 100-Item Scale, a Behavioral Dataset Across 25 Models, and a Diagnostic for LLM-as-Judge Validation
- AnnotateAnything: Automatic Annotation of 3D Assets for Robot Manipulation
- Anomaly Detection in Particle Physics Data with Denoising Diffusion Models
- AnomalyRex: Comparative Reasoning for Generalist One-Shot Industrial Anomaly Detection
- An Online-learning Spiking Neural Network for Robust Gas Concentration Regression
- An Operator Approximation Theorem for Deep Q-Learning
- Answering Counterfactual Queries on Graph Datasets
- Antidote: Reliable Model Evaluations Beyond Arena-Style Voting (Extended)
- Any3D: Pushing Native 3D Generation Models towards Image Distribution
- AnyEvo: Toward Open-Ended Algorithm Discovery via Interactive Plan and Program Coevolution
- AnyGroundBench: A Specialized-Domain Benchmark for Video Grounding in Vision-Language Models
- Anytime Pretraining: Horizon-Free Learning-Rate Schedules with Weight Averaging
- Anytime-Valid Sequential Parameter Inference on Network Using E-Value Sharing
- A One-Step Neural FBSDE Solver for Mean-Field Sovereign Default Equilibria
- ApET-V: Approximation-Error Guided Token Compression for Efficient Video VLMs
- APEX: Amplitude Anchors and Phase Priors for Target-Scarce Higher-Frequency Wave Prediction
- APEX: Autonomous Policy Exploration for Self-Evolving LLM Agents
- A Phase Transition Theory of Geometry-Dependent In-Context Learning Dynamics
- A Physics-Informed Multimodal Framework for Pediatric Short Stature: Integrating Bone Age Assessment and Variant Prioritization
- APOLLO: On efficiency of localized learning and zeroth-order optimization for Large Model pre-training
- A Position on Causal Representation Learning Without Intervention - EEG as a Case Study
- APO: Unsupervised Atomic Policy Optimization for 3D Structure Prediction of Atomic Systems
- APPO: Agentic Procedural Policy Optimization
- Approximate Algorithms for the Chamfer Distance Under Translation
- Approximate Matrix–Vectors Under a Bounded $\ell_1$ Assumption and Applications to Kernel Matrices
- Approximate Policy Iteration for Zero-Sum POSGs
- Approximation Algorithms for GPU Pricing under Finite Capacity
- Approximation and Learning in Network Security Games with Contagion and Uncertainty
- Approximation of Maximally Monotone Operators : A Graph Convergence Perspective
- A Preference-Moment Theory of Direct Preference Optimization
- A Primal-dual Approach for Semi-Infinitely Constrained Reinforcement Learning
- A prism hierarchy of learning regimes in large linear autoencoders
- A Probabilistic Framework for Multimodal Open-Set Test-Time Adaptation
- A probabilistic model of visual segmentation explains early visual cortical dynamics
- A proximal gradient algorithm for composite log-concave sampling
- AptaBench: A Benchmark for Aptamer-Small Molecule Binding
- APTUS: Adaptive Personalization for Tracer-specific Uptake Synthesis
- AQBENCH: Benchmarking Neural Surrogates for Air Quality Forecasting
- AR1-ZO: Topology-Aware Rank-1 Zeroth-Order Queries for High-Rank LoRA Fine-Tuning
- Arbitrarily Conditioned Hierarchical Flows for Spatiotemporal Events
- Arbor: Explicit Geometric Conditioning for Controllable 3D Asset Generation
- Arbor: Tree Search as a Cognition Layer for Autonomous Agents
- ARCADE: Arabic Reasoning for Causality Across Dialectal Expressions
- ArcFlow: Unleashing 2-Step Text-to-Image Generation via High-Precision Non-Linear Flow Distillation
- Architecture-agnostic Lipschitz-constant Bayesian header and its application to resolve semantically proximal classification errors with vision transformers
- ARC-STAR: Auditable Post-Hoc Correction for PDE Foundation Models
- ARDrive: Adaptive Retrieval-Augmented Generation for Autonomous Driving
- AReaL-Hex: Accommodating Asynchronous RL Training over Heterogeneous GPUs
- A Recoverability View of Layer-wise Approximation in Deep Architectures
- A Regime Theory of Controller Class Selection for LLM Action Decisions
- Are Graph Attention Networks Able to Model Structural Information?
- A Regret Perspective on Online Multiple Testing
- Are Hypervectors Enough? Single-Call LLM Reasoning over Knowledge Graphs
- Are LLMs Smarter than Chimpanzees? Estimating Knowledge States and Action Potential of Story Characters
- Arena as Offline Reward: Efficient Fine-Grained Preference Optimization for Diffusion Models
- A Replicated MATH Counterexample to Sub-Point Public Benchmark Tie-Breaking
- A Residual-Correction View of Spectral Optimization
- A Retained-Signal Interface for LLM Watermark Robustness under Paraphrase
- AReUReDi: Annealed Rectified Updates for Refining Discrete Flows with Multi-Objective Guidance
- A Reusable Preference Reward Model for Machine Learning Engineering Agents
- Are Watermarked Images Editable? SafeMark for Watermark-Preserving Text-Guided Image Editing
- Are We Benchmarking Fairly? Bias and Redundancy in Video Anomaly Detection Datasets
- Are We Really Benchmarking Forecasting Models? The Impact of Preprocessing on Time Series Performance
- ARGO: Asymmetric Replay Grounding with Optimistic Planning
- ARGOS: Automated Functional Safety Requirement Synthesis for Embodied AI via Attribute-Guided Combinatorial Reasoning
- Argus: Evidence Assembly for Scalable Deep Research Agents
- ARIA: A Diagnostic Framework for Music Training Data Attribution
- Aria: Variance-Controlled Dimensionality Expansion for Anomaly Detection
- ARKA: Adaptive Routing and Knowledge Assemblyfor Prompt-Efficient Long-Context Agents
- A Robust Training Method for Federated Learning with Partial Participation
- Arrow: A Foundation Model for Causal Discovery
- ART: Attention Run-time Termination for Efficient Large Language Model Decoding
- ArtCrafter: Feed-Forward Generation of Articulated 3D Object with Analytic Joint Derivation
- ARTEdit: Adaptive Region and Timestep for Image Editing
- ART: From Common Semantics to Unique Details via Decoupled Feature Interaction for RGBT Detection
- Articulation in Prime: Primitive-Based Articulated Object Understanding from a Single Casual Video
- Artificial Aphasias in Lesioned Language Models
- ASAP: Amortized Doubly-Stochastic Attention via Sliced Dual Projection
- ASAP: Assembly-Source Aligned Pseudocode Refinement For Binary Decompilation
- ASAP: Single-Step Adversarial Purification with Spliced Attention Guidance
- ASCA: An Adaptive Semantic Cues Aggregation Framework for Remote Sensing Image-Text Retrieval
- A Scalable PyTorch Abstraction for Multi-GPU Gaussian Splatting
- A Scaling Recipe for Generative World Renderer: A Closer Look
- ASCOT: Structure-Aware Low-Budget Solver Configuration for Repeated Convex Optimization
- A second order regret bound for NormalHedge
- A Semantic-Functional Fusion Framework for DNA Sequence Representation Learning
- A Semantic-Sampling Framework for Evaluating Calibration in Open-Ended Question Answering
- A Set-Sequence Model for Time Series
- ASH: Agents that Self-Hone via Embodied Learning
- ASI-Evolve: Towards AI-Driven Discovery Across the AI Research Stack
- A Similarity-Based Proximal Window KV Cache Compression Method
- A Skill Repair Framework with Self-Evolution for Cost-Effective adaptation of Web Agents in Dynamic Environments
- Asking Less, Adapting Better: Feedback-Efficient Test-Time Adaptation for Vision-and-Language Navigation
- Ask the Answer Graph: Uncertainty Estimation for Open-Ended Visual Question Answering
- ASO Atlas 2.0: Evaluating antisense oligonucleotide prediction across the preclinical pipeline
- A Spectral Framework for Closed-Form Relative Density Estimation
- ASPIRE: Anchored Subspace Projection for Incremental Rank Exploration
- ASSET2SIM: Automatically Improving Simulation Physics of Articulated Objects
- Assign and Add: A Mechanistic Study of Compositional Arithmetic
- A Stability Analysis of AdamW: Unstable Equilibria and Non-Convergence
- A Statistical Framework for Algorithmic Collective Action with Multiple Collectives
- A Statistical Theory of Gated Attention through the Lens of Hierarchical Mixture of Experts
- A Stoichiometrically Constrained Hybrid ODE–NN Core Model for Fed-Batch Fermentation
- ASTRA: Deployment-Time Rescue of Frozen RL Agents via Grounded Graph Reasoning
- ASTRA: Statistically Robust Model Selection From Cross-Validation
- A Stratified Multi-Rater Evaluation of LLM-Based Virtual Standardized Patients with a Deployed Data-Generation Platform
- ASTRIDE: Adaptive Semantic Token Routing with Hierarchical Context Distillation for Efficient Vision Transformers
- Astro: Activation-guided Structured Regularization for Outlier-Robust LLM Post-Training Quantization
- AstroAlertBench: Evaluating the Accuracy, Reasoning, and Honesty of Multimodal LLMs in Astronomical Classification
- AstroMind: A High-Fidelity Benchmark for Spacecraft Behavior Reasoning Based on Large Language Models
- A Structural Theory of Position Bias in Transformers
- A Structure-Aware Higher-Order Message Passing Framework on Walk States for Graph Classification
- Asymmetric EMA for Federated LoRA: Feed Back A, Not B
- Asymmetric Scaling Laws from Sparse Features
- Asymptotically Optimal Best Arm Identification with Fixed-Budget under Differential Privacy
- Asymptotically Optimal Learning for Parametric Prophet Inequalities
- Asymptotic Anytime-Valid Inference for Federated Learning
- AsymVLM: Asymmetric Token Pruning for Efficient Vision-Language Model Inference
- Asynchronous SGD with Markovian Data Sampling
- Async-SNN: Efficient Spike-Driven Large Language Models with Asynchronous Computing Paradigm
- AsyncVLA: Asynchronous Flow Matching for Vision-Language-Action Models
- A Systematic Evaluation of Co-folding Model Representations for Small-Molecule Learning
- A Systematic Evaluation of Molecular Mixture Behavior Prediction
- ATCG: Auditable Decision Logic via Structural Trait Molding for Combinatorial Auditing
- AT-Dec-POSG: Formalizing and Solving Adversarial Team Games with Decentralized Partial Observability
- A Temporal-Aware Self-Supervised Diffusion Model for Complex Robot Manipulation
- Athena: Execution-Consistent Graph Imitation for Constrained Edge–Cloud DNN Inference
- ATHENA: Risk-Sensitive Local Scheduling for SLO-Compliant LLM Inference
- A Theoretical Analysis of amortized Gaussian Homotopy Optimization
- A Theoretical Analysis of Discrete Flow Matching Generative Models
- A Theoretical Analysis of Test-Driven Code Generation
- A Theoretical Analysis on the Emergence of Implicit Curriculum in Transformers
- A Theoretical Reduction From Membership Inference to Reconstruction Attacks Against Learners
- A Theoretical Study of Model Routing with Distributional Losses
- A Theory of Atomic Features and Four Testable Predictions
- A Theory of Frugal Equivalence of Neural Networks
- A Theory of Spatial Continuous Attractors in Hopfield Energy Landscapes
- A Theory of Training Profit-Optimal LLMs
- A Threshold Exceedance Framework for CBRN Uplift Evaluation in Frontier Language Models
- ATLAS: Adaptive Teacher Learning with Asymmetric Supervision
- ATLAS: Adaptive Temporal Learning for Single-Cell Multi-Omics Alignment and Dynamics
- ATLAS: Agent Trace Laboratory for Anomaly Semantics
- ATLAS: A Large-Scale Evaluation Benchmark for Adversarial LiDAR Perception
- ATLAS: The Landscape of Approximate Similarity Search — Two Decades of Algorithmic Advances
- AtlasULP: Domain-aware Universal Link Prediction via Relation Atlas
- AtlasVA: Self-Evolving Visual Skill Memory for Teacher-Free VLM Agents
- AtlasVid: Efficient Ultra-High-Resolution Long Video Generation via Decoupled Global-Local Modeling
- Atomistic Language Modeling Closes the Gap to Geometry-Aware Architectures
- AtomUP: Unsupervised Physics-Informed Diffusion for Atomic-Resolution STEM Simulation-to-Real Synthesis
- AtomWorld-Mem: Memory-Restored World States for Long-Horizon Atomistic Evolution
- A Topological Sorting Criterion for Random Causal Directed Acyclic Graphs
- A Trainable Optimizer
- A Transformer-Derived Iterative Preconditioner
- A Trust Region Approach for Learning Schrödinger Bridges
- Attack Anything: Recursive Self-Evolving Attack Tree Search for Multi-Turn LLM Red-Teaming
- Attack-Free Adversarial Vulnerability Assessment via Second-Order Statistics of Predictive Probabilities
- Attack logics, not outputs: Towards efficient robustification of deep neural networks by falsifying concept-based properties
- AttChain: Hijacking Agent Tool Chain via Multi-stage Malicious Tool Forgery
- AttenA+: Rectifying Action Inequality in Robotic Foundation Models
- Attend Locally, Remember Linearly: Linear Attention as Cross-Frame Memory for Autoregressive Video Diffusion
- Attention-based PCA
- Attention Hijacking: Response Manipulation Across Queries in Vision-Language Models
- Attention is All You Need to Kill Your Latent Neural Process
- Attention Is Not Retention: The Orthogonality Constraint in Infinite-Context Architectures
- Attention Sinks and Outliers in Attention Residuals
- Attention Sinks Induce Gradient Sinks: Massive Activations as Gradient Regulators in Transformers
- Attractor Inversion: A Geometric Account of Adversarial Manipulation in Human Decision-Making
- Attractors of Inter-Layer Connectivity in Deep Neural Networks
- Attribute Inference on Differentially Private and Fair Learning
- Attributing Emergence in Million-Agent Systems
- ATTRIB: Workshop on Data Attribution and Provenance
- A typed tensor language for federated learning
- Audible World Models: Spatially Aware Sound Generation for 3D Worlds
- AudioCALM: Continuous Autoregressive Language Modeling for Universal Audio Generation
- AudioGuard: Toward Comprehensive Audio Safety Protection Across Diverse Threat Models
- AudioSphere: Towards Self-Supervised Spatial Audio Representation Models
- Auditable Neuro-Symbolic Medical Image Diagnosis with Synthesized DeepProbLog Programs
- Auditing Agent Harness Safety
- Auditing Classifier Fairness with Sorted Residual Distributions Helps Name the Disadvantaged Group
- Auditing Explanation Preservation in Sparsified GNN Pipelines
- Auditing Frozen-Decoder Token Coarsening in Long-Context Serving
- Auditing Games for Sandbagging
- Auditing GUI State Predicate Verification
- Auditing Length Confounds in Hallucination Detection Benchmarks
- Auditing Missing Mechanistic Links in Biomedical QA Under Matched Budgets
- Auditing Packet Approval Under Fixed Retrieval: When Answer F1 Misranks Compact Evidence Packets
- Auditing Prompt-Injection Benchmarks with a Neuro-Symbolic Detector
- Auditing Sender Dependence in Latent LLM Communication
- Auditing Single-Query Recoverability in Self-Supervised Representations
- Auditing the Judge: Human-Grounded Bias Discovery, Quantification, and Mitigation in LLM Judges
- Auditing Verifier Placement in Partial-Ontology Canonical-Schema Resolution
- Auditing Workflow-Control Placement in Externally Phase-Labeled LLM Workflows
- Augmented Lagrangian Method for Last-Iterate Convergence for Constrained MDPs
- Augmenting Doob's Lagrangian with the Committor: A Freidlin–Wentzell Perspective on Rare Event Sampling
- Augment Smarter, Not Harder: Mitigating Dimensional Collapse in Self-Supervised Learning
- A Unified Approach for Computing Wasserstein Barycenters of Discrete and Continuous Measures
- A Unified Embedding for Faces, People, and Objects: A Foundation Model Approach to Multi-Category Visual Identification at Human Performance Levels
- A Unified Framework for Change of Measure Inequalities: Applications to Generalization, Memorization and Privacy
- A Unified Framework for Critical Scaling of Inverse Temperature in Self-Attention
- A Unified Framework for Hyperrectangular Conformal Prediction Sets via Rectangularization
- A Unified Framework for Token-Level Watermarks
- A Unified Framework for Uniform-Price Resource Allocation Mechanisms
- A Unified Graph Language Model for Multi-Domain Multi-Task Graph Alignment Instruction Tuning
- A Unified Image and Video Encoder for Multimodal LLMs
- A Unified Neural Architecture for Variable-Wise Shape Constraints
- A Unified Risk View of Uncertainty: Posterior Risk for Disentanglement and Evaluation Beyond Proxies
- A Unified Uncertainty Representation for Graph Neural Networks via Doubly-Spectral Stochastic Expansion
- A Unified View of Jointly Evolving Representations and Images in Diffusion Models
- A Unifying Perspective on Language Model Interpretability
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- Auteur: Language-Driven Cinematographic Framing for Human-Centric Video Generation
- Auto-Annotation with Expert-Crafted Guidelines: A Study through 3D LiDAR Detection Benchmark
- Auto-Compression from First Principles: Signal Propagation in Summation Architectures
- Auto-Dreamer: Learning Offline Memory Consolidation for Language Agents
- AutoHoney: Automating, Deploying, and Evaluating Scheming Honeypots Across Production Codebases
- Autolabeling Satellite Imagery: Large Language Models for Forest Understanding
- Autolearn: Learn by Surprise, Commit by Proof
- Automated Discovery of Category-Specific Refusal Axes in LLMs via Knowledge-Graph Retrieval
- Automated Hypothesis Discovery for Characterizing Annotation Disagreement
- Automated Manipulation Learning
- Automated VNN Solver Configuration Selection Using Large Language Models
- Automatically Refining Coding Rules for AI Coding Agents
- Automatic Constraint Policy Optimization based on Continuous Constraint Interpolation Framework for Offline Reinforcement Learning
- Automating ML for Science: Can Frontier Agents Climb Scientific Hills in the Wild?
- AutoMechInterp: A Reproducible Claim-Verification Benchmark for Mechanistic-Interpretability Evidence
- Automotive-ENV: Benchmarking Multimodal Models in Automotive Cockpit Environments
- Autonomous Agents Need Memory That Consolidates, Forgets, and Monitors Itself
- Autonomous and Self-Adaptive AI-Generated Image Identification Systems
- Autonomous Driving with Evolution-Aware Memory Verification and Physics-Aware Memory Retrieval
- Autonomous Scientific Discovery via Iterative Meta-Reflection
- Autonomous Surrogate Engineering: LLM-Driven Diagnostic Reasoning for Industrial Simulators
- Autonomous weapons are no longer a future-tense debate, but humans must remain responsible.
- AutoOR: An LLM-based Automatic Modeling Workflow for General Operations Research Problems
- Autoregressive Diffusion World Models for Off-Policy Evaluation of LLM Agents
- Autoregressive Generation as a Noisy Contractive Dynamical System
- AutoRPG: Harnessing Systematic AI for Non-Intrusive Low-Latency Mobile RPG Combat
- A Variational Nonconvex Envelope Activation Function with Learnable Curvature Modulation for Stable and Expressive Deep Neural Networks
- AVE-Compass: Towards Holistic Evaluation for Audio-Video Editing Abilities
- Averaged Kernels are Better Kernels
- AVID: A 5T fMRI Dataset for Benchmarking Auditory-induced Visual Mental Imagery Decoding
- AVI-HT: Adaptive Vision-IMU Fusion for 3D Hand Tracking
- AVIS: Adaptive Test-Time Scaling for Vision–Language Models
- AwareCompiler: Agentic Context-Aware Compiler Optimization via Knowledge-Data Synergy
- A Word Is Not a Smell - Towards Continuous Operator Learning for Olfactory Telepresence
- AXIOM: Foundations of Efficient Deep Learning
- BACE: Behavior-Adaptive Connectivity Estimation from Multi-Region Neural Recordings
- Backdoor Channels Hidden in Latent Space: Cryptographic Undetectability in Modern Neural Networks
- Back to Basics: Revisiting Exploration in Reinforcement Learning for LLM Reasoning via Generative Probabilities
- Back to the Roots: Rethinking Multilingual LLMs through Foundational Learner Vocabulary
- BACON: Budgeted Human Calibration for Modeling and Evaluation with Multiple AI Judges
- BaKron: Efficient Quantization with Kronecker-Factored Hessians
- Balancing Frequencies and Pixels in Flow Matching
- Balancing Guidance and Control for Novice Users: A Wizard-Guided, Node-Based System for AI Video Generation
- BalCapRL : A Balanced Framework for RL-Based MLLM Image Captioning
- Baleen: Self‑Interpretable, Robust SSMs with Stochastic Selective Memory
- BanditSched: Contextual Bandits for Data Scheduling in Reasoning Reinforcement Learning
- Bandit Simulation for Average Reward Inference
- BAO-Net: Agentic Bayesian Grounding for Hallucination-Resistant Reasoning across $350\text{B}+$ Parameters
- BAPM: Boundary-Aware Prompt Mining for Training-Free Few-Shot Medical Image Segmentation
- BARM: Bayesian Assistance Routing for Embodied Robot Manipulation over VLM and Human Experts
- Base Items Overfit, New Items Underfit: Hidden Cost of Joint Training in Incremental Adaptation
- BASIL-DCM: Biophysical Amortized Scalable Inference for Latent Dynamic Causal Modeling
- BA-T: An Iterative Transformer for Two-View Bundle Adjustment
- Batch-Aware Hierarchical Routing for Memory-Efficient Mixture-of-Experts with Theoretical Guarantees
- Batch-Conditioned Semantic Anchors for Robust Transductive Adaptation of Vision--Language Models
- BatteryOCV: A Paired Dataset and Benchmark for High-to-Low Rate Profile Reconstruction
- BayesAT: Bayes-Guided Progressive Distillation for Semi-Supervised Adversarial Training
- Bayesian Agentic Medical Diagnosis
- Bayesian Anytime Pareto Set Identification for Multi-Objective Multi-Armed Bandits
- Bayesian-Aware Interactive Diagnostic Reasoning over Dynamic Causality-Informed Graphs
- Bayesian Decision Making around Experts
- Bayesian Filtering Transformer: Precision is What You Really Need
- Bayesian Hawkes Kernel Synthesis: Compositional Triggering Kernels via Probabilistic Program Synthesis
- Bayesian Hierarchical Time-Series Forecasting with Coherent Aggregation
- Bayesian Latent Space Models for Graphs Are Misspecified: Toward Robust Inference via Generalized Posteriors
- Bayesian Model Merging
- Bayesian Optimization in Sequence-to-Architecture Latent Space for Zero-Shot NAS
- Bayesian Optimization with Early Trial Termination for Speeding Up Parallel Neural Network Training
- Bayesian Optimization with Structured Measurements: A Vector-Valued RKHS Framework
- BCCT-Hub: A Benchmark and Toolkit for Measuring Representation Convergence Across Model Families
- BCG-FM: A Foundation Model for Ambient Cardiac Health Sensing
- BCVQ: Binary-coding Vector Quantization for Accurate Acceleration of Large Language Models
- BEACON: A Multimodal Dataset for Learning Behavioral Fingerprints from Gameplay Data
- BEACON: Bridging Offline Priors and Online Adaptation for Efficient Zero-Shot Coordination
- BEACON: Cross-Domain Co-Training of Generative Robot Policies via Best-Effort Adaptation
- BEAM: Binary Expert Activation Masking for Dynamic Routing in MoE
- BECON: Belief-Conditioned Constrained Multi-Objective Reinforcement Learning under Drifting Preferences and Budgets
- BEDA: Bounded-Edge Drift-Adaptation for Label-Free Continual Learning on Wearable Devices
- Before Attention: Token Roles Emerge from Projector-Space Modality Alignment
- Behavioral Deception Detection in Instructed LLM Roleplay Is Dominated by Correction-Marker and Instruction-Following Signals: A Three-Control Audit (3B--70B, English)
- Behavioral Foundation Models for Quality Diversity
- Behavioral Geometric Supervision Aligns Video Foundation Models with Human Social Perception
- Behavioral Integrity Verification for AI Agent Skills
- Behavior Amplification: Discovering and Steering Model-Specific Reasoning Strengths in LLMs
- Behavior-Aware Reward Shaping for Personalized Preference Optimization
- Behavior Cue Reasoning: Monitorable Reasoning Improves Efficiency and Safety through Oversight
- Behavior Pack Optimization for Video MLLM Post-Training
- Behaviour4All: A Dependency-Aware Toolkit for in-the-wild Facial Behaviour Analysis
- Belief-conditioned Predictive Latent Embedding for Zero-shot Multi-agent Coordination
- BeliefFlow: Quality-Aware Belief Packets for Calibrated Financial Multi-Agent Inference
- Belief Memory: Agent Memory Under Partial Observability
- Belief-Shielded MAPPO for Semantic Offloading with Imperfect Server-Health Observability
- Believable but not Measurable: A Psychometric Audit of LLMs on Validated Psychology Questionnaires
- Bellman Contraction under MMD: A Unified Framework
- Bellman-Corrected Dualization for Safe Vision-Language-Action Models
- BELLS-O: Evaluating the Operational Trade-offs of LLM Supervision Systems
- Benchmark Accuracy Is Not Time Series Reasoning: Model, Task, and Evaluation Design Must Be Coupled Around Temporal Evidence
- Benchmark Dataset for Catalysis on 2D MXenes
- Benchmark Everything Everywhere All at Once
- Benchmarking and Improving Monitors for Out-Of-Distribution Alignment Failure in LLMs
- Benchmarking and Optimizing Multimodal Structured Generation: The OracleGraph Dataset and PRISM Framework
- Benchmarking Attention for Tabular Foundation Models
- Benchmarking Fine-Grained Spatio-Temporal Awareness in Embodied Brain Models
- Benchmarking In-Context Experiential Learning from Heterogeneous Web Feedback
- Benchmarking LLM Agent Propensities for Lock-In Risk
- Benchmarking Multi-Modal Graph-based Social Media Popularity Prediction
- Benchmarking Multimodal Mathematical Reasoning with Explicit Visual Dependency
- Benchmarking Vision-Language Models on Ambiguous Images
- Benchmark Validity Under Audited, Withheld Interface Drift
- BenchPress: Benchmark Generation as an Evaluation Task
- BenchPress: Probing Visual Grounding in Video Question Answering via Counterfactual Entity Removal
- BENDER: A Cross-taxon IDP Simulation Database Reveals Conserved Sequence-Ensemble Laws Across the Tree of Life
- Beneath The Code: Detecting LLM-Generated Codes Through Dynamic Execution Footprints
- Benign Overfitting Does Not Occur in Diffusion Models
- Benign Reinforcement Learning Can Amplify Latent Backdoors
- Bernini: Latent Semantic Planning for Video Diffusion
- Best-of-Both-Worlds for Combinatorial Semi-Bandits with Graph Feedback: Beyond the m-Set Case
- Best-of-Tails: Bridging Optimism and Pessimism in Inference-Time Alignment
- Bet Imaginatively, not Historically in Independent-Data Sequential Testing
- Better generalization comes from lower expected Fisher rank
- Better Predictions, Worse Schedules: Selecting Scheduling Predictors on the Wrong Metric
- Better Source, Better Flow: Learning Condition-Dependent Source Distribution for Flow Matching
- Beyond $L_2$: Sobolev-Kinematic Flow Matching for Extreme Sparse Physical Inversion
- Beyond 3 Million Tokens: A Multi-Modal Foundation Model for Full-Resolution Heliophysics
- Beyond Accuracy: A Risk–Coverage–Accuracy Framework for Uncertainty in Clinical AI Screening
- Beyond Accuracy: Competence-Based Evaluation of Logical Reasoning in Large Language Models
- Beyond Accuracy: Information-Theoretic Decoder Evaluation Under Latent Intent
- Beyond Accuracy: Stable Predictions, Unstable Topology in Transformer Representations
- Beyond AI Detection: Measuring the Amount of Human Contribution in Textual Outputs
- Beyond Answer-Level Scaling: Consolidating Reasoning Paths for Inference-Time Scaling
- Beyond AUROC: A Trajectory-Aware Audit Framework for Alarm-Driven Risk Models
- Beyond Average Performance: Dynamic Instance Clustering and Specialized Algorithm Design in LLM-Assisted Evolutionary Search
- Beyond Bag-of-Words: Diagnosing Compositional Binding Failures in Vision-Language Models
- Beyond Behavioral Benchmarks: Surfacing Narrow Hidden Bias in Fine-Tuned Language Models
- Beyond Binary Correctness: Confidence Estimation for Continuously Graded Responses in Large Language Models
- Beyond Bit Matching: Orthogonal Watermarks for Collusion-Resistant Image Fingerprinting
- Beyond Box Overlap: Correspondence-Aware Evaluation for Video Copy Localization
- Beyond Brightening: A Physically-Consistent Domain Adaptive Enhancer for Nighttime UAV Tracking under Complex Illumination
- Beyond Chamfer Distance: Granular Order-aware Evaluation Metric For Online Mapping
- Beyond Classification: Pairwise Ranking with Curriculum Learning for Generalizable Multilingual Readability Assessment
- Beyond Closed-Pool Video Retrieval: A Benchmark and Agent Framework for Real-World Video Search and Moment Localization
- Beyond Contraction: Geometry-Faithful Supervised Dimensionality Reduction for Data Visualization
- Beyond Copy-Paste: How Well Do Subject-Driven Video Models Understand Their Subjects?
- Beyond Correctness: Robustness-Driven Evolutionary Self-Training for Large Language Models
- Beyond Data Augmentation: Distilling Mixup's Regularization Kernel via Gradient Dynamics
- Beyond Deterministic Fusion: Distributional Multi-Domain Reinforcement Learning for Point Cloud Domain Generalization
- Beyond Domains: Reusing Web Skills via Transferable Interaction Patterns
- Beyond Drug Discovery: The Nanotechnology Molecular Optimization (NMO) Benchmark
- Beyond Empirical Quantiles: A New Framework for Conformal Prediction
- Beyond Empirical Support: Structured Outlier Generation via Sinkhorn Optimal Transport
- Beyond Encoder Accumulation: Measuring Encoder Roles in Multi-Encoder VLMs
- Beyond Endpoint Sufficiency: Inner-Time Trajectory as a Representation Channel in Iterative Systems
- Beyond Entropy: Learning from Token-Level Distributional Deviations for LLM Reasoning
- Beyond Evidence Access: Execution-Time Evidence Management for Retrieval-Augmented Generation
- Beyond Exemplar Selection: Value-Aware Memory Allocation in Replay-Based Continual Learning
- Beyond Expected Returns: Time-Average Regularized Policy Optimization for Trajectory-Level Performance
- Beyond Expert Identity: Routing Stability Diagnostics for Time-Series Mixture-of-Experts: A Closer Look
- Beyond Feature Disruption: Boundary-Diverting Unlearnable Examples against Linear Probing
- Beyond Feature-Space Priors: View-Specific Label Decomposition for Multi-View Multi-Label Learning
- Beyond Gaussian Bottlenecks: Topologically Aligned Encoding of Vision-Transformer Feature Spaces
- Beyond Generation: Unlocking Discriminative Representations from Diffusion Models
- Beyond Greedy Heuristics: Learning Task-Server Compatibility for Dynamic Load Balancing
- Beyond Ground Truth: Evaluating Non-Verifiable Reasoning in LLMs through Moral Robustness
- Beyond In-Distribution Generalization: Benchmark Machine Learning for Ground State Property Prediction of Quantum Systems
- Beyond Langevin: Sampling Multimodal Densities using the Witten Laplacian on 1-forms
- Beyond Linear Analogies: Probabilistic and Non-Commutative Semantic Transformations in Structured Latent Spaces
- Beyond Linear and Overcomplete Regimes: A Mean-Field Analysis of Bottleneck Autoencoders
- Beyond Linear Decoders: Dynamic Expert-Coupled Optimal Decoding for Time Series Forecasting
- Beyond Markov: Attention-Driven Policy Optimization for LLM Decision Making
- Beyond Maximum Likelihood: Variational Inequality Estimation for Generalized Linear Models
- Beyond Medical Diagnostics: How Medical Multimodal Large Language Models Think in Space
- Beyond MMSE: Enhancing PnP Restoration with ProxiMAP
- Beyond MoCap: Scaling Motion Tokenizers with Synthetic Human Motion for Generative Modeling
- Beyond Mode-Seeking RL: Trajectory-Balance Post-Training for Diffusion Language Models
- Beyond Myopia: Pre-trained Optimization Models with Global Population Dynamics
- Beyond Next-Token: Evaluating Text Embeddings for Next-State Dynamics
- Beyond Next Token Prediction: Diffusion and Flow Models for Next-Generation Decoding
- Beyond Normalization: Hierarchical Periodic Encoding for Time Series Forecasting
- Beyond One-Shot Retrieval: Learning Corpus-Aware Planning for Iterative Search
- Beyond Outcome Rewards: Process-Aware Optimization for Search Agents
- Beyond Oversquashing: Understanding Signal Propagation in GNNs Via Observables
- Beyond Pairwise Preferences: Listwise Reward-Aware Alignment for Diffusion Models
- Beyond Patch Slimming: Discriminative Reweighting for Fine-Grained Cross-Modal Alignment
- Beyond Pixel Distance: A Difficulty-Stratified, Multi-Metric Framework for Evaluating Inverse Vector Graphics in Multimodal Language Models
- Beyond Pixels: Bridging the Cognitive Gap in AI-Generated Video Detection
- Beyond Prediction: Steering VLM Agents with Retrospective World Modeling
- Beyond Private Training: The New Landscape of AI Privacy
- Beyond Reasoning: Reinforcement Learning Unlocks Parametric Knowledge in LLMs
- Beyond Red-Teaming: Formal Guarantees of LLM Guardrail Classifiers
- Beyond Residuals: An Energy-Aware Sampling Paradigm for Physics-Informed Neural Networks
- Beyond Return Distributions: Reconstructing Distributional Critic Benefits in Continuous Control
- Beyond Reward Correctness: Distribution-Driven Mechanisms in LLM Reinforcement Learning
- Beyond Risky Activities: Bridging the Supervision Gap for Situational Risk Reasoning
- Beyond Robust Algorithmic Recourse: Sharing the Burden of Validity through Multiplicity
- Beyond Sample Anchors: Distributional Wasserstein Trust Regions for High-Dimensional Continuous Control
- Beyond Scalar Distances: Semantic Attribute Gradients from Frozen MLLMs for Visual Embeddings
- Beyond Scaling: Hierarchical Predictive Coding Enables Few-Shot Generalization and Efficient Exploration
- Beyond Selection: Token Parameterization for Extreme Visual Token Compression
- Beyond Self-Exploration: Stable Off-Policy Reinforcement Learning for LLMs
- Beyond Semantic Alignment: Geometric Incomparability in Multi-Oracle Soft Fusion
- Beyond Semantic Eclipsing: Learning Desemanticized Textual Anchors for AI-Generated Image Detection
- Beyond Sequential Plans: A Three-Facet Evaluation of Agentic Planning under Dependency-Preserving Augmentation
- Beyond Shapley: An Influence-Based Data Auditing Pipeline for LLM Alignment and Evaluation
- Beyond Solving: Prescriptive Probing for Neural Combinatorial Optimization
- Beyond Sparse Linear Models: Sparse Input Neural Networks for High-Dimensional Feature Selection
- Beyond Spatial and Temporal Priors: A Generalizable Approach for Dense Correspondence Matching
- Beyond Spatial Compression: Interface-Centric Generative States for Open-World 3D Structure
- Beyond Stabilization: Dual-EMA Teachers for Global–Local Semantic Learning in Semi-Supervised Medical Image Segmentation
- Beyond Stale Credit: Single-Pass Actor-Critic Preference Optimization for LLM Agents
- Beyond Staleness Decay: History-Enhanced Residual Acceleration in Asynchronous Federated Learning
- Beyond Static Bias: Adaptive Multi-Fidelity Bandits with Improving Proxies
- Beyond Static Charts: Benchmarking Interactive Chart-to-Code Generation for Multimodal Large Language Models
- Beyond Static Sandboxing: Learned Capability Governance for Autonomous AI Agents
- Beyond Strings: MORSE, a GNN-Native Operator Code for Molecules and Chemical Reactions
- Beyond Structural Agnosticism: Stable-Rank-Guided LoRA for Structure-Aware Fine-Tuning
- Beyond Success Rates: Trainability and Extractability in Offline GCRL
- Beyond Surface Semantics: Logic-Guided Privacy-Preserving Cloud-Edge LLM Reasoning
- Beyond Surface Style: Aligning Multi-Turn User Simulators with Behavioral Consistency
- Beyond Synthetic Splits: A Benchmark for Federated Learning on Real-World Medical Data Classification
- Beyond Text Conditioning: A Systematic Study of MLLM-DiT Fusion for Video Generation
- Beyond the Half Approximation: Fair and Efficient Online Class Matching
- Beyond the Prompt: Leveraging Pre-Decoding States for Jailbreak Detection in dLLMs
- Beyond the Synaptic Connectome: Inferring Active Neural Circuits Using Diffusion Scores
- Beyond the Trade-Off: Defining a Constrained Optimization to Regularize with the Least Training Error
- Beyond the VAE Bottleneck: One-Step Pixel Diffusion from Latent Teachers
- Beyond the Window: An Information-Theoretic Account of Sequence Memory
- Beyond Truthfulness: Evaluating Honesty in Large Language Models
- Beyond Uniform Credit Assignment: Selective Eligibility Traces for RLVR
- Beyond Uniformity: Selective Control for Text-Guided Music Editing
- Beyond Uniform Lipschitzness: Pathwise Jacobian Stability for Flow Matching Samplers
- Beyond View Consistency: Rotation-Consistent Alignment for Viewpoint-Robust Robotic Manipulation
- Beyond Weisfeiler–Lehman: Topological Features Increase Graph Neural Network Expressivity
- Beyond What to Select: A Plug-and-play Oscillatory Data-Volume Scheduling for Efficient Model Training
- Beyond When: How Root Content Propagates as Information Cascade Trees
- Beyond Zooming: Learning Multi-Tool Visual Reasoning for Ultra-High-Resolution Remote Sensing
- Bian Que: An Agentic Framework with Flexible Skill Arrangement for Online System Operations
- BIAS-ID: A Framework for Analyzing Transformation Biases in AI-Generated Image Detectors
- Bid-Agnostic Auction for Client Selection in Online Federated Edge Learning
- Bidirectional-Guided Diffusion Model for Multi-Scale Time Series Forecasting
- BiFlow: First-Order Bi-Level Coarse-to-Fine Flow Matching Policy for Continuous Control
- Bifurcations Break Neural Surrogates: An Approximation Barrier and a Deployment-Time Certificate
- Bilevel Data Curation for LLM Fine-tuning: Offline Selection and Online Self-Refining Generation
- BillSim: A Synthetic Heterogeneous Graph Benchmark for Non-Adversarial Anomaly Detection in Usage-Based Billing
- BiLoCo: Binary Low-Rank Corrections for LLM FP4 Decode
- BIM-Edit: Benchmarking Large Language Models for IFC-Based Building Information Modeling
- Binary Rewards and Reinforcement Learning: Fundamental Challenges
- Binary-RL: Binarized Policy Learning via Continuation Training
- Binding Crystallizes Before the Image Forms
- Binding Decoupling: A Scale-Aware Diagnostic of When LLMs Know but Refuse to Say
- BindingGYM: A Large-Scale Mutational Dataset Toward Deciphering Protein-Protein Interactions
- Binding-Output Coupling Explains Phase Transitions in Compositional Generalization
- BioBlobs: Unsupervised Discovery of Functional Substructures for Protein Function Prediction
- Bioheat-PINN: A Physics-Informed Neural Surrogate for 3D Hepatic Thermal Ablation Planning
- BioInteract: A Large-Scale Multimodal Dataset for Evaluating Fine-Grained Semantic Understanding of Biotic Interactions
- Bio-JEPA: Learning Semantic Representations for Drug Repositioning via Protein-Ligand Affinity Prediction with Quantum Fingerprint Validation
- Biological Sequence Analysis Using the Bezier Curve
- BioLT: Streaming Biosignal Modeling
- BioMicroAgents: A Co-evolutionary Multi-Agent Framework for High-Fidelity Biomicroscopy Imaging
- BioMM-Eval: A Benchmark for Diagnosing Modality Reliance in Multimodal Biological Models
- BiomniBench: Process-level Evaluation of LLM Agents for Real-world Biomedical Research
- BioTrack: Bio-Inspired Resource-Aware Track with Cross-Temporal Interaction Networks
- Birds of a feather: The indirect reinforcement effect in data synthesis
- BitMoE: Post-Training Binarization for MoE-based Large Language Models
- Black-Box Followers, White-Box Leaders: Partial Zeroth-Order Methods for MPECs
- Black-box model classification under the discriminative factorization
- Black-Mamba: Biologically-Inspired Leaky Accumulation for Conceptual Knowledge under Distribution Drift
- BLANP: Memory-Efficient Backpropagation-Free Local Training via Antithetic Node Perturbation
- BLAZE: Bias-Driven Load-Aware Zero-Overhead Expert Routing for MoE LLM Inference
- BlenderFORGE: Framework for Optimizing Reactive 3D-Graphics Editing Ability of MLLMs
- Blind Perception: Model Curvature-based Membership Inference
- Block-Based Double Decoders
- Block Optimism for Nonstationary Bandits with Latent Linear Dynamics
- Block Sparse Flash Attention
- Block Sphere Vector Quantization
- BlueFin: Benchmarking LLM Agents on Financial Spreadsheets
- Blur Issue Matters for Thermal Novel View Synthesis: A Floating Gaussian Suppression Approach
- Boltzmann-Expected Molecular Design with Decoupled Annealing Flows
- BONSAI: Bayesian Optimization with Natural Simplicity and Interpretability
- Boosting LLM Reasoning via Human-Inspired Reward Shaping
- Boosting Multiagent Reinforcement Learning at High Replay Ratios with Ensemble Reset
- Boosting Reinforcement Learning with Verifiable Rewards via Randomly Selected Few-Shot Guidance
- Boosting Supports Max-Min Fairness Without Identifying Demographics
- BoostLoRA: Growing Effective Rank by Boosting Adapters
- BoostNAM: Distributional Neural Additive Models with Multi-Stage Boosting and Per-Feature Uncertainty Diagnostics
- Bootstrapped Monitoring: Leveraging Transparent Reasoning to Oversee Stronger AI Agents
- Bosonic Walk Observables for Controlled Graph Expressivity
- Boundary Optimality Activation: A Geometric Perspective on Learning Enhanced Neural Representations near Feasibility Boundaries
- Bound-Conditioned Latent Inference for Progressive Image Compression
- Bounded-Abstention Multi-horizon Time-series Forecasting
- Bounding Hallucinations: Merlin-Arthur Protocols for Mutual-Information Bounds in Language Models
- BPP: Beyond Pixels and Points - A Robust Robot Policy with View-Invariant Motion Representations
- BRACE: Bipolar Reference-Aware Calibration and Estimation for Incomplete Multimodal Learning
- Bradley-Terry Policy Optimization for Generative Preference Modeling
- Brain2voice 2.0: High-performance voice synthesis brain-computer interface
- Brain Disease Detection Based on Bottom-Up Hierarchical Graph Representation Learning
- BrainExplore: Large-Scale Discovery of Interpretable Visual Representations in the Human Brain
- BrainFlow: Latent Flow Matching for Subject-preserving Longitudinal 3D Brain MRI Generation
- BrainGenFlow: Modeling Longitudinal Progression via Residual Flow Matching and Gaussian Imputation
- BrainWorld: A Structural-Prior-Conditioned Generative Model for Whole-Brain 4D fMRI Dynamics
- Breadcrumbing Search Agents: Per-Turn Scheming Over Long-Horizon Trajectories
- Breakeven complexity: A new perspective on neural partial differential equation solvers
- Breaking Symmetry Bottlenecks in GNN Readouts
- Breaking the Bubble: Asynchronous Pipeline Parallel Training with Bounded Weight Inconsistency
- Breaking the Diffusion Bottleneck: Train-Free Acceleration via Stage-Aware Model Allocation
- Breaking the Group Size Barrier: Parameter-Efficient Group Dance Generation with Chain-of-Dancers
- Breaking the Historical Action Shortcut: Mitigating Visual Modality Collapse in Gaming VLA Models via Historical Action Noise Injection (Extended)
- Breaking the Second Barrier: Sub-Second Timestamped Omni-Modal Captioning
- Breaking the Silo Between Software and Robotic Agents is Essential for General Edge Intelligence
- Breaking the Tokenizer Barrier: On-Policy Distillation across Model Families
- Break the Confirmation-Bias Loop: A Dedicated Pseudo-Labeler for Class-Imbalanced Semi-Supervised Learning
- Break the Optimization Barrier of LLM-Enhanced Recommenders: A Theoretical Analysis and Practical Framework
- BReD: Block Replay Dithering for Stable Low-Bit EMA Optimizer States
- Bricker to BRACE: A Bracket Exposure RAW Dataset and Restoration Model for Flicker-Banding
- BRICKS: Compositional Neural Markov Kernels for Zero-Shot Radiation-Matter Simulation
- BRIDGE: A Bias Register-Integrated Dataset for Generalized Evaluation of LLM Bias Detection
- BridgeMVS: Bridging Multi-View Stereo and Monodepth via Bidirectional Dynamic Fusion
- Bridge-Rescued RAG: Fixed-Budget Evidence Shaping for Multi-Hop QA
- Bridge the Deformation Gap: Elastic Flow-guided Few-shot Medical Image Segmentation
- BridgeTwist: Twisting Schrödinger Bridges for Training-Free Conditional Sampling
- Bridging Academia and Industry: A Comprehensive Benchmark for Attributed Graph Clustering
- Bridging Back to Clean Semantics: One-Sided Anchoring Enables Robust Spoken Language Understanding
- Bridging Code and Docs: Natural Language Enrichment in Code Knowledge Graphs for Repository-Level Question Answering
- Bridging Constraints and Stochasticity: A Fully First-Order Method for Stochastic Bilevel Optimization
- Bridging Convection and Diffusion Regimes via a Structure-Preserving Neural Operator
- Bridging Disparate Emotional Granularities: A Hierarchical Consistency Framework for Emotion Recognition
- Bridging Hypergraphs and LLMs for Pharmacological Relation Discovery: A Benchmark and a Simple Framework
- Bridging Local Observation and Global Simulation in Closed-Loop Traffic Modeling
- Bridging Modalities, Spanning Time: Structured Memory for Ultra-Long Agentic Video Reasoning
- Bridging Modalities, then Aligning Cells: Cross-Modal Pathology Cell Representation Learning without One-to-One Correspondence
- Bridging Optimal Transport, Learning and Structured Data: Toward Geometric Distributional Learning
- Bridging Perceptual and Analytic Dynamics via Function Alignment
- Bridging Reasoning and Adaptation: Multi-View Distillation with Drift-Aware Verbalizer in Bullet Chats for Recommendation Intent Identification
- Bridging Simulation and Reality: Geometry and Decision Alignment for Autonomous Driving
- Bridging the Annotation Gap: Automated Egocentric Action Annotation with MLLMs
- Bridging the Gap Between Latent and Explicit Reasoning with Looped Transformers
- Bridging the Last Mile of Circuit Design: PostEDA-Bench, a Hierarchical Benchmark for PPA Convergence and DRC Fixing
- Bridging the Simulation-to-Experiment Gap with Adversarial Distribution Alignment
- Bridging the Statistical Mismatch: Higher-Order Moment Alignment for Weight Initialization
- Budget-Adaptive Research Plan Tree Search for Autonomous Research Agents
- Budget-Conditioned Clipping Policies for Differentially Private Federated Learning
- Budgeted Agentic Workflow with Pareto Frontiers
- Budgeted Multiple-Expert Deferral
- Budgeted Multi-Source Counterfactual Annotation for Off-Policy Evaluation
- Budgeted One-Pass Best-Arm Identification
- Budgeted Quotient-Residual Guidance for Frozen Pocket-Conditioned Molecular Diffusion
- Budgeting Discretion: Theory and Evidence on Street-Level Decision-Making
- BuildArena: A Physically Grounded Ecosystem for Evaluating Model Intelligence by Artifacts it Creates
- Building Bridges to Global Optima
- BUNO: Integral-Consistent Basis-Universal Neural Operators for Diverse PDE Classes
- Buying Data of Unknown Quality: Statistical Information Procurement Auctions
- Buying Time: Net Slack as a Control Variable for Proactive Real-Time Embodied Control
- bViT: Investigating Single-Block Recurrence in Vision Transformers for Image Recognition
- C3P: Contrastive promoter-protein pretraining yields representations capturing bacterial gene regulation
- C3VD-DEFCOL: A Deformable Colonoscopy Dataset with Time-Resolved 3D Ground Truth and Realistic Appearance
- CacheRDA: Repurposing Prefix KV Cache for Efficient Reasoning Depth Augmentation
- CADBench: A Multimodal Benchmark for AI-Assisted CAD Program Generation
- CAD-Zero: Efficient Training-Free Zero-Shot Camouflaged Object Segmentation
- CAFE: Causally-Guided Automated Feature Engineering with Multi-Agent Reinforcement Learning
- CAIRN: Cross-Room 3D Scene Understanding with Topology-Aware Large Multimodal Models
- CA-Judge: Teach Large Models to Judge Anomalies via Comparison for Video Anomaly Detection
- CalArena: A Large Scale Post-Hoc Calibration Benchmark
- Calibrated Low-Rank Adaptation for Continual Visual Instruction Tuning
- Calibrate, Don't Curate: Label-Efficient Estimation from Noisy LLM Judges
- Calibrated Partial Resets: Preventing Policy Collapse in Continual Reinforcement Learning
- Calibrated Predictions or Signals? Performance and Sense of Agency in AI-Assisted Decision Making
- Calibrated Surrogate Losses for Adversarial Classification With a Reject Option
- Calibrating LLMs with Semantic-level Reward
- Calibrating Scientific Foundation Models with Inference-Time Stochastic Attention
- CALLIOPE: Private Insights Into AI Use, Without Embeddings
- CALM: A Continual Adaptive Learning Model with Self-Aware Readiness Agent for Reliable Few-Shot Vision Systems
- CALYREX: Cross-Attention LaYeR EXtended Transformers for System Prompt Anchoring
- CAMAL: Improving Attention Alignment and Faithfulness with Segmentation Masks
- CAMEO: Towards Online Active Learning in the Open World
- Camera-Aware Prompting for Debiased Clustering in Unsupervised Visible-Infrared Person ReId
- CAM: Question Answering on Entity-Centric Videos with Continuous Extraction and Adaptive Querying
- CanadaWildFireDaily: A Large-scale Dataset for Daily Wildfire Spread in Canada
- Can Agent Benchmarks Support Their Scores?
- Can AI Agents Synthesize Scientific Conclusions?
- Can an LLM Reason Like a Lawyer? Benchmarking the ability of LLMs to map the facts of a case to the elements of the applicable legal rule
- Canary-Controlled Feedback Interleaving for Emergent Misalignment in LLM Fine-Tuning
- Can Bits Seal Language?
- Can Circuit Alignment Predict OOD Generalization?
- Can Complementary Signals Bridge Similarity Islands? Manifold-Augmented Graph Embedding for Multimodal Recommendation
- Can Hybrid-Parallel Planning Support Alternating Model–Strategy Design? Dependency-Keyed Per-Layer Primitives for Re-Planning
- Can In-Sample Offline RL Learn Critics First and Extract Policies Later?
- Can Language Models Actually Retrieve In-Context? Drowning in Documents at Million Token Scale
- Can Linguistic Reasoning Vectors Enhance Multimodal Reasoning Ability?
- Can LLM Agents Communicate Without Autoregression? Thought Blocks for Latent Multi-Agent Collaboration
- Can LLMs Beat Classical Hyperparameter Optimization Algorithms? A Study on autoresearch
- Can LLMs Improve at Debate Through Self-Play? A Pipeline and RL System to Gamify Competitive Debate with Branching and Hierarchical Rewards
- Can LLMs Predict Polymer Physics Just by Reading Synthesis and Processing Prose?
- Can LLMs Serve as Reference Anchors for the German Kangaroo Competition?
- Can LLMs Take Retrieved Information with a Grain of Salt?
- Can Memory Enhance Security? Reusing Agent Memory Against Indirect Prompt Injection
- Can Model Merging Improve Aggregation in DiLoCo?
- Cannistraci-Hebb Channel-wise Dynamic Sparse Training of Convolutional Neural Networks with Contextual Modulation
- Canonical Paraphrasing: A Provable Defense against Steganography in Language Models
- Can RL Teach Long-Horizon Reasoning to LLMs? Expressiveness Is Key
- Can Tainted Pixels Expose Deepfake Videos?
- Can Unified Generation and Understanding Models Maintain Semantic Equivalence Across Different Output Modalities?
- CanvasAgent: Enabling Complex Image Creation and Editing via Long-Horizon Tool Orchestration
- Can Vision-Language Models Locate the World from Above?
- Can We Break LLMs Out of Self-Loops? Fine-Grained Reasoning Control with Activation Steering
- Can We Model the Artifacts Explicitly? Disentangle Artifacts via Pairwise Edit Relations for Image Manipulation Localization
- Can We Really Learn One Representation to Optimize All Rewards?
- Can we train a computer? Two mechanisms for content addressing in transformers
- Can We Trust AI Evaluation? Robustness, Causality, and Risk in Modern AI Assessment
- Can We Trust Item Response Theory for AI Evaluation?
- Can We Trust the Judge? Building Reliable Evaluation for Language Models
- Can You Hear, Localize, and Segment Continually? An Exemplar-Free Continual Learning Benchmark for Audio-Visual Segmentation
- Capability Control Should be a Separate Goal From Alignment
- Capacity Allocation at the Source: Sparse Target Optimization for LLM Knowledge Editing
- CAP: Candidate Acquisition Policy for Data-Scarce Bayesian Optimization
- CAPS: Cascaded Adaptive Pairwise Selection for Efficient Parallel Reasoning
- Car4Cast: A Dataset and Benchmark for LLM-Based Motion Forecasting and Spatial Reasoning in Autonomous Driving
- CarCrashNet: A Large-Scale Dataset and Hierarchical Neural Solver for Data-Driven Structural Crash Simulation
- CAREBench: Evaluating LLMs' Emotion Understanding by Assessing Cognitive Appraisal Reasoning
- CARE: Calibrated Reward for Hallucination-Aware Reinforcement Learning of Large Language Models
- CaRE: Compute-aware Remasking Evaluation Protocol for Masked Diffusion Language Models
- CASAM: Consistency-Anchored Sharpness-Aware Minimization for Improved Model Generalization
- CASCADE: Context-Aware Relaxation for Speculative Image Decoding
- Cascade-DMD: Closing the Train-Test Gap in Multi-Resolution Video Generation via End-to-End Distribution Matching Distillation
- CASCADE: Mitigating Privacy Over-Refusal via Streaming Semantic Remediation
- CascadeNet: Debiased Network Inference from Cascade Data
- Case Based Reasoning with Any Classifier Type
- Case-Grounded Evidence Verification: A Framework for Constructing Evidence-Sensitive Supervision
- Casper: A Projection-Based Neurosymbolic Layer for Scalable & Guaranteed Constraint Satisfaction
- CASPIAN: Online Detection and Attribution of Cascade Attacks in LLM Multi-Agent Systems via Cross-Channel Causal Monitoring
- CAST: Causal Anchored Simplex Transport for Distribution-Valued Time Series
- CAST: Conditional Conformal Filtering for LLM-Generated Sentences in Human–AI Hybrid Texts
- CAST: Credit Assignment on Selective Reasoning Trees
- Catch-Only-One: Non-Transferable Examples for Model-Specific Authorization
- Categorical Bayes filtering for computational phenotyping in adaptive learning
- Categorical Drifting Models
- CATS: Acceptance-Oriented Critical Token Adaptive Selection for Multimodal Speculative Decoding
- Cauchy-Riemann Regularization for Extrapolative Latent Representations
- CausalCompass: Evaluating the Robustness of Time-Series Causal Discovery in Misspecified Scenarios
- Causal Decomposition of Misinformation Diffusion under Latent User-Environment Factors
- Causal Density Functions
- Causal Discovery in Structural VAR Models Under Equal Noise Variance (Extended)
- Causal Discovery of Spatiotemporal Systems From Limited Physics Knowledge and Data
- CausalDriveBench: Evaluating Causal Reasoning in Vision-Language-Action Models for Autonomous Driving
- Causal-Driven Feature Evaluation for Cross-Domain Image Classification
- Causal Effect Identification with a Single Agnostic Proxy
- Causal Evaluation of Membership Inference Attacks
- CausalEvolve: Towards Open-Ended Discovery with Causal Scratchpad
- Causal Fairness for Survival Analysis
- CausalFE: Causal Graph-Driven Automated Feature Engineering for Tabular Data
- Causal Foundation Models with Continuous Treatments
- Causal Gating under Proxy Confounding for Multi-Modal Financial Prediction
- Causal Hyperbolic Hypergraph Networks for Pan-Cancer Survival Prediction
- Causal Integrated Gradients: Scalable Feature Attribution Along Causal Paths
- Causal learning with the invariance principle
- Causally Constrained Optimal Transport for Multimodal Relation Extraction
- Causal Machine Unlearning via Counterfactual World Modeling
- CausalMem: Counterfactually Validated Memory for Training-Free Visual Reasoning Agents
- Causal Multi-Task Demand Learning
- CAUSALNAV: Neuro-Symbolic Reasoning over Learned Causal World Models
- Causal Pruning: The Interventional View of Taylor-FO Scales to Modern LLMs
- CausalReason: Counterfactual Frame Necessity and Bootstrap Collapse in Zero-Annotation Video Grounding
- Causal Reinforcement Learning for Complex Card Games: A Magic The Gathering Benchmark
- Causal Representation Meets Stochastic Modeling under Generic Geometry
- Causal Spatiotemporal State Evolution for Monocular 4D Visual Geometry
- Causal Variational Deep Embedding: A Family of Interventional Generators for Confounded Images
- Causal World Models for Evaluating RAG and Agentic AI Beyond LLM-as-a-Judge via Knowledge and Step Attribution
- Causing Crashes Without Having One: Training Realistic and Risk-Averse Adversarial Driving Agents
- CCL-Bench 1.0: A Trace-Based Benchmark for LLM Infrastructure
- CDC-OIL: Class-Domain Coupled Online Incremental Learning
- CD-RMOT-Bench: Benchmarking the Cross-Domain Referring Multi-Object Tracking
- CELEUS: Certifiable and Efficient LLM Evaluation via E-Processes
- CellLens: Hybrid Vision-Language and Optical Flow Analysis for Cardiomyocyte Microscopy Videos
- CellU-Net: Fusing U-Net with Position-Enhanced Neural Cellular Automata for Face Restoration
- Censored LLMs as a Natural Testbed for Secret Knowledge Elicitation
- Centered Mirror Descent Policy Optimization for Stable and Efficient Reinforcement Learning
- Central Dogma Transformer III: Interpretable AI Across DNA, RNA, and Protein
- Certifiable Robustness from Approximate Gaussian Mixture Structure in Pretrained Latent Spaces
- Certificate-Constrained Policy Learning
- Certification from Examples is Hard for Circuits and Transformers under Minimal Overparametrization
- Certified Adaptive Trust Regions for Safe Policy Optimization via Fisher Uniqueness
- Certified Bias Control for Counterfactual Guided Machine Unlearning
- Certified Diffusion Smoothing for Robust Reinforcement Learning
- Certified Output-Feedback Proposal Sampling with Diffusion
- Certified Policy Optimisation for Nested Causal Bandits via PAC-Bayes Risk
- Certified Robust Interpretability via Concept-Space Stability under Interventional Proxies
- Certified Robustness via Sparse Second-Order Cone Relaxations
- CFC26: Building Evaluations for Deployment in Sonar-Based Fish Counting
- CGA-Bench: From Action-Set Scoring to Trace-Level Conformance in Clinical-Agent Evaluation
- CGD-Net: A Weakly-Supervised Framework for Macro-Constrained GDP Disaggregation via Dual-Branch Graph Learning
- C-GRPO: Conformal Group Relative Policy Optimization
- Chain-Divergence: A Cross-Vendor Measurement and Mechanism Probe for Replay-Attestable LLM Agents
- Chain-of-Replicas Theory for Block Gibbs Sampling in Restricted Boltzmann Machines
- Chain-of-Route: State-Aware LLM Routing for Multi-Turn Conversations
- Chain-of-thought needs a substrate: the case for typed, governed decision graphs in organizational LLM agents
- Challenges in Evaluating Contextual Integrity for Large Language Models
- Chance-constrained Flow Matching for High-Fidelity Constraint-aware Generation
- Chance-Constrained Monte Carlo Tree Search for Online Stochastic Scheduling Problems
- Changing Modalities: Adapting Remote Sensing Models to New Satellites and Sensors
- Chaos-Aware Spatio-Temporal and Cross-City Knowledge Transfer for Traffic Flow Prediction
- Characterizing Gestalt Organization in Deep Visual Representations
- Characterizing Jaggedness Aids Safety & Usability
- Characterizing the Aesthetic Defaults of Generative Image Models
- Characterizing the Effect of Cross-Task Contamination on Meta-Learning
- Characterizing the Generalization Error of Random Feature Regression with Arbitrary Data-Augmentation
- Characterizing Trainability of Instantaneous Quantum Polynomial Circuit Born Machine
- Chart2CSV: Can VLMs Faithfully Convert Complex Charts into Structured Tables?
- CHART: Causal Hidden-state Attention Routing Transfer via Delta-Aware Distillation for High-Throughput LLM inference
- Cheap Talk, Real Stakes: Commitment and Exploitation in Human-LLM Strategic Interaction
- Chebyshev Center-Based Direction Selection for Multi-Objective Optimization and Training PINNs
- Child Safety in AI
- CholMA: A Multi-Expert Annotated Benchmark for Surgical Action Triplet Recognition
- Choosing the Laplacian Over the Gaussian Kernel for Explainable Clustering
- CHoRD
- CHORUS: Adapting the Number of Demonstrations via Test-Time LLM Consensus
- CIG: Exploration via Conditional Information Gain
- Cine-JEPA: A Joint-Embedding Predictive Architecture for Hierarchical EEG Decoding during Naturalistic Movie-Watching
- CIPO: Counterfactual Imagination Policy Optimization for Adaptive Tool Granularity Selection
- CIPQA-9K: A Large-Scale Dataset and Learning Framework for Perceptual Quality Assessment of Colorized Images
- CIR-CoT: Towards Interpretable Composed Image Retrieval via End-to-End Chain-of-Thought Reasoning
- CircuitBench: Evaluating Reasoning and Visual Understanding in Large Language Models
- CISyn: Human Mesh Recovery in Close Interactions with Physics-Simulated Synthetic Data
- CitaStat: A Labeled Dataset for Citation Classification
- CiteVQA: Benchmarking Evidence Attribution for Trustworthy Document Intelligence
- CKDU: Certified Unlearning for Distilled Teacher-Student Models
- CKT-WAM: Parameter-Efficient Context Knowledge Transfer Between World Action Models
- CLASP-GRPO: Feasibility-Preserving Grouped Reinforcement Learning for LLM Combinatorial Optimization
- Classroom Final Exam: An Instructor-Tested Reasoning Benchmark
- Claude Coke: Prevent Automated Crime by Agents
- Claw-Eval-Live: A Live Agent Benchmark for Evolving Real-World Workflows
- Claw-SWE-Bench: Benchmark of Coding Agent on OpenClaw and Harness
- CLEAR: Comprehensive Pseudo Label Evaluation And Refinement for Semi-Supervised Camouflaged Object Detection
- Clearer Sight, Fewer Lies: Oriented Pickup Preference Optimization for Multimodal Hallucination Mitigation
- CLeVar: Contrastive Learning on Variation space
- Cliff Tokens: Identifying Single-Token Failure Triggers in LLM Mathematical Reasoning
- Climate Downscaling with Stochastic Interpolants
- Clinical Trajectory Alignment for Medical Vision-Language Pre-training
- ClinMAS: A Knowledge-Grounded Multi-Agent Simulation Framework for Evaluating Clinical Reasoning in LLMs
- ClinStab: Stability-Oriented Learning for Medical Time Series via Dual-Stream Alignment
- CliSurv: Coupling Risk and Survival Distributions in Deep Survival Modeling
- Closed-Form Last Layer Optimization
- Closed-Form Linear-Probe Dataset Distillation for Pre-trained Vision Models
- Closed-form predictive coding via hierarchical Gaussian filters
- Close the Loop: Synthesizing Infinite Tool-Use Data via Multi-Agent Role-Playing
- Closing Open Combinatorial Conjectures with a Solver-LLM-Lean Loop: Three Formalized Theorems in Ramsey and Design Theory
- Closing the Approximation Gap in Simulation-free Latent SDEs
- Closing the Confidence-Faithfulness Gap in Large Language Models
- Closing the Distribution Gap in Adversarial Training for LLMs
- Closing the Loop on Latent Reasoning via Test-Time Reconstruction
- Closing the Loop with Fixed-Point Self-Attention
- Closing the Marginal–Subgroup Calibration Gap in Retrieval Fusion
- CLUE: Correlated Latent Uncertainty for Single-Pass Deep Uncertainty Estimation
- Clustering-aware fine-grained self-supervised learning
- Clustering Evaluation Should Break Epistemic Circularity to Produce Falsifiable Results
- Clustering with Weak Distance Oracles
- ClusterSplat: Semantic Cluster Selection for 3D Visual Grounding in Gaussian Splatting
- CME–SpectrumBench: Can LLMs Analyze Condensed Matter Spectral Data?
- CMKL: Modality-Aware Continual Learning for Evolving Biomedical Knowledge Graphs
- Co3DVG: Occlusion-Aware Cooperative 3D Visual Grounding for Autonomous Driving
- Coarsened Causal Calibration: Conformal Prediction under Interventions without Fine-Grained DAG Recovery
- Coarsening Linear Non-Gaussian Causal Models with Cycles
- Coarse-to-Fine Spectral Cascading for Parameter-Efficient LLM Adaptation
- Coarse-to-Real: Generative Rendering for Populated Dynamic Scenes
- CoCoT-EEG: Contrastive-Pretrained Multiscale Convolutional Transformer for EEG Decoding
- COCOTree: A Dataset and Benchmark for Open Tree-Structured Visual Decomposition
- CODA: Coordinating the Cerebrum and Cerebellum for a Dual-Brain Computer Use Agent with Decoupled Reinforcement Learning
- Code as Policies via Latent Geometry Inference
- Codebook-Guided Cross-Modal Knowledge Distillation for Structurally Heterogeneous Features
- Codebreaker-Bench: Measuring Skill Composition in AI Agents via Sequential Investigation
- Codec-derived Information Priors for Video Token Sparsification in Vision Language Model Inference
- CodeClinic: Evaluating Automation of Coding Skills for Clinical Reasoning Agents
- CodedNN: Coding-Theoretic Redundancy for Robust Neural Classification
- CodegenBench: Can LLMs Write Efficient Code Across Architectures?
- CodeGraderBench: Evaluating LLMs on rubric-based grading of beginner-level source code
- CodeGuard: Auditable Pre-Execution Defense against Repository-Level Poisoning for Coding Agents
- “Code is cheap, show me the specification”: Training LLMs for End-to-End Verified Code Generation
- Code-MT-Bench: Scaling and Evaluating Code Large Language Models in Multi-Turn Dialogues
- CoDeQ: Compression with Dead-Zone Quantizer for Differentiable Joint Pruning--Quantization
- CoDeRNet: Selective Cross-Task Routing under Heterogeneous Supervision for Change Detection and Captioning
- Coding Agent Is Good As World Simulator: A Closer Look
- Coding Agents are Effective Long-Context Processors
- Co-Evolving Interpolants and Flows via Path-Flow Alignment
- Co-Exploration and Co-Exploitation via Shared Structure in Multi-Task Bandits
- Cognitive Alignment: AI Should Align to How Humans Think
- Cognitive Anchoring Systems: Counteracting Social Conformity and Protecting Intellectual Independence in LLMs
- Cognitive constraints and thalamocortical architecture explain systematic biases and neural signatures in human hierarchical decision-making
- Cognitive Models and AI Algorithms Provide Templates for Designing Language Agents
- CogWM: Building Cognitive Worlds through Test-Time Practice
- Coherence-Aware Transition-Intent Fusion for LTL Planning under Uncertain Semantic Maps
- Coherence over Correlation? A Spectral Temporal Kernel for Neural Manifold Learning
- Coherent Behavioral Misalignment Under Sensor Corruption in Embodied VLM Agents
- Coincidence Calculus on Graphs: Max-Entropy Walks, Schrödinger Bridges, and Coalescent Free Energies
- CoLD: Counterfactually-Guided Length Debiasing for Process Reward Models in Mathematical Reasoning
- CoLe: Turning Agentic Context into Dense Training Signals for Reinforcement Learning: A Closer Look
- Collaborative Real2Sim–Sim2Real Agential Learning for Geometric and Generative Human Dynamics
- Collaborative Reasoning Distillation via Cross-Feedback and Coherent Curation
- Collapse Hunter: Tackling the Dimensional Degeneration in Generative Ranking
- COLLAR: Cascaded Object-Level Latent Refinement for High-Fidelity Conditional Generation
- Colored Noise Diffusion Sampling
- ColorimetrYI: High-Throughput Colorimetry of Transparent Samples using Python
- CO-MAP: A Reinforcement Learning Approach to the Qubit Allocation Problem
- COMB: An Encoder Plug-in for Native Position-Independent-Caching
- Combinatorial Bioisosteric Analog Composition via Unbalanced Optimal Transport
- COMET: COmpressed Memory-Efficient Topology for Large-Scale Heterophilic Graphs
- COMET: COoperative Multi-modal Experts with Text-guidance for Multi-Modal Object Re-Identification
- COMET: Decoupled Distillation, Routing, and Capacity Control for Task-Agnostic Continual Vision--Language Learning
- CoMMa: Contribution-Aware Medical Multi-Agents for Decentralized Oncology Decision Support
- Commit Then Explore: Reasoning Models Diversify, Not Converge
- Common-Agency Games for Multi-Objective Test-Time Alignment
- Common Approaches to Spiking Neural Network Training Mask Adverse Sensitivity to Temporal Discretisation
- Commutator-Induced Uncertainty in VAEs
- Comp$^2$VLM: A Hybrid Framework Combining Quantization and Lossless Compression for Efficient Vision-Language Models
- Compact Latent Manifold Translation: A Parameter-Efficient Foundation Model for Cross-Modal and Cross-Frequency Physiological Signal Synthesis
- Compact Representations of Impact-Based Fair-Ranking Policies
- Comparing Explanations is not Enough,Explain the Change: New Standards are Needed to Explain Behavioral Shifts in Large Language Models
- COMPASS: Composable Policy-Amortized Structured Search for LLM-Based Optimization Modeling
- COMPASS: Sample-Efficient Preference Alignment on High-Dimensional Behavioral Manifolds
- Compatible Likelihoods for Flow Matching on Manifolds
- Competing to Discover the Rare: A Game-Theoretic Framework for Multi-Agent Outlier Detection
- Competitive Algorithms for Multi-Agent Ski-Rental
- CompJudge: Fine-Grained Comparative Evaluation using Multimodal LLM for Subject-Driven Generation
- Complementary Cache Guidance with Gradient Disentanglement for Continuous Test-Time Adaptation
- Complementary Preservation–Modification Scoring for Zero-Shot Composed Video Retrieval
- Complexity Aware Continuous Level of Details for Gaussian Splatting
- Complexity Guarantees for Heterogeneous Federated Bilevel Optimization with Lower-Level Constraints
- Complexity-guided Regularization for Generalizable Human Gaussian Splatting
- Complexity-theoretic tradeoffs between precision and expressivity in Transformers
- Compliance-Critical LLM Agents Should Be Specified Above the Impossibility Bound
- Component-Based Out-of-Distribution Detection
- Composable Crystals: Controllable Materials Discovery via Concept Learning
- Composable Latent Action Components for Active Inference Robotic Control
- COMPOSE: Hypergraph Cover Optimization for Multi-view 3D Human Pose Estimation
- CompoSET: A Controlled Single-Edit Benchmark for Vision-Language Compositionality
- Composing Diffusion Priors with Explicit Physical Context via Generative Gibbs Sampling
- Compositional Depth Scaling in Residual Networks
- Compositional Generalization via Geometric Abstractions in Sequential Decision-Making
- Compositional meta-learning through probabilistic task inference
- Comprehensive video captioning via hierarchical multi-dimensional verifiable rewards
- Compress at Input, Recover at Depth: Improving the Efficiency–Performance Trade-off for MLLMs
- Compressed-Domain Attention: Tensor-Core Decode for 4-Bit Codebook KV Caches
- Compressed Sensing Beyond Sparsity: Unique Recovery of Bordered Signals from Underdetermined Measurements
- Compressible Representations: Functional Spines in Deep Neural Networks
- Compressing Collections of Trees with Decision Equivalence
- Compression at a Cost: Interpreting Information Bottlenecks in Safety-Aligned Reward Models
- Compression-Based Actuator Scheduling for Dynamical Systems with Stability Guarantees
- Computationally Efficient Replicable Learning of Parities and Applications
- Compute Efficiency and Serial Runtime Tradeoffs for Stochastic Momentum Methods
- Computing All Optimal Partial $p$-Wasserstein Matchings on the Line
- CompWorld: Compositional Action Control for Video World Models
- Comp-X: On Defining an Interactive Learned Image Compression Paradigm With Expert-driven LLM Agent: A Closer Look
- Conceal, Reconstruct, Jailbreak: Exploiting the Reconstruction--Concealment Tradeoff in MLLMs
- ConceptAlign: A Human-in-the-Loop Large Language Model Framework for Concept Alignment
- Concept-Aligned Neuronal Representation Transfer Across Brain States
- Concept-Based Mechanistic Interpretability Needs a Concrete Evaluation Paradigm
- Concept-Localized Generative Representations
- Concept Modulation Models: A Unified Framework for Identifiability and Extrapolation
- Concepts are not globally linear: contextual geometry of learned representations
- Concepts in Motion: Temporal Concept Bottleneck Model for Interpretable Video Classification
- ConceptTree: Bringing Semantic Transparency to Black-Box Decision Making for Robotic Manipulation
- Concorde: Geometry-Aware Link Prediction via Decoupled Energy Minimization
- Concordia Discors: Phase-Consistent Control for Efficient Disaggregated LLM Serving
- Concord-SLAM: Cross-Render Concordance for Boundary-Native Semantic Gaussian SLAM
- Concurrency without Model Changes: Future-based Asynchronous Function Calling for LLMs
- Concurrent model evidence computation and posterior sampling in continuous attractor network subspaces
- Condensation on Demand: An Adaptive Budget Allocation Framework for Graph Condensation
- CondiFair: Enhancing Fairness with Conditional Visual Reprogramming
- Conditional Attribute Estimation with Autoregressive Sequence Models
- Conditional Compatibility Learning for Context-Dependent Anomaly Detection
- Conditional Counterfactual Mean Embeddings: Doubly Robust Estimation and Learning Rates
- Conditional Dependence Structure in Sparse Autoencoder Features
- Conditional Flow Models Convert Epistemic into Aleatoric Uncertainty
- Conditional independence and graphical models for rankings
- Conditionally Minimal Sufficient Representations for Debiased Visual Learning
- Conditional Phase Diagnostics for Gaussian Random-Effects Panel Samplers: Orthogonalized Interweaving as a Safe Practical Default
- Conditional Ranking of Off-Policy Evaluators in Retrieval-Augmented Generation
- ConFADS: Continuous-Time Forecast-Aware Downscaling for Short-range Weather Forecasting
- ConfDet: Learning Reliable Confidence for MLLM-based Detection
- Confidence-Aware Multi-Teacher Prediction-Powered Semi-Supervised Learning
- Confidence-Aware Tool Orchestration for Robust Video Understanding
- Confidence-Calibrated Temporally Averaged Regression for Semi-Supervised Low-Light Image Enhancement
- Confidence-Conditioned Slot Composition for Structured Retrieval under Noisy Grounding
- Confidence Estimation via Decoupled Smoothing for Dynamic LLM Routing and Aggregation
- Confidence Laundering in Agent Systems: Why Uncertainty Needs a Latent Carrier
- Conflict-Resilient Multi-Agent Reasoning via Signed Graph Modeling
- Conflict-Resolving and Sharpness-Aware Minimization for Generalized Multi-Update Knowledge Editing (Extended)
- Conformal Cache: Reliable Proxy-Discrepancy Caching for Fast Generative Inference
- Conformal Meta-learner for Individual Treatment Effects under Unmeasured Confounding
- Conformal-Triggered Mixture-of-Experts Adaptation for Uncertainty-Aware Inductive Graph Learning
- ConFoThinking: Consolidated Focused Attention Driven Thinking for Visual Question Answering
- Connectome-Wired Reservoirs Align with Vision-Language Models During Abstract Visual Reasoning
- ConQuR: Corner Aligned Activation Quantization via Optimized Rotations for LLMs
- Conservative Continuous-Time Treatment Optimization
- Conservative Image Editing with Intention-Aware Flow Matching
- Conservative Pareto Set Amortization for Offline Multi-Objective Optimization
- ConSFT: Forget Less with Conservative SFT for Flow-Matching VLA Models
- Consistency as a Testable Property: Statistical Methods to Evaluate AI Agent Reliability
- Consistency Training for Aligning Language Models’ Explanations and Behaviors
- Consistent Low-Rank Aggregation for Federated LoRA Fine-Tuning
- Consistent One-vs-All Losses Robust to Misspecification of the Weak Label Transition Model
- Consistent Region-Informed Self-supervised Pretraining
- Constant-Target Energy Matching: A Unified Framework for Continuous and Discrete Density Estimation
- Constrained Code Generation with Discrete Diffusion
- Constrained Diffusion Models with Primal-Dual Inference
- Constrained Diffusion Policy Optimization for Offline Reinforcement Learning
- Constraint, Calibration, and Coordination (C$^3$): Factual Grounding and Evidential Fusion for Holistic and Reliable Multimodal Neurocognitive Disorder Detection
- Constraint Tree Exploration for Learning from Language Feedback
- Consumer Search and Social Learning in Agentic Markets
- Contact-Topology Lock-In in Trajectory Diffusion Models
- Context Binding and Reusable Leakage in Threshold Decryption
- Context-Conditioned Residual Uncertainty for Active Feature Acquisition
- Context Engineering Arises from and Must Evolve with the Human–Machine Intelligence Gap
- Context-Gated Associative Retrieval: From Theory to Transformers
- Context Inference Attacks without Jailbreaks
- Context-Level Differential Privacy for Retrieval-Augmented Generation
- Contextual Agentic Memory is a Memo, Not True Memory
- Contextual Candour: Role-Aware Graduated Disclosure for LLMs over Mixed-Sensitivity Contexts
- Contextualized Hypotheses for Real-Time Online Video Anomaly Detection
- Context Value Informed In-Context Reinforcement Learning
- Context-weighted Discrete Flow Matching
- Continual Learning as Bayesian Filtering
- Continual Learning Bench: Evaluating Frontier AI Systems in Real-World Stateful Environments
- Continual Learning for Enterprise AI Agents
- Continual Learning in the Era of Foundation Models and Embodied Agents
- Continual On-Policy Distillation from Experts
- Continual Self-Improvement with Lightweight Experiential Latent Memories
- Continual World Models
- ContinuLoc: Continuous Pose Inference over Neural Fields for UAV Geo-Localization
- Continuous Audio Thinking for Large Audio Language Models
- Continuous Personalized Diffusion Model via Spinor-Component Forward Geometry
- Continuous-Time Distribution Matching for Few-Step Diffusion Distillation
- Continuous-Time Dynamic Graph Coarsening for Scaling Temporal Graph Neural Networks
- Continuum Mechanics of Flow Matching: Lagrangian Mass Conservation in Finite-Time Probability Transport
- Contract-and-Discretize: An Algebraic Characterization of $k$-FWL GNN Expressivity
- ContractBench: Can LLM Agents Preserve Observation Contracts?: A Closer Look
- Contractive Monoids: The Algebra Behind Stable Residual Propagation in Deep Graph Neural Networks
- Contractive Restoring Flows: Robust Reasoning Distillation via Orbital Stability
- CONTRA-KD: Continuous Trajectory Alignment for Knowledge Distillation
- Contrastive Concept-Tree Search for LLM-Assisted Algorithm Discovery
- Contrastive Distribution Matching for Amortized Sequential Monte Carlo in Discrete Diffusion
- Contrastive Nonmyopic Objective Cost-Tradeoff Acquisition for Longitudinal Data
- Contrastive Pretraining Scales Agentic Exploration
- Contrastive Representation Learning for Exposure-Invariant Image Correction
- Contrastive representation learning for instance-adaptive active feature acquisition
- Contrast-X: A Multi-Modal Contrast Image Synthesis Benchmark and Universal Modality Flow Matching
- Contribution Audits for Diffusion-Assisted Incremental Recognition: Separating Generation, Replay, and Representation Effects
- Control in Stable Matchings via Capacity Transfers and Augmentation
- ControlJEPA: Principled Trajectory Regularization via Lyapunov Tube Loss
- Controllable and Content Based Recommendations
- Controllable Driving Scenario Synthesis Platform for Open-Loop Evaluation
- Controllable Road Marking Generation
- Controllably Efficient Language Models
- Controlla: Learning Controllability via Graph-Constrained Latent Geometry
- Controlled Temporal RSA Reveals Post-Stimulus EEG Alignment with Visual Representational Geometry
- Controlling Temporal Pseudo-Label Marginals for Stable Online Test-Time Adaptation
- Controlling Your Image via Simplified Vector Graphics
- Control Protocols for Mitigating Covert Malicious Finetuning Attacks from Internally Deployed Agents
- Control-Space Projection for Zero-Shot Constrained Diffusion via Pontryagin's Maximum Principle
- Control-TS: Uncertainty-Guided Diffusion for Probabilistic Time Series Forecasting
- Control Your Queries: Heterogeneous Query Interaction for Camera-Radar Fusion
- Convergence of Consensus-Based Particle Methods for Nonconvex Bi-Level Optimization
- Convex-Geometric Error Bounds for Positive-Weight Kernel Quadrature
- Convexity in Disguise: A Theoretical Framework for Nonconvex Low-Rank Matrix Estimation
- CoolData: Benchmarking Machine Learning Methods for Electronics Cooling
- Cool Graphs: Active Property Search Towards Quantum Nano-Refrigerators
- CoopLight: Cooperative Vision-Language Agents for Long-Tail Traffic Control
- Coordinate-Wise Adaptive Optimization Under Heavy-Tailed Gradient Noise
- CoPE-GL
- CoPeP: Benchmarking Continual Pretraining for Protein Language Models
- COPRA: Conditional Parameter Adaptation with Reinforcement Learning for Video Anomaly Detection
- Copying Under Repetition: From Prefix Matching to Position Hashing
- CORAL: In-Context Inverse-Operator Generation via Causal Low-Rank Accumulation
- CORA: Per-Slice Coherent Orthogonal Rotation for SVD-based Low-Rank Adaptation
- CORE-BREW: LLR-Based Soft Decoding for Robust Multi-Bit LLM Watermarking
- CORE: Contrastive Reflection Enables Rapid Improvements in LLM Reasoning
- CoreCover: Risk Constrained Graph Recommendation via Probabilistic Core Promotion Coverage
- CoreErase: Inversion-Aware Core Region Concept Erasure
- CORP: Closed-Form One-shot Representation-Preserving Structured Pruning for Transformers
- CorrCLR: Correlation-based Contrastive Learning of Representations
- Correct Before It Drifts: Timely Intervention on Partial Reports to Mitigate Factual Hallucinations in Radiology
- Correct but Unselectable: The Hidden Interface Tax in Multi-Candidate Reasoning
- Correcting Diffusion Hallucinations via Uncertainty-guided Trajectory Steering
- Correct Is Not Enough: Training Reasoning Planners with Executor-Grounded Rewards
- Correlation-Aware and Gaussianity-Preserving Robust Latent Angular Watermarking for Diffusion Models
- Correpondence Alignment For Improved Virtual Try-On
- Correspondence Pruning by Iterative Structural Rectification
- CORTEG: Foundation Models Enable Cross-Modality Representation Transfer from Scalp to Intracranial Brain Recordings
- Cortically-Regularized Transfer of Expert Adapters for Cross-Dataset Aesthetic Assessment
- CoScan: Multi-Scale Content-Adaptive Space-Filling Scans for Causal State-Space Image Restoration
- CoSG: Counterfactual Explainers for Signed Graphs
- Cosine-Gated Adam-Decay: Drop-In Staleness-Aware Outer Optimization for Decoupled DiLoCo
- COSMO-Agent: Tool-Augmented Agent for Closed-loop Optimization, Simulation, and Modeling Orchestration in Industrial Design
- Cost-Aware Adaptive Conformal Inference for Runtime Assurance in Dynamic Environments
- CoTEgo: A Benchmark of Egocentric State Reasoning and Chain-of-Thought under Recursive Manipulation
- CoT-Guard: Small Models for Strong Monitoring
- CoTrek: Toward Scalable On-Policy Distillation for Long Chain-of-Thought Reasoning
- CounterCount: A Diagnostic Framework for Counting Bias in Vision Language Models
- Counterexample-Guided Shortcut Suppression without Shortcut Labels
- Counterfactual Discovery of Motion-Effect Circuits in Video Diffusion Transformers
- Counterfactual Distillation: Internalizing Reflective Experience into LLM Agent
- Counterfactual Explanations for Time-Series Classification via Constrained Flow Matching
- Counterfactual Identifiability for Dynamical Systems via Schrodinger Bridges
- Counterfactual Maps: What They Are and How to Find Them
- Counterfactual Niche Fields: Learning Editable Microenvironment Representations for Spatial Omics
- Counterfactual Rollout Replay: Forkable Environments as Free Process Rewards for Software Engineering Agents
- Counting without Scale: Scale-Consistent Error Correction for Crowd Counting
- Coupled Integral PINN for Discontinuity
- Coupling-Aware Reinforcement Learning for Co-Evolving Graph Games
- Coupling Models for One-Step Discrete Generation
- Coupling Noisy Pairwise Knowledge to the DAG Posterior for Causal Discovery
- Covariance-Aware Spectral-Filter Widths for Graph-Gaussian Signals
- CoVER: Group Relative Policy Optimization with Code-Verified Process Rewards in Stateful Code Execution Environments
- Covering Human Action Space for Computer Use: Data Synthesis and Benchmark
- Cover meets Robbins while Betting on Bounded Data: $\ln n$ Regret and Almost Sure $\ln\ln n$ Regret
- CoVisIT: Cross-modal Prior Guided Diffusion Model for Visible-to-Infrared Image Translation
- CPCANet: Deep Unfolding Common Principal Component Analysis for Domain Generalization
- CPGBANK: A Multimodal Benchmark for LLM Adherence to Clinical Practice Guideline Flowcharts
- CPO: Out-of-Distribution Detection with Contrastive Pseudo-Outliers
- CPO-RL: CoT-Policy Co-Optimization for Embodied Reinforcement Learning
- CPR: Causal Physiological Representation Learning for Robust ECG Analysis under Adversarial Perturbations
- CPRKF: A Consistency-based Possibilistic Robust Kalman Filter
- CRA-Bench: Benchmarking Conversational Recommender Agents under Ambiguous User Intent
- CRaFT: A Causality-Guided Approach to Missing Value Imputation in Time Series
- CRAFT: Corrector-Retrieval Augmented Fine-Tuning for Time Series Forecasting
- Crafting Reversible SFT Behaviors in Large Language Models
- CREATE: Testing LLMs for Associative Creativity
- Creating Multi-Layer Belief Systems in AI: A Structured Belief Stack for Gradual Belief Updating, Coherence, and Alignment
- CRE: Continuous 2D rotation equivariance on grids
- Credit Bandwidth Lower Bounds for Diffusive Cortical Learning
- CRISP: Calibrated Routing for Imbalanced Selective Prediction under Long-Tailed Data
- CriterAlign: Criterion-Centric Rationale Alignment for Code Preference Judging
- Criticality and Saturation in Orthogonal Neural Networks
- CRoCoDiL: Continuous and Robust Controlled Diffusion for Language
- CRONOS: Benchmarking Counterfactual Physical Consistency in Video Models
- Cross-Agent Campaign Attribution: Linking Asynchronous Attacks Across LLM Agents
- Cross-Attention and Encoder–Decoder Transformers: A Logical Characterization
- Cross-Cell-Line Perturbation Prediction Needs Controls
- Crosscoding Through Time: Sparse Feature Discovery Across Sequence Positions
- Cross-Cohort Generalization in Behavioral Data Requires Semantic Anchoring
- CROSS: Constrained Residual Optimization for Subspace Separation in Few-Shot Class-Incremental Learning
- Cross-Epoch Adaptive Rollout Optimization for RL Post-Training
- Cross-Fidelity Generalization in Quantum Circuit Classification: When Data Diversity Beats Contrastive Learning
- Cross-Hospital Transfer Peaks Before Maximum Alignment in Frozen Pathology Foundation Models
- Cross-Layer Evolution Graph Learning for Fine-grained VLM Hallucination Detection
- Cross-Modal Prior-Guided Training with Visual Foundation Models for Unsupervised LiDAR Point Cloud Registration
- Cross-Modal Unified Generation Framework for Fine-Grained Visual-Textual Alignment
- Cross-model activation monitoring via relative probing
- Cross-Model KV Cache Transfer in LLM Families: A Closed-Form Linear Mapping for Prefill Reuse
- CrossPoE: Task-Signal-Preserving Latent Translation for Robust Multimodal Learning Under Modality Missingness
- Cross-Question Reliable Reinforcement Learning
- Cross-Scale Latent Predictive Pretraining for Whole-Slide Image
- Cross-System Predictability Ordering via Gauge-Fixed Ordinal Network
- CrossWeave: Emergent Cross-Modal Scene and Instance Retrieval from Sparse 2D-3D Alignment
- CRPS-LAM: Probabilistic Regional Weather Forecasting with Continuous Ranked Probability Score
- Crucible: Emergent Deception in LLM Social Dilemmas Through Private Reflection
- CrudeOilMix: A Million-Scale Multimodal Benchmark for Crude Oil Characterization
- CryoPilot: Scientific Workflow Memory for Dual-Mode Automation of High-Throughput Cryo-EM
- Crystalite: A Lightweight Transformer for Efficient Crystal Modeling
- CSLA: Sparse-Linear Attention with Learnable Routing and Quantization-aware Training
- CTMAD: Causal Transformer for Multi-Scale Attack Detection in Cross-Chain Bridge Protocols
- CTRL: Continual Test-Time Reinforcement Learning for Large Language Models
- CtrlFormer: Efficient Non-Interactive Secure Transformer Inference without Feature Explosion
- Ctrl-VI: Controllable Video Synthesis via Variational Inference
- CTTS: Collective Test-Time Scaling
- CUDABeaver: Benchmarking LLM-Based Automated CUDA Debugging
- CUDAHercules: Benchmarking Hardware-Aware Expert-level CUDA Optimization for LLMs
- CuFF: A Collaborative and Unified Framework for Few-Shot Visual Recognition
- Cumulative Meta-Learning from Active Learning Queries for Robustness to Spurious Correlations
- CupOFMoCA: Coupled Objective-Guided Discrete Flows for Molecular Conjugate Assembly
- CURe: Conservative Unlearning with Soft-Gating Regularization for Offline Reinforcement Learning
- CURE: Coupled User-Grouped Reinforcement Learning for Cross-Domain Recommendation with Non-Overlapping Users
- Curriculum-Driven Degradation-Aware Diffusion Transformers for Real-World Old-photo Face Restoration
- Curriculum Learning-Guided Progressive Distillation in Large Language Models
- Curriculum Reinforcement Learning Can Incentivize Reasoning Capacity in LLMs Beyond the Base Model
- Curvature Allocation Dynamics in Deep Learning: Sharpness, Magnitude and Spectral Concentration
- Curvature-Dependent Generalization Bounds for Learning on Riemannian Manifolds
- Curvature-Dependent Lower Bounds for Frank-Wolfe
- Curvature Determines the Fate of Compounding Error in Imitation Learning
- Curved Trajectories in Transformer Representations
- Cutting-Plane Attention: Geometric Feasible-Region Refinement for Expressive and Universal Transformer Architectures
- CyberEvolver: Structured Self-Evolution for Cybersecurity Agents On the Fly
- CycleFormer: Reconstructing Transformer with Periodic Components for Multivariate Time Series Forecasting
- Cyclic Denoising Reveals Ultrastable Memories in Diffusion Models
- Cylindrical Geodesic Flow Matching for Quasiperiodic Physiological Signal Transformation
- D$^2$Quant: Accurate Low-bit Post-Training Weight Quantization for LLMs
- D4SAT: Deep Cooperation of CDCL and Discrete Diffusion Model for SAT
- DAB-Attack: Defense-Agnostic Persistent Backdoors in Federated Learning
- DACA-GRPO: Denoising-Aware Credit Assignment for Reinforcement Learning in Diffusion Language Models
- DAD4TS: Data-Augmentation-Oriented Diffusion Model for Time-Series Forecasting with Small-Scale Data
- DAGE: Differentiable Design of Context-Specific Combinatorial Gene Perturbations for Cellular Rejuvenation
- DAG topological generalization for inductive transfer learning of quantum fidelity via Graph Transformer
- DAPAI: Dual-Level AMR-Enhanced model for Perturbation-Robust AI-Generated Text Detection
- DARE: Difficulty-Adaptive Reinforcement Learning with Co-Evolved Difficulty Estimation
- DARK: Diagonal-Anchored Repulsive Knowledge Distillation for Vision-Language Models under Extreme Compression
- DarkVGGT: Seeing Through Darkness Using Thermal Geometry without Daylight Tax
- DARLING: Detection Augmented Reinforcement Learning with Non-Stationary Guarantees
- DARS: Dynamic Alignment Reward Steering for Robust Adaptive Reasoning
- Darwinian Competition in Representation Space: A Causal Framework for Attention Head Fitness in Transformer Language Models
- DashengTokenizer: One layer is enough for unified audio understanding and generation
- DASNR: One-Shot Domain-Aware Structural Neuron Removal for LLMs
- Data Density Scaling Laws for Image Self-Distillation
- Data-Driven LinUCB with Near-Optimal Regret
- Data-driven Projection for Solving Semidefinite Programming Problems
- Data-Driven Soft Labeling Scales DNA Read Classification to Whole-Body Cell-Type Deconvolution
- Data-Efficient Autoregressive-to-Diffusion Language Models via On-Policy Distillation
- Data Highways for LLMs: Global Path Search for Data-Aware Routing and Pruning
- Data Needs Tuning Too: Reassessing Machine Learning Performance through a Data-centric Lens
- Data Poisoning via User-Generated Content Poses a Systemic Vulnerability in GenAI
- Dataset Collections: Challenges of Large-Scale Data Aggregation in 3D Medical Image Datasets
- Dataset Publishing as a Mechanism for Controlling Task Affinity
- DataValve: Selecting-while-Training for Efficient Visual Instruction Tuning
- Davinci-Math: A Routing-Aware Data Pipeline for End-to-End Mathematical Reasoning
- DCEdit: Dual Chain-of-Thought with Textual and Visual Reasoning for Face Image Editing
- DCI
- DCRL: Decoupling and Coupling Reinforcement Learning via Policy-Reward Manifold Alignment
- DC-VLAQ: Query--Residual Aggregation for Robust Visual Place Recognition
- DDCSC: Efficient and Accurate LLM Reasoning via SPRT-Grounded Contrastive Probabilistic Framework
- DDGE: Disentangled Dirichlet Geodesic Evaluation for Robust Few-Shot Learning
- DDTime: Dataset Distillation with Spectral Alignment and Information Bottleneck for Time-Series Forecasting
- DDx-TRACE: A Benchmark for Medical Diagnostic Trajectories in VLMs
- Debiased Counterfactual Generation via Flow Matching from Observations
- Debiasing Message Passing to Mitigate Popularity Bias in GNN-based Collaborative Filtering
- Decentralized $\mu^2$-SGD: Narrowing the Parallelism Gap to Centralized Learning
- Decentralized RAG via Locality-Preserving Hashing and Progressive Routing
- DECEPT-Bench: A Multi-Mechanism Behavioral Audit of Language-Model Deception
- Deceptive Algorithm Research Must Be Encouraged and Governed by What It Enables, Not by Who Conducts It
- Decision-Aware Proximal Bridge Learning for Optimal Treatment Selection
- DecisionBench: Benchmarking Skill-Aware Emergent Orchestration in Long-Horizon Agentic Workflows
- Decision-Preserving Compression for Compact World Models
- Decision-Weighted Flow Matching for Contextual Stochastic Optimization
- Declarative Data Services: Structured Agentic Discovery for Composing Data Systems
- Decodability ≠ Steerability: Path-Faithfulness Bounds Where Activation Steering Can Work
- Decoding Alignment without Encoding Alignment: A Critique of Similarity Analysis in Neuroscience
- Decomposed Representations Mitigate the Alignment–Specificity Trade-off in Multi-Omics
- Decomposing Effects in Neural Causal Models
- Decomposing the modulation of interactions between neuronal populations
- DecompressionLM: Deterministic, Diagnostic, and Zero-Shot Concept Graph Extraction from Language Models
- DeCoRL: Decomposed Consistency Reinforcement Learning for Multi-Image Composition
- DeCoS: Decoding by Contrasting Steered Features Reduces Hallucinations in Large Language Models
- Decoupled Intensity Modeling for Neural Temporal Point Processes
- Decoupled Motion-Modality Aggregation for Skeleton-Based Micro-Action Recognition
- Decoupling Baseline from Condition in Single-Cell Perturbation Data with Split Deep Generative Decoder
- Decoupling Connectivity from the Optimization Objective in Graph Clustering
- Decoupling Time and Space in Selective CDEs Using Tensor Train Decomposition
- Decoy Direction Optimization: Mechanistic Weight Editing Against LLM Abliteration
- Deductor: Turning Evaluation Requests into Auditable Benchmark Workflows
- Deep-ARViM: A Perceptual Visibility Metric and Benchmark for Augmented Reality Displays
- Deep Common Principal Component Analysis
- Deep Ensembles for Epistemic Uncertainty: A Frequentist Perspective
- DeepImagine: Clinical Trial Outcome Prediction via Stepwise Local Counterfactual Imaginations
- Deep Learning-based Algebraic Reynolds Stress Closures for RANS simulations of Turbulent Flows
- Deep Minds and Shallow Probes
- Deep Predictor-Corrector Networks for Robust Parameter Estimation in Non-autonomous System with Discontinuous Inputs
- DeepReviewer 2.0: A Traceable Agentic System for Auditable Scientific Peer Review
- DeepTumorVQA: A Hierarchical 3D CT Benchmark for Stage-Wise Evaluation of Medical VLMs and Tool-Augmented Agents
- Default Feature Representations of the Cognitive Map
- Defeating the Training-Inference Mismatch via FP16
- Defense Against Prompt Inversion Attacks: An Information-Theoretic Approach for LLM Collaborative Inference
- Deferred Contextualization for Efficient Long-Context KV Reuse
- Deformable 2D Gaussian Splatting
- DEFT: Disentanglement-Enhanced Fine-Tuning for EEG Foundation Models
- Déjà View: Looping Transformers for Multi-View 3D Reconstruction
- Delayed homomorphic reinforcement learning for environments with delayed feedback
- Delayed Robustness Is Not Grokking
- DEL: Digit Entropy Loss for Numerical Learning of Large Language Models
- Delightful Distributed Policy Gradient
- Delta Attention Residuals
- DeltaFugue: Orchestrating Spatial and Associative Memory for Algorithmic Length Generalization
- DELTA-TTS: Adapting Autoregressive Model into a Diffusion Language Model for Text-to-Speech
- Democratizing Tool Learning with Environments Fully Simulated by a Free 8B Language Model
- Demonstrating Generalization Failures via Mixtures of Conditional Policies: A Closer Look
- Demystification of a class of activation functions
- Demystifying Variance in Circuit Discovery of LLMs
- DenoSP: Selective Prediction for Diffusion Language Models via Calibrated Denoising Risk Scores
- De Novo Design of Protein Switches with Diffusion-Based Ensemble Sampling
- Dense2MoE: Pushing the Pareto Frontier of On-Device LLMs via Unified Pruning and Upcycling
- Density-Ratio Losses for Post-Hoc Learning to Defer
- DePHI: Dynamically Evolving System Framework for PHI-Safe and Auditable LLM Generation
- DePICT: Decision-Preserving Interface for Constrained Downstream Tasks
- Depth Completion Decoders are Linear System Solvers: From Neumann Series to State-Dependent Recurrence
- Depth-CoT Tradeoffs in Transformers: A Look at Path Following
- Depth or Tokens, Not Parameters: A Theory of Sequential Reasoning via Chain Distance
- Depthwise Flow Signatures for Hallucination Detection and Refinement in Large Language Models
- Derivative-Informed Training of Neural Operators On-the-Fly via Sketched Tangent Consistency
- DermVerse-500K: A Large-Scale Expert-Annotated Dataset with VLM Assisted Clinical Feature Extraction
- Design Conditions for Intra-Group Learning of Sequence-Level Rewards: Token Gradient Cancellation
- Designing Effective Monitor-Based Interventions for Mitigating Reward Hacking During RL
- DeSQ: Dead-Zone-Aware Joint Sparsification and Quantization for LLMs
- Detect, Correct, Abstain: Conformal Control of Hallucinations in Diffusion Inverse Problem Solvers
- Detecting Deception, Not Deepfakes: Why Media Forensics Needs Social Theories
- Detecting Deliberative Convergence in Multi-Agent LLM Systems: A Geometric and Relational Saturation Framework
- Detecting LLM Hallucinations with Manifold-Calibrated Pseudo-Supervision
- Detect the Composition, Not the Pixel: A Position on Forensic Media Verification
- DetectViT: Test-time Backdoor Detection for Vision Transformers via Inter-Head Attention Discrepancy
- DetM-Tracker: A Deterministic and Memory-Efficient VIS Framework for Industrial Autonomous Inspection
- DETS: An Interval-Censored Evidential Sampling Framework for Cross-Domain Scientific Discovery
- DevAI: Developmental Perspectives on AI
- DeVI: Physics-based Dexterous Human-Object Interaction via Synthetic Video Imitation
- DeVOT: Decomposed Variational Optimal Transport for Task-Free Continual Learning
- Dex2HOI: Dexterous Bimanual Two-Object Interaction Generation
- DexAdapter: Learning Perceptive Physics-aware Retargeting for Dexterous Manipulation
- DexOPE: 6D Object Pose Estimation in Dexterous Manipulation
- Dexterous Point Policy: Learning Point-based Dexterous Hand Policies from Human Demonstrations
- Dexterous Skill Discovery via Topology-Aware Wasserstein Dependency
- D-GAP: Improving Out-of-Domain Robustness via Dataset-Agnostic and Gradient-Guided Augmentation in Amplitude and Pixel Spaces
- DGN: Disagreement Graph Networks for Robust Learning from Multiple Annotators
- DGPO: Directional Gain Policy Optimization for Fine-Grained Credit Assignment in Reasoning
- DGRAF: Observation-Quality-Aware Reinforcement Learning for Dynamic Reconfigurable Batteries
- DHAFT: Mitigating Stage Misalignment in Soft Prompt Tuning for Coherent LLM Reasoning
- DIABLO & PAWS: Parameter-Space Detection and Obfuscation of Encrypted Backdoors: A Closer Look
- Diagnose, Don't Patch: A Falsifiable Framework for Sample-Efficient World Models
- Diagnosing Alignment Through Geometry: SPD Analysis of LLM Persona Structure
- Diagnosing and Mitigating Modality Interference in Multimodal Large Language Models
- Diagnosing and Repairing Post-Construction Mismatch in Sparse Transformer--SSM Hybrids
- Diagnosing and Repairing Visual Collapse in Compact Medical Multimodal LLMs
- Diagnosing Prediction Fragility in Image Classifiers with Degradation Curves
- DiagnosticIQ: A Benchmark for LLM-Based Industrial Maintenance Action Recommendation from Symbolic Rules
- Diagonalizing the Softmax: Hadamard Initialization for Tractable Cross-Entropy Dynamics
- DialBandit: Adaptive Sequential Search with Tunable Evaluation Fidelity
- DIAMOND: Directed Inference for Artifact Mitigation in Flow Matching Models
- DIBench: Benchmarking Decision Integrity of GUI-based Mobile Agents Under Deceptive Injections
- DICE: Decoupling Capability from Intervention Necessity in LLM Tutoring
- DICEQuant: Distortion-Compensated Rounding with Dual-Ended Shrinkage for LLM Quantization
- Dictionary Based Pattern Entropy: A Causal Discovery Framework for Symbolic Sequences
- DiDE: Direct Injection with Color-Texture DEcoupling for 3D Stylization
- Diff3R: Feed-forward 3D Gaussian Splatting with Uncertainty Aware Differentiable Optimization
- DiffAtom: Learning Physical Atoms for Flexible Off-Grid Compressed Sensing
- Diff-CA: Separating Common and Salient Factors with Diffusion Models
- Diffeomorphic Optimization
- Difference of Convex Programming in the Wasserstein Space with Applications to MMD Optimization
- Differentiable Bit-Widths: Co-optimizing Pruning and Quantization via SVD for Ultra-Efficient LLM Compression
- Differentiable DAG Discovery under Structured Shift Interventions
- Differentiable Escape Moves over Permutations
- Differentiable Exact Learning of Algorithms
- Differentiable Ray Tracing with Gaussians for Unified Radio Propagation Simulation and View Synthesis (Extended)
- Differentiable Topic-Based Soft Contrastive Learning for Heterogeneous Representation Learning
- Differential Attention for State Space Models
- Differential Harm Propensity in Personalized LLM Agents
- Differentially Private Clipping-Free Stochastic Sign Descent
- Differentially Private Empirical Cumulative Distribution Functions
- Differentially Private Federated Averaging with James-Stein Estimator
- Differentially Private Inference for Geodesic Regression on Riemannian Manifolds
- Differentially Private Model Merging
- Difficulty-Adaptive Reward Normalization for Multi-Reward Policy Optimization
- Difficulty-Aware Reweighting for Zeroth-Order Fine-Tuning of LLMs
- DiffPDHG: A Primal-Dual Plug-and-Play Method with Diffusion Priors for Inverse Problems
- DiffPtr: Diffusion Pointer Networks for Multi-Agent Combinatorial Optimization
- Diff-Target: Target-Specific Causal Path Estimation with Counterfactual Diffusion
- Diffuse and Bound
- DiffuseGene: Aligning Identity Injection with Diffusion Dynamics for Kinship Face Synthesis
- Diffusion-based 3D Human Pose Estimation on LiDAR Point Clouds
- Diffusion-based Dual Variance Control for Partition-Robust Open Set Wireless Signal Recognition
- Diffusion-based language models capture temporal structure in neural dynamics of speech production
- Diffusion Based Symbolic Regression: A Closer Look
- Diffusion Editing with Soft Mask: Pixel Level Redo of Image and Video with Adjustable Strength
- Diffusion Fine-Tuning: Iterative Refinement for Advanced Grounding with Diffusion Large Language Models
- Diffusion Guidance Is a Controllable Policy Improvement Operator
- Diffusion Language Models Are Natively Length-Aware
- Diffusion Language Models: Foundations, Efficiency, and Reasoning
- Diffusion LLMs are Natural Adversaries for any LLM
- Diffusion Masked Pretraining for Dynamic Point Cloud
- Diffusion Meta-Prompting and Steering for Generalizable Foundation Model Adaptation
- Diffusion-State Policy Optimization for Masked Diffusion Language Models: A Closer Look
- Dimensionality Reduction for Robust Federated Learning: A Theoretical Analysis and Convergence Guarantee
- Dimension-free Bounds for Covariance Estimation with Tensor-Train Structure
- Dimension-Free Convergence Rates for Stochastic Interpolants under Arbitrary Couplings
- DINA: Disentangled Noise Abduction for Counterfactual Image Generation
- DINO-SOAR: Multi-Level Photometry via Semantic Outlier Anchor Routing for Anomaly Detection
- DIP-GS: Deep Image Prior For Gaussian Splatting Sparse View Recovery
- DIPHINE: Diffusion-based $\Phi$-ID Neural Estimator
- DIPrune: Task-Aware Token Pruning with Dual Importance for Efficient Multimodal Language Models
- Direct Contextual UCB under Exact Cauchy Noise
- Direct Estimation of Schrödinger Bridge Time-Series Drifts: Finite-Sample, Asymptotic, and Adaptive Guarantees
- Directional Confusions Reveal Divergent Inductive Biases Through Rate-Distortion Geometry in Human and Machine Vision
- Directional Consistency as a Complementary Optimization Signal: The GONO Framework
- Directional Faithfulness in Learned Discretization for Geometric Representation Learning
- Directional Personalization: Preference Geometry and Multi-Team Decision Making
- Directional Reasoning Trajectory Change (DRTC): Identifying Critical Trace Segments in Reasoning Models
- Direction-Aware Low-Resource Multilingual Document Parsing with Heterogeneous Synthesis
- Direct Preference Density Alignment for Conversational Audio Equalization
- DiRecT: Safe Diffusion-Based Planning via Receding-Horizon Denoising
- DirectSpeech2LLM: A Simple End-to-End Framework to Mitigate Prompt Overfitting in Speech-LLMs
- DirectUV: Image-Conditioned UV Texture Generation with Surface-Aware Positional Encoding
- DIS-Bench: Evaluating LLMs on System Testing via Directed Input Synthesis
- DISCO: Dispersion-Guided Sparse KV Cache Compression for Efficient Long-Context Inference in Large Language Models
- DiscoLoop: Looping Discrete Embeddings and Continuous Hidden States for Multi-hop Reasoning
- Discovering dynamical parameters of synthetic multicellular systems from image sequences
- Discovering Interpretable Failure Modes of Vision Language Models
- Discovering KV Cache Eviction Policies via LLM-Guided Program Evolution
- Discovering Learning-Friendly Generation Orders for Sequential Computation
- Discovering Multi-Agent Learning Algorithms with Large Language Models
- Discovering Structurally Plausible and Interpretable Cognitive Models with Large Language Models
- Discovering What You Can Control: Interventional Boundary Discovery for Reinforcement Learning
- Discrete Bridges for Mutual Information Estimation
- Discrete Diffusion Models with Interpolation-Based Controllable Resampling
- Discrete Flow Matching: Convergence Guarantees Under Minimal Assumptions
- Discrete MeanFlow: One-Step Generation via Conditional Transition Kernels
- Discrete Resonance Channels in Selective State Space Models
- Discrete vs. Continuous Action Spaces for Portfolio DRL: A Walk-Forward Benchmark
- Discretizing Continuous Time Series for Imputation with Masked Diffusion Training
- Discriminating Out-of-Distribution Actions in Offline Reinforcement Learning via Quantile Advantage
- Disentangled Learning Improves Implicit Neural Representations for Medical Reconstruction
- Disentangled Multimodal Learning for Scalable Dynamic IR-Drop Analysis
- Disentangled Sparse Representations for Concept-Separated Diffusion Unlearning
- Disentanglement with Holographic Reduced Representations
- Disentangling Channel Semantics in Vision Transformers via Token Decorrelation and Composition-Aware Modulation
- Disentangling KL Direction and Prefix Source in LLM Distillation
- Disentangling Latent Representations by Groups, not Dimensions, via Partial Correlation (Extended)
- Disentangling Self-Preservation in Language Models: Post-Training Gating, Geometric Structure, and Koan-Derived Agentic Steering
- Disentangling Semantics and Acoustics in Low-Bitrate Text-Aligned Speech Tokenization
- DishTKG: Bridging Graph Structures and Large Language Models via Discrete Representation Alignment for Temporal Knowledge Graph Reasoning
- Dissecting LLM-Generated CUDA: Proposal, Search, and Environment Compatibility
- Dissecting Long-range Dependency in Graph Implicit Models
- Dissipative Liquid Alignment: Thermodynamic Stability Guarantees for Robust Vision-Language Learning
- Dist$^2$ill: Distributional Distillation for One-Pass Uncertainty Estimation in Large Language Models
- Distance-Aware Muon: Adaptive Step Scaling for Normalized Optimization
- Distance Marching for Generative Modeling
- Distill AR to Diffusion for effiecient vla for autonomous driving
- Distillation-Guided Modality Dropout for Robust Token Fusion Under Non-Registered Modality Absence
- DistillBench: An Evaluation Benchmark and Framework for Dataset-Level Training Data Attribution (Extended)
- Distill'em All: Data-Free Flow Matching via Discretization Density Mixing
- Distilling Bayesian Uncertainty into a Single Forward Pass
- Distilling Reliable Guidance from Suboptimal Trajectories via Hierarchical Expert-Calibrated Alignment
- Distilling Sequential Computation in Transformer Language Models
- Distilling Visual Chains of Thought into Text Space
- Distilling What Matters: Confidence-Aware Selective Distillation for Large Language Models
- Distill-on-Call: Energy-Aware Dynamic Distillation for Serverless LLM Inference
- Distributional Alignment of Generative Models via Dual-Reward Optimization
- Distributional Equivalence of Kernel $k$-means and Spectral Clustering
- Distributional Evaluation of Generative Models via Relative Density Ratio
- Distributionally Robust Decision-Making under Reject Inference Context
- Distributionally Robust Listwise Preference Optimization
- Distributionally Robust LLM Augmentation for Discrete Choice Models (Extended)
- Distributionally Robust Multi-Objective Optimization
- Distributionally Robust Multi-Task Reinforcement Learning via Adaptive Task Sampling
- Distributionally Robust Optimization under Overlapping Contamination via $\gamma$-Divergence
- Distributionally-Robust Policy Learning from Observational Data
- Distributional Multi-Objective Reinforcement Learning with Consistency-Model Critics
- Distribution Corrected Decision Transformer for Offline Reinforcement Learning with Imbalanced Datasets
- Distribution-informed Efficient Conformal Prediction for Full Ranking
- Distribution-Preserving Dataset Distillation Using General-Purpose Diffusion Model
- Distribution Shift in Missing Data Imputation: A Risk-Based Perspective and Importance-Weighted Correction under MAR
- Dis-WSPO : Incorporating Dispreferred Samples Enhances Weak-to-Strong Preference Optimization
- DiT4DiT: Jointly Modeling Video Dynamics and Actions for Generalizable Robot Control
- DiTTo: Scalable Order-aware All-in-One Image Restoration Agent
- Divergence between Gradient Backpropagation and the Human Brain during Image Processing
- Diverse Sampling in Diffusion Models with Marginal Preserving Particle Guidance
- Diversity Curves for Graph Representation Learning
- Diversity Maximization: Algorithms for Distant $k$-Subsets
- Div, Grad, Curl, and All That: Electrostatic Models for Score Matching
- Divide-and-Conquer CoT: RL for Reducing Latency via Parallel Reasoning
- DLR: Zero-Inference-Cost Latent Residuals for Low-Rank Pre-Training
- D-model and D-Module: Structure-Sensitive Admission for Plug-in Control of LLM Representation Evolution
- DNPMorph: Direct Noise-Path Optimization for Smooth and Consistent Diffusion Morphing
- Do Composed Image Retrieval Benchmarks Require Multimodal Composition?
- DocReward: A Document Reward Model for Structuring and Stylizing
- DocScope: Benchmarking Verifiable Reasoning for Trustworthy Long-Document Understanding
- Dodge-It: Learning Collision-Aware VLA Models for Robotic Manipulation
- Does AlphaFold Understand Physics or Memorize Templates?
- Does an LLM Know Before It Speaks? Prompt-Side Attention Topology for Pre-Generation Error Detection
- Does Certified Robustness Generalize? A Neural Network Verification Study of Distribution Shift
- Does Latent Count? Introducing Negative Binomial Recurrent Explicit Duration Switching Linear Dynamical System
- Does My LLM Handle Indeterminacy Well? An Omnimodal Evaluation Across Uncertainty Triggers and Response Strategies
- Does Seeing More Mean Knowing More? Mono-Anchored Advantage Normalization for Multi-Source Visual Reasoning
- Does the Question Really Matter? Training-Free Data Selection for Vision-Language SFT
- Does Weight Decay Enhance Training Stability?
- Does Your Large Language Model Have An Intuitive Sense of The Difficulty of A Question?
- Dogged Backdoor Attacks: On the Incompleteness of Machine Unlearning in Diffusion Models
- Do Language Models Dream of Binding Molecules? Benchmarking LLMs under Spatial Constraints
- Do LLMs Know When They Hallucinate? A Representation Learning Approach
- Domain Generalization in-the-Wild: Disentangling Classification from Domain-Aware Representations
- Domain-Informed Conditional Fields for Plasma Etching Surrogate Modeling
- Domain Knowledge Matters: Energy-Guided Tabular Data Generation
- Done, But Not Sure: Disentangling World Completion from Self-Termination in Embodied Agents
- Do Neural Networks Lose Plasticity in a Gradually Changing World?
- Do Not Let Spikes Flip: Margin-Resculpted Learning for Robust Spiking Neural Networks
- Don't Adapt Attention: Spectral Evidence for MLP-Only LoRA
- Don't Call LLM Social Simulators (In)valid Without Saying at What Aspect
- Don't Discard Your Rollouts: Reusing Teacher RL Traces for Student Distillation
- Don't gamble, GAMBLe: An Analytical Framework for AI-Driven Research Systems
- Don’t Learn What You Can Compute: Arithmetic Residual Blocks for Exact Arithmetic in Transformers
- Don’t Let Bandit Feedback Pull Continual LLM-Recommender Updates Off Target
- Don't Match the Noise: Distribution Matching under Unknown Measurement Error with an Audit Sample
- Don't Stop Me Yet: Sampling Loss Minima via Dissipative Riemannian Mechanics
- Dooly: Configuration-Agnostic, Redundancy-Aware Profiling for LLM Inference Simulation
- DoPE: Denoising Rotary Position Embedding
- DORA: Dynamic Online Reinforcement Agent for Token Merging in Vision Transformers
- D-ORCA: Dialogue-Centric Optimization for Robust Audio-Visual Captioning
- Do Reasoning LLMs Refuse What They Infer in Long Contexts?
- Do Smaller Language Models Think More Like Humans? A Scaling Analysis of Psycholinguistic Surprisal Effects
- Do Speech BCIs Need Larger Models? Rethinking Neural Decoding beyond Scaling
- Double Preconditioning: Optimization for Test-Time Performance, not Validation Loss
- Do‑Undo Bench: Reversibility for Action Understanding in Image Generation
- Do We Really Need to Cache Keys and Values? Structure-Aware Surrogate Caching for Long-Context LLM Inference
- Do You CARE to Generalize? Extracting Robust Concept Directions from LLMs
- DP-$\lambda$CGD: Efficient Noise Correlation for Differentially Private Model Training
- DPBench: Structural Determinants of Multi-Agent LLM Coordination Under Simultaneous Resource Contention
- DP-BiSF: Differentially Private Fine-Tuning via Bilateral Subspace Projection
- DPMPC: Diffusion Policies as Expressive Priors for Model Predictive Control
- DP-RFT: Learning to Generate Synthetic Text via Differentially Private Reinforcement Fine-tuning
- DPrivBench: Benchmarking LLMs' Reasoning for Differential Privacy
- DRA-CLIP: Decoupled Residual Adaptation for CLIP Out-of-Distribution Generalization
- DRACO: A Cross-Domain Benchmark for Deep Research Accuracy, Completeness, and Objectivity
- dRAE: Representation Autoencoder with Hyper-Spherical Codes
- DrafterBench: Benchmarking Large Language Models for Auditable Tool-Calling in Drafting Workflow
- Draft, Verify, \& Improve: Toward Training-Aware Speculative Decoding
- Draw2Think: Harnessing Geometry Reasoning through Constraint Engine Interaction
- Draw What LLMs Intend: Prompt Alignment Is Not Enough for Text-to-Image Generation
- DreamAvoid: Critical-Phase Test-Time Dreaming to Avoid Failures in VLA Policies
- Dream-Cubed: Controllable Generative Modeling in Minecraft by Training on Billions of Cubes
- DREAM: Diffusion-guided Distribution Rebalancing with Latent Deformation for Realistic Long-Tailed Semi-Supervised Learning
- DREAM: Error-Guided Counterexample Learning for Robust Early Pancreatic Cancer Detection
- Dreaming Smoothly and Sample Efficiently with Gradient-Penalized Latent Dynamics
- DReSS: Data-driven Regularized Structured Streamlining for Large Language Models
- DressFlow: Dynamic Garment Reconstruction by Structured Latent Geometry Propagation
- D-REX: A Benchmark for Detecting Deceptive Reasoning in Large Language Models
- DrGait: Biomechanically Grounded Visual Reasoning for Interpretable Clinical Gait Analysis
- Drift-admissibility via Optimal Transport
- Drift-Constrained Optimization: Only Direction Matters in Fine-Tuning Instruct Models
- Drift-Correcting Neural Emulation via Conditional Decoupling
- Drift Flow Matching
- DRILL: Training World Models to Improve Policies, Not Predict Pixels
- DriveFuture: Future-Aware Latent World Models for Autonomous Driving
- DriveHierarchy: A Benchmark for Diagnosing VLM Driving Capabilities from Open-Loop Understanding to Closed-Loop Execution
- DriveJudge
- DriveMind: Mind-Evolving Belief Tracking for Closed-Loop Autonomous Driving
- DriveReward: A Comprehensive Dataset and Generative Vision-Language Reward Model for Autonomous Driving
- Driver Vigilance Prediction Based on Deep Feature Fusion of Multi-Modal Physiological Signals with Dempster-Shafer Evidence Theory and Contrastive Distillation
- DriveWAM: Video Generative Priors Enable Scalable World-Action Modeling for Autonomous Driving
- Driving Intents Amplifies Planning Oriented Reinforcement Learning
- Driving-LRM: 4D Large Reconstruction Model for Driving Scenes
- Driving Video Retrieval for Complex Queries with Structured Grounding
- Dropout-GRPO: Variational Stochasticity for Continuous Latent Reasoning
- DR-SNE: Density-Regularized Stochastic Neighbor Embedding
- DrugBench: Evaluating AI Control Protocols for Medication Harm Mitigation
- DrugSAGE: Self-evolving Agent Experience for Efficient State-of-the-Art Drug Discovery
- DSAQuant: Denoising-Stage-Aligned Quantization-Aware Training for Video Generation
- DSCA-KT: Dual-State Context-Aware Model for Dynamic Knowledge Tracing
- DS-Cone: Dynamic Sequential Lorentz Cone for Knowledge Graph Path Reasoning
- DSD-GS: Dynamic-Static Decomposition of Gaussian Splatting for Efficient and High-Fidelity Dynamic Scene Reconstruction
- DSDyn-VLA: A Dual-Stream Dynamic Manipulation Framework with Motion Perception, Future Awareness, and Realtime Correction
- DS-Flow: Dual-Stream Dynamical System Correction for Few-Step Diffusion Sampling
- DSSP: Diffusion State Space Policy with Hierarchical Full-History Conditioning
- DT-SAFD: Learning Differentiable and Task-Driven Spectral Representations in RAW Object Detection
- DTS: Distractor-Guided Temperature Scaling for LLM Calibration
- Dual Advantage Fields
- DualChem: Can LLMs Provide Dangerous Uplift in Dual Use Chemistry?
- Dual Feature-Relational Alignment for Transferable Targeted Attacks on MLLMs
- Dual-Grained Agent Memory and Shapley Context Attribution for Multimodal Agentic Learner
- Dual-Hypothesis Vision-Language Reasoning for AI-Generated Image Detection
- Duality Models: An Embarrassingly Simple One-step Generation Paradigm
- Dual-Level Contrastive Alignment for Robust Representations in Long-Tailed Medical VQA
- DualNoise-ClinIE: A Calibrated Benchmark for Joint OCR and Annotation Noise in Semi-Structured Clinical Information Extraction
- Dual-Primal Graph VAEs for Noisy Label Aggregation
- DualQuant: Dual-scale Calibration-free Weight-and-Activation Quantization for LLMs
- Dual-Space Preconditioning for Variational Inequalities and Root-Finding Problems
- DualSteer: Dual-Space Steering for Robust Jailbreak Mitigation of Large Vision Language Models
- DualStreamFlow: Structured Conditional Flow Matching for Frequency-Aware End-to-End Autonomous Planning
- DualWorldBench: Can Agents Plan Deliveries across Symbolic and Grounded Worlds?
- DUIL: Deep Unsupervised Inverse Learning for in situ Macromolecular Morphology Identification
- DUMap: Dirichlet Uncertainty-Guided Online Vectorized HD Mapping under Environmental and Geographic Domain Shifts
- D-Value Bench: A News-Driven Dynamic and Comparative Value Evaluation Benchmark
- DVAO: Dynamic Variance-adaptive Advantage Optimization for Multi-reward Reinforcement Learning
- DVD: Deterministic Video Depth Estimation with Generative Priors
- DVD: Discrete Voxel Diffusion for 3D Generation and Editing
- DyCoPart: Dynamic Body-Part Coordination for Text-to-Motion Generation
- DynamAuction: a reinforcement learning environment for repeated auction with dynamic value
- Dynamical Anatomy of Emergent Abilities in Large Language Models
- Dynamical System Discovery from Specifications and Partial Observations
- Dynamic Budget Allocation for Federated Class-Incremental Learning
- Dynamic Cross-Modal Prompt Generation for Multimodal Continual Instruction Tuning
- Dynamic Delayed Tree Expansion For Improved Multi-Path Speculative Decoding
- Dynamic Depth Routing of Large Language Models with Mixture of Segment Operators
- Dynamic FFNs Improve Representation Learning in Transformer Pretraining
- DynamicManip: Enabling Dynamic Manipulation from a Single Static Demonstration
- Dynamic Mask Attention: End-to-End Trainable Sparse Attention
- Dynamic Minimax Regret Optimization for Robust LLM Post-Training
- Dynamic Neural Optimal Transport for Subset Alignment
- Dynamic Optimisation of Discount and Trace Decay for Value-Based Reinforcement Learning
- Dynamic Optimistic Constrained OCO with Memory via Delay Equivalence
- Dynamic Quadtree Tokenization for Autoregressive PDE Forecasting
- DynamicRad: Content-Adaptive Sparse Attention for Long Video Diffusion
- Dynamic Range Collapse: Feature-Space Density for Adaptive Conformal Prediction
- Dynamic Regret in Online Convex Optimization with Indicator Switching Costs
- Dynamic Regulation for Continual Learning via Layer-Wise Feedback Control
- Dynamic Representation Modeling for Federated Medical Image Domain Generalization
- Dynamics at the Frontiers of Optimization, Sampling, and Games
- Dynamic Shapley Computation
- Dynamics of Stochastic Momentum with Sparse Updates in High Dimensions
- Dynamic SVD via Token-Adaptive Basis Selection
- Dynamic Texturing for 2D Multibody Systems
- Dyna-Style Safety Augmented Reinforcement Learning: Staying Safe in the Face of Uncertainty
- DynCodon: Multi-objective Dynamic Learning Strategies for Robust Codon Optimization
- DynEdit: Dynamic Entropy-Guided Sequential Editing for Large Language Models
- EACO-RAG: Edge-Assisted Collaborative Orchestration for Scalable Cloud–Edge Retrieval-Augmented LLM Serving
- EAD-LoRA: Expert-Aware Differential Initialization for MoE-LoRA Frameworks
- EA-DST: Energy-Aware Dynamic Sparse Training for Sustainable Deep Learning — Beyond FLOPs
- Eagle-Embodied: Seeing in 2D, Acting in 3D
- Early-Bird Decoding: Accelerating Diffusion LLMs with Learnable Block Sizes and Parallel Sampling
- Early Semantic Commitment in Diffusion Sampling
- Early Stopping Without Validation via Unlabeled Surrogate Risk
- EASE Configuration Facilitates A Reproducible Science of LLM Social Simulations
- EasyLens: A Training-Free Plug-and-Play Subtle-Lesion Representation Amplifier for Medical Vision-Language Models
- EatBench-2.7K: A Benchmark for Fine-Grained Eating Action Grounding in Videos
- EAT: Eviction-Aware Training for Long-Context LLM Inference on Edge Devices
- EBM-CoT: Energy-Based Calibration for Implicit Chain-of-Thought
- EBPO: Empirical Bayes Shrinkage for Stabilizing Group-Relative Policy Optimization
- ECG Dataset with Multi-Expert Annotations and Delineations
- ECGT : Evolutionary Causal Graph Transformer for Amortized Causal Discovery
- ECHO-2: A Large-Scale Distributed Rollout Framework for Cost-Efficient Reinforcement Learning
- ECoGen: Learning to Complete Chemical Reactions with Electron-Constrained Generation
- ECOKV: Geometry-Aware KV Cache Eviction via Complementary Diversity Metrics
- EconML: Economics for Machine Learning
- Economy of Minds: Emerging Multi-Agent Intelligence with Economic Interactions
- ECSO: Evidence-Certified Safe Overrides for Multi-Hop Question Answering
- EDEN: Emergent Dynamics in Evolutionary Neural-networks for Robust Continuous Control
- Edge of Stability Selectively Shapes Learning Across the Data Distribution
- EDITOR-Bench: Edit-Level Evaluation of Evidence Use and Risk Control in Clinical Record Updating
- Editor's Choice: Evaluating Abstract Intent in Image Editing through Atomic Entity Analysis
- EduBench: Benchmarking Pedagogical Quality in Large Language Models for Education
- EduVidBench: A Benchmark for Evaluating Timing-Aware Educational Animation Generation
- EEG-X: Toward Device-Agnostic and Noise-Robust Foundation Models for EEG
- Eevee: Towards Test-time Prompt Learning in the Real World for Self-Improving Agents
- Effective Context in Transformers: An Analysis of Fragmentation and Tokenization
- Effective Residual-Stream Depth of Language Models
- Effective Test-Time Scaling of Discrete Diffusion through Iterative Refinement
- EffGS: Efficient and High-Fidelity Gaussian Splatting
- Efficiency of Diffusion Models for Infinite-dimensional Data Generation: Dimension Independent Approximation and Estimation Errors
- Efficient Active Learning for Continuous-Time Experiments
- Efficient Adaptive Data Analysis over Dense Data Distributions
- Efficient Agentic GPU Kernel Optimization with a Compact Domain-Specific Language and Speed-of-Light Guidance
- Efficient Agentic Reasoning Through Self-Regulated, Simulative Planning
- Efficient Algorithms for Contextual Apple Tasting with Log-Loss
- Efficient Algorithms for Distributed Saddle Problems
- Efficient Analytic Uncertainty Quantification for Multimodal Regression
- Efficient and Interpretable Transformer for Counterfactual Fairness
- Efficient and Robust Physical 3DGS-MPM Simulation via Interior Filling and Text-Physics Optimization
- Efficient Attention Adaptation with Binary Addressing and Sparse Delivery
- Efficient Benchmarking Is Just Feature Selection and Multiple Regression
- Efficient Best-Of-$N$ with Radial Consensus Score
- Efficient Diffusion Distillation via Embedding Loss
- Efficient Diffusion Policy Fine Tuning with Latent Noise Representation Bridging
- Efficient Diffusion Transformer Inference via Token-Adaptive Mixture Caching and Grouped Forecasting
- Efficient Dynamic Algorithms for Graph Neural Networks with Non-Linearity
- Efficient Error-Driven Model Revision via Near-Hit and Near-Miss Augmentation
- Efficient evaluation and error pattern discovery for blackbox AI systems
- Efficient Exploration in Model-Based RL with Convex Neural Network Posteriors
- Efficient Fine-Tuning for Structured Sparsity Under Group Repartitioning
- Efficient Generative Transformer Operators for Million-Point PDEs
- Efficient Gradient-Aware Asynchronous Reinforcement Learning for LLM Post-Training
- Efficient Long-Horizon Learning for Learned Optimization
- Efficiently Challenging the Leader in Matroid Bandit Leveraging Unimodality
- Efficient Multinomial Logistic Bandit via Frequent Directions
- Efficient Neural Field Learning via Adaptive Coverage and Focused Sampling
- Efficient One-Step Diffusion Restoration Model with Compact Token Compression and Linear Attention
- Efficient Online Proportional Sampling with Applications to Smoothed Online Learning
- Efficient Pre-Training of LLMs through Truncated SVD Layers
- Efficient Private Adaptation of Multimodal Large Language Models via Token and Update Sparsification
- Efficient Regression Models for Scan Statistics
- Efficient RLVR Training via Weighted Mutual Information Data Selection
- Efficient Scaling of LLM Training with Flexible Context Parallelism
- Efficient Spatio-Temporal Grounding with Multimodal Large Models via Second-Level Tracking and RL Verification
- Efficient Spike‑Driven Large Vision‑Language Models with Hierarchical Binary Spiking Neurons
- Efficient Test-Time Adaptation For Robot Policies
- Efficient Tool-Calling Agentic Workflows with Continuous Chain-of-Thought
- Efficient Training of Deep Spiking Neural Networks with Input-Driven Derivative-Free Updates
- Efficient Transferable Optimal Transport via Min-Sliced Transport Plans
- EGCA: A Spectral Perspective on Forward Process Design in Diffusion Models
- EgoBodyRAG: Coarse-to-Fine Retrieval-Augmented Human Mesh Recovery in Egocentric Video
- EgoGen: Egocentric Whole-body Human Motion Generation via VLA Models
- EgoXR-GUI: Benchmarking GUI Grounding in Physical–Digital Extended Reality
- Eigen Neural Networks: Training Deep Layers in Learned Orthogonal Coordinates
- EKG-Scientist: Evolutionary Knowledge Graph Search for Automated Scientific Discovery
- Elastic Attention Cores for Scalable Vision Transformers
- Elastic Certified Removal: Continual Machine Unlearning with Formal Privacy Guarantees
- Elastic Representations via Hyperbolic Geometry
- ELDOR: A Dataset and Benchmark for Illegal Gold Mining in the Amazon Rainforest
- ElectroDiT: Unifying Prior Knowledge and Structural Constraints for Efficient Electrolyte Design
- Elevation-FS4K: A Factorial Benchmark for Diagnosing Multi-View Spatial Reasoning in Vision-Language Models
- ELF: Embedded Language Flows (Extended)
- Elicited Adaptation: Auditable Localized Fairness via Pairwise Queries
- Eliciting Interaction across Observation, Retrieval, and Reasoning for Medical Multimodal LLMs: A Closer Look
- ELLIS Workshop on the Foundations of LLM Post-Training in Changing Environments
- ELPAC: Endpoint-Anchored Latent Progression with Stage-Varying Multimodal Coordination
- Elucidating the Design Space of Sequential EHR Generators
- EMABench: An Extensible Multimodal Agent Benchmark for Trace-Aware Diagnosis
- EMAG: Differentiable 4D Gaussian Mixture Splatting for EEG Spatial Super-Resolution: A Closer Look
- Embedding Dimension Lower Bounds for Universality of Deep Sets and Janossy Pooling
- Embedding Foundation Model Predictions in Discrete-Choice Models with Structural Guarantees
- Embedding Perturbation may Better Reflect Intermediate-Step Uncertainty in LLM Reasoning
- Embedding Physical Reasoning into Diffusion-Based Shadow Generation Under the Sun and Sky
- Embodied Action Hijacking: How Vulnerable Are Your Embodied VLM Agents?
- Embodied Forcing: Physics-Informed Video Generation for Robotic Manipulation
- Embodied Neurocomputation: A Framework for Interfacing Biological Neural Cultures with Scaled Task-Driven Validation
- Emergence of Physical Intelligence via Controllable Information Production
- Emergent Biological Capabilities in a Foundation Model for Molecular Interactions
- Emergent Misaligned Communication in Long-Horizon Multi-Agent LLM Commerce
- Emergent Misinformation Genesis in Multi-Agent LLM Clinical Pipelines
- Emergent Region-Level Facial Correspondence in Frozen Vision Foundation Models
- Emergent Specialization in Populations of Self-Supervised Collaborative Experts
- EM-JEPA: An Episodic Memory-Augmented Joint-Embedding Predictive Architecture for Large Language Models
- EML Trees are Universal Approximators
- Emotion OS: A State-Space Control Framework for Stabilizing Emotional Interaction in Conversational AI
- Emotion-Trained Vision Models Do Not Necessarily Learn EEG-Aligned Facial Dynamics
- EMPA: Evaluating Persona-Aligned Empathy as a Process
- Enabling Functional Encryption-Based Inference on Residual Architectures
- Enabling Preference-driven Unlearning in Few-step Distilled Text-to-Image Diffusion Models
- ENACT: Single-Image Human-Scene Interaction Motion from Language via Foundation-Model Orchestration
- EncoderMoE: Classifying User-Defined Gestures Using MoE Encoders and Meta-Learning
- Encode, Think, Decode: Scaling test-time reasoning with recursive latent thoughts
- Encoding RNA Topology into Synthetic Alignments for 3D Structure Prediction
- EndoSCOP-V: A Multi-Turn Video Understanding Evaluation Framework for Multimodal Models in Endoscopy Reporting and Clinical Reasoning
- End-to-End Context Compression at Scale
- End-to-End Neural Modeling of EM Response and Design Performance for Free-Form RFIC Passives
- End-to-end PDDL Planning with Hardcoded and Dynamic Agents
- End-to-End Training for Unified Tokenization and Latent Denoising
- End-to-End Verification of Neuro-symbolic Automata via Contrastive Logit-Gaps
- EnergyBench: Benchmarking of Time Series Foundation Models for energy load forecasting and anomaly detection
- Energy-Efficient Coding in Spiking Neural Networks via Coefficient-of-Variation Regularisation
- Energy-Gated Trajectory Optimization (EGTO)
- Energy-Guided Learning for Parabolic PDEs: Optimal Estimation and Generalization Bounds
- Enforcing Constraints in Generative Sampling via Adaptive Correction Scheduling
- Engineering Embodied Prompts to Expose Sparsity in LLMs
- Enhanced convergence guarantees of score-based generative models in $\mathcal{W}_2$-distance beyond log-concavity
- Enhanced Generation in Diffusion Models via Joint High-to-Low Denoising Schedule and Model Sparsity with Theoretical Guarantees
- Enhancing Bayesian Neural Networks with Functional Priors and Partial Stochasticity through Deep Weight Factorization
- Enhancing GNNs Performance on Combinatorial Optimization by Recurrent Feature Update
- Enhancing Knowledge Injection with Surrounding Backgrounds in Continual Training LLMs
- Enhancing Reinforcement Learning for Autonomous Driving with Outcome-Related Reward Shaping
- Enhancing Software Engineering Through Closed-Loop Memory Optimization
- Enhancing Time Series Forecasting via Spectral Entropy-guided Training
- Ensemble Modeling for Time Series Forecasting: an Adaptive Robust Optimization Approach
- Ensemble Selective Classification
- Ensembles of Embedding Models for Motor Imagery Classification
- Ensembling Language Models with Sequential Monte Carlo
- EnterpriseLab: A Full-Stack Platform for developing and deploying agents in Enterprises
- EntiRE: Invariant Learning for Robust Concept Erasure in Text-to-Image Generative Models
- EntiWeave: Weaving Entity Dependencies for Deep Research Reasoning
- Entropic Frontdoor Adjustment: Falsification and Partial Identification under Weak Confounding
- Entropy Aware Reward Guidance for Diffusion Language Model Alignment
- Entropy-Aware Token Ordering in Discrete Diffusion Language Models
- EntropyCache: Decoded Token Entropy Guided KV Caching for Diffusion Language Models
- Entropy Distribution as a Fingerprint for Hallucinations in Generative Models
- Entropy Guided Dynamic Patch Segmentation for Time Series Transformers
- EntropyMaG-1: A System for Constructing Harder Mathematical Problems via Multi-Generation Evolution with Auditable Quality Control
- Entropy Pacing Policy Optimization for Multi-Task Agentic Reinforcement Learning
- Entropy Polarity in Reinforcement Fine-Tuning: Direction, Asymmetry, and Control
- EnvFaultBench: Benchmarking LLM Agents on Software Environment-Fault Troubleshooting
- Environment Learning: A Vector-Viability Framework for Self-Modifying Adaptive Agents
- EnvRL: Learn from Environment Dynamics in Agentic Reinforcement Learning
- EnzymeLM: Chemistry-Native Language Modeling for Balanced Enzymatic Reaction Prediction
- EO-WM: A Physically Informed World Model for Probabilistic Earth Observation Forecasting
- EPAS: A Multi-Dimensional Benchmark for Evidence Use in Parliamentary Discourse
- EpiCARD: Graph-Prior-Free Epidemic Forecasting with Card-Guided Case Semantics
- EPIC: Efficient Predicate-Guided Inference-Time Control for Compositional Text-to-Image Generation
- EpiPivot: Learning to Control the Simplex Method under Epistemic Uncertainty
- EpisodeBench: A Full-Cycle Benchmarking Pipeline for Long-form Interactive Story Generation with Controllable RL
- Episode Significance under Minimal Windows Frequency
- Epistemic Role Framing Elicits Deep Collaboration in LLM Multi-Agent Systems
- Epistemic Uncertainty for Test-Time Discovery
- EPSI: A Sketch-and-Precondition Framework for Randomized Low-Rank Approximation via Error-Powered Sketched Inverse Iteration
- Equidistribution of Solution Gradients in Physics-Informed Neural Networks via Coordinate Transformations
- EquiFiLM: Charge-Conditioned Equivariant Force Fields via Feature-wise Linear Modulation
- Equilibrium Residuals Expose Three Regimes of Matrix-Game Strategic Reasoning in Language Models
- EquiMark Robust Watermarking via Equivariant Convolutions and Print Augmentations
- Equivariant Force Field Calibration for Flow-based Protein Design
- Equivariant Weak-Form Bayesian Discovery of Partial Differential Equations
- ER-Reason: A Benchmark Dataset for LLM Clinical Reasoning in the Emergency Room
- Error-Guided Self-Distillation: Turning Student Failures into Distillation Signals
- Error Taxonomy-Guided Prompt Optimization
- Escaping Confidence Trap: Evolutionary Decoding for Mathematical Reasoning in Diffusion LLMs
- Escaping Iterative Parameter-Space Noise: Differentially Private Learning with a Hypernetwork
- Escaping the Mode Lottery: Multi-Response Training Improves Language Model Generalization
- Escaping the MSE Trap: Stable and Spectrally Faithful Prediction of 3D Transonic Turbulence
- ESP-YOLO: Stabilizing Incremental Object Detection via Eigenvalue Scaled Projection Regularization and Confidence-Aware Distillation
- Estimand Mismatch in RLVR: A Structural Fix and Symmetric Held-Out Analysis
- Estimating and Orthogonalizing Unknown Pre-training Gradients for Continual Fine-tuning of Large Language Models
- Estimating Structure-Preserving Velocities for Image Editing via Low-Rank and Sparse Decomposition
- Estimating the expected output of wide random MLPs more efficiently than sampling
- Estimation of the Label-Noise Transition Matrix with Performance Guarantees via Selective Classification
- Estimation of the sub-Gaussian parameter
- Euclidean Embedding of Data Using Local Distances
- Euclid-Omni: A Unified Neuro-Symbolic Framework for Plane Geometry
- Evaluating AGENTS.md: Are Repository-Level Context Files Helpful for Coding Agents?
- Evaluating AI-Generated Video Detectors under Real-World Distribution Shifts
- Evaluating Deployable Inference-Time Error Prediction in Vision MoEs
- Evaluating Multimodal Narrative Understanding of Popular Hollywood Films
- Evaluating Physical Reasoning in LLM Agents Requires Construction Benchmarks
- Evaluating Probabilistic Label Consensus Techniques for Foundation Model-Driven Pseudo Labels
- Evaluating Spatial World Modeling in Video Generators via 3D Camera Trajectory Generation
- Evaluating Synthetic ECG Pretraining: When Can Patient-Free Simulators Substitute for Real ECG Data?
- Evaluating the Relevance of Uncertainty Estimators for LLM Hallucination: A Closer Look
- Evaluation Awareness in Language Models Has Limited Effect on Behaviour
- Evaluation-Conditioned Training: Teaching Models to Generalize to Stronger Oversight Regimes
- Evaluation--Deployment Sensitivity in LLMs: When Prompt Disguise Changes Behavior
- Evaluative Fingerprints: Auditing Whether LLM Judges Are Interchangeable
- E-Values: From Statistics to ML
- EVaR-Optimal Arm Identification in Bandits Models
- EvenFlow: Benchmarking Shared-Space Navigation from Real Human Motion
- Even Sharper Bounds for Transductive Learning and Its Applications
- Event based Multi-Velocity-Scale Imaging
- EventCert: Certified Robustness to Structured Deletions via Delete-Aligned Smoothing
- Event-Guided Fusion Network for Ultra-Low-Light Video Enhancement
- Events as Triggers for Behavioral Diversity in Multi-Agent Reinforcement Learning
- EventSym: An Event-Based Dataset Exploring Scaled Resolutions of Standard Signs and Symbols
- Every9D: Large-Scale Canonicalization of Everyday Objects
- Every Bit, Everywhere, All At Once: A Binomial Multibit LLM Watermark
- EvidCoT-Surg
- Evidence and Instability Auditing in Medical Image Classification via Geometry-Consistent Energy Inference
- Evidence Over Plans: Online Trajectory Verification for Skill Distillation
- Evidential Game-Theoretic Fusion: Resolving Multi-View Conflicts in 3D Perception
- Evidential Latent World Models for Safe Model-based Reinforcement Learning
- Evidential Semantic Uncertainty for Autoregressive LLM Generation
- EvilGenie: a Reward Hacking Benchmark
- EvoArena: Tracking Memory Evolution for Robust LLM Agents in Dynamic Environments
- EVOCHAMBER: Test-Time Co-evolution of Multi-Agent System at Individual, Team, and Population Scales
- EvoClinician: A Self-Evolving Agent for Multi-Turn Medical Diagnosis via Test-Time Evolutionary Learning
- EvoCodeBench: Evaluating Coding Agents in Multi-Turn Iterative Interactions
- EvoCUA: Evolving Computer Use Agents via Learning from Scalable Synthetic Experience
- Evo-Depth: A Lightweight Depth-Enhanced Vision-Language-Action Model
- EvoDrive: Self-Evolving Autonomous Driving via Latent World Model-Guided Post-hoc Reflection
- EvoGuard: Neurogenetic Search of Guardrail Architectures for Regulated AI
- EvolveMem: A Self-Evolving Memory Architecture for LLM Agents
- Evolving Agent Teams
- Evolving Layer-Specific Scalar Functions for Hardware-Aware Transformer Adaptation
- Evolving Robustness–Exploration Trade-offs in Online Bayesian Reinforcement Learning
- EvoMem: Memory-Augmented Evolution for Code Optimization
- EvoReasoner: Evolutionary Expert for Recursive Reasoning via Latent Space Learning
- EvoRubric: Self-Evolving Rubric-Driven RL for Open-Ended Generation
- EvoScene-VLA: Evolving Scene Beliefs Inside the Action Decoder for Chunked Robot Control
- EVQ-Cosh: Variational Frequency Allocation for Rotary Position Embedding
- EVT-Reranker: An Extreme-Vaule-Theory-based Weight Reranking Model for Long-Tail Retrieval-Augmented Generation
- Exact Channel Decoupling via Joint Diagonalization and Uniform Splicing for Diffusion Transformer Quantization
- Exact Combinatorial Optimization for Partial Permutation Synchronization
- Exact Distributed Structure-Learning for Bayesian Networks
- Exact Executable Ancestry Beyond Strong Non-DAG State: A Matched-Control Study of Contract-Rich Repository Repair
- Exact Federated Forecasting for Heterogeneous Streams: A Closed-Form Incremental Approach
- Exact Flow Linear Attention: Exact Solution from Continuous-Time Dynamics
- Exact-Form Regret and Conservative Correlated Equilibria
- Exact Identifiability in Causal Representation Learning via Global Automorphism Construction
- Exact Labels, Wrong Target: Continuation-Consistent Replay for Closed-Loop Agent Control
- Exactness Matters for Physical Rule Enforcement
- Exact Unlearning via Quantized Sufficient Statistics
- ExEBench: A Benchmark and Roadmap for Extreme Event Prediction and Mapping
- ExecTune: Effective Steering of Black-Box LLMs with Guide Models
- Exemplar Partitioning for Mechanistic Interpretability
- ExLLM: Efficient Experience-Enhanced LLMs for Large-Scale Discrete Optimization
- ExMAG: Learning of Maximally Ancestral Graphs
- ExoST: Multi-System Exogenous Variable Modeling for Spatio-Temporal Forecasting
- Expander Sparse Autoencoders: Parameter-Efficient Dictionaries for Mechanistic Interpretability
- Expanding LLM Agent Boundaries with Strategy-Guided Exploration
- Expanding the Role of Diffusion Models for Robust Classifier Training
- Experience Makes Skillful: Enabling Generalizable Medical Agent Reasoning via Self-Evolving Skill Memory
- ExpertNavigator: Functionally Coherent Expert Grouping and Pairwise-Ranked Routing for High-Fidelity Dense-to-MoE Conversion
- Expert-Validated STEM QA
- Explainability matters: The effect of liability rules on the healthcare sector
- Explainable Network Dynamics via Lagrangian Formulation of Training Regimes
- Explainable Novel Category Discovery in Semantic Concept Space
- ExplainerPFN: Towards tabular foundation models for model-free zero-shot feature importance estimations
- Explaining and Preventing Alignment Collapse in Iterative RLHF
- Explaining Attention with Program Synthesis
- Explaining Cross-Modal Model Behavior with Gradient-Estimation-Based Feature Interaction
- Explanation-Guided Distillation for Human-Aligned LLM-Judges
- Explanation of Dynamic Physical Field Predictions using WassersteinGrad: Application to Autoregressive Weather Forecasting
- Explicit Admissibility Interfaces: Positive and Null Regimes under Mixed Specifications
- Explicit Critic Guidance for Aligning Diffusion Models
- Exploiting Depth Maps for Blind Image Quality Assessment via Spatial-Structural Cues
- Exploiting Fine-Tuning Structures to Improve Adversarial Transferability on Downstream SAM
- Exploiting Mode Connectivity for Out-of-Distribution Detection via Low-Dimensional Subspace Modeling
- Exploiting Neural Network Memory in a Structured Model Embedding Space
- Exploiting the Structural Affinity of Cage-Based Deformation for Computational Shape Design
- Exploratory Experience Shapes the Geometry of Predictive Representations
- EXPLORING EXPERT FAILURES IMPROVES LLM AGENT TUNING
- Exploring MLLM-Diffusion Information Transfer with MetaCanvas
- Exploring Multi-Order Self-Similarity for Motion Understanding
- Exploring Multiple High-Scoring Subspaces in Generative Flow Networks
- Exploring Parallel Guidance Reconstruction for Multi-class Unsupervised Anomaly Detection
- Exploring Self-Learning for Next-Generation Multimodal Foundation Models
- Exploring the AI Obedience: Why is Generating a Pure Color Image Harder than CyberPunk?
- Exposing and Mitigating Temporal Attack in Deepfake Video Detection
- Exposure-State Conditioned Multi-Scale Photo Exposure Correction
- Expressiveness Limits of Autoregressive Semantic ID Generation in Generative Recommendation
- Expressive, Scalable, and Efficient Architectures for Oscillatory State-Space Models
- ExRD: Preventing Zero-Shot Degradation in Continual Learning via Exemplar Replay with Distillation
- Extending 3D Reconstruction Models to Any Camera
- Extending Myerson's Optimal Auctions to Correlated Bidders via Neural Network Interpolation
- Extracting and Composing Function Vectors for Analogical Reasoning over Latent Relations
- Extracting Recurring Vulnerabilities from Black-Box LLM-Generated Software
- Extracting Search Trees from LLM Reasoning Traces Reveals Myopic Planning
- Extracting Training Data from Diffusion Language Models via Infilling
- Extragradient and Past Extragradient Methods for $(L_0,L_1)$-Lipschitz Variational Inequalities
- Extrapolation Guarantees for Perturbation Modeling Under the Additive Latent Shift Assumption
- Extreme-Tail Conformal Calibration: An Order-Statistic Phase Transition and a Bias--Variance Perspective
- FABSVer: Faster Training and Better Self-Verification for LLM Mathematical Reasoning
- FACET
- FACETS: Cross-Granularity Vision--Language Modeling for 3D Anomaly Detection
- Factored Diffusion Policies: Compositionally Generalized Robot Control with a Single Score Network
- Factorized Gradients for Scalable Highly-expressive Parametric Diffeomorphisms
- Factorized Story-Level Control for Multimodal Storybook Generation from Child-Authored Cues
- FactoryBench: Evaluating Industrial Machine Understanding
- FADE: Frame-Aware Diffusion-Transformer-based Multi-Concept Erasure for Video Unlearning
- Failing Forward: Adaptive Failure-Informed Learning for Vision-Language-Action Models
- Failure Modes of Complaint Classification in Healthcare: A Large-Scale Evaluation under Operational Constraints
- Fair Clustering Ensembles with Assignment-Induced Allocation Weights: A Statistic-Level Risk Theory
- Fair Division Meets Scheduling: Approximately Envy-Free Interval Scheduling
- Fair Division Under Inaccurate Preferences
- Fair Ensemble Learning for Imbalanced Data Streams
- FairGAD: Mitigating Influence Bias for Dynamic Graph Anomaly Detection
- Fair Indivisible Payoffs through Shapley Value
- FairMT: Fairness for Heterogeneous Multi-Task Learning
- Fair on the Surface? Benchmarking Hidden–Output Fairness Gaps in LLM Recommenders
- FairSplit: Decomposing the Embedding Space for Fair Classification
- FairTune Market: A Fair and Trustworthy Marketplace for Fine-Tuned LLMs via Posted-Price & Proper-Scoring Mechanisms
- Faithful Summarization via Atomic Fact Decomposition and Circular Consistency Reward
- Faithful to the Persona, Unfaithful to the Decision: A Mechanism for Chain-of-Thought Unfaithfulness
- FA-LAM: Focus-Aware Large Avatar Model for One-Shot 4D Animatable Gaussian Head
- FALCON: Aligning Vision and Language Models with Few-Shot Supervision
- FALD: Fine-grained Adaptive Latent Detection for LLM Hallucination
- FALS-C4: Per-Tensor Distributional Refinement for Layerwise Sparsity Allocation in LLM Pruning
- False Convergence: Representation-Level Predictors of Incorrect Consensus in Multi-Agent LLM Systems
- Family-FDR Control for Open-Vocabulary Detection under Known Prompt Sets
- FAMO: Frequency-Aware Multi-Objective Learning for Unified Image Restoration
- Fantastic Adaptive Taxonomies and How to Use Them
- FARBench: How Far Are LLM Agents from Autonomous Research?
- FaroeWRF-500: A Diagnostic Benchmark for Weather Model Failure Analysis over Complex Terrain
- FashionChameleon: Towards Real-Time and Interactive Human-Garment Video Customization
- FAST3DIS: Feed-forward Anchored Scene Transformer for 3D Instance Segmentation
- Fast Adversarial Attacks with Gradient Prediction
- Fast and Accurate Probing of In-Training LLMs' Downstream Performances
- Fast and Consistent Structure Learning in Graphical Models via Approximate Cross-Validation
- Fast and slow gradient descent dynamics of logistic regression through weak alignment
- Fast Approximate $\ell_p$ Chamfer Distance via Lopsided Embeddings and Structured JL
- Fast-dDrive: Efficient Block-Diffusion VLM for Autonomous Driving
- FastDSAC: Unlocking the Potential of Maximum Entropy RL in High-Dimensional Humanoid Control
- Fast & efficient online alignment with multidimensional reward model
- Faster Dynamic Graph Clustering with Hierarchical Graph Contraction
- Faster-GCG: Efficient Discrete Optimization Jailbreak Attacks against Aligned Large Language Models
- Faster Optimization and Machine Learning via Data Splitting
- Faster Policy Learning from Multiple Experts: Stitch, Imitate, Optimize
- FASTER: Rethinking Real-Time Flow VLAs
- FastSAM3：Accelerating SAM3 with Lazy Detection and Tracking
- FastSAMv2: Fully Convolutional Networks for Fast Segment Anything in Images and Videos
- Fast Sandwich Products in Clifford Algebra
- Fast, Sparse, and Accurate Kernels for Node-Attributed Graphs
- Fast Text-to-Audio Generation with One-Step Sampling via Energy-Scoring and Auxiliary Contextual Representation Distillation
- FateCode: Conditional Flow Matching over Learned Fate Distributions for Program-Level Cell Fate Analysis
- FaultArena: A Benchmark for State-Dependent Long-Horizon Industrial Fault Diagnosis
- FaVChat: Hierarchical Prompt-Query Guided Facial Video Understanding with Data-Efficient GRPO
- FC-DGCN: Deep Graph Convolutional Network for Face Clustering and Recognition
- FC-SSM: Frequency-Conditioned Selective State Spaces for Non-Stationary Spatiotemporal Dynamics
- FDSRL: Fidelity-Driven Synergistic Representation Learning for Robust Multi-Domain Fusion
- FeatMark: Feature-level Watermark Protection against Mimicry Attacks with Diffusion Models
- Feature Collapse Is the Bottleneck: A Benchmark Study of Selective Prediction Under Medical Distribution Shift
- Feature Starvation as Geometric Instability in Sparse Autoencoders
- FedCF: Fair Federated Conformal Prediction
- FedDCT: One-Shot Federated Learning via Frequency-Domain Aggregation
- FedeKD: Energy-Based Gating for Robust Federated Knowledge Distillation under Heterogeneous Settings
- Federated learning approach for the Erdős-Rényi random graph
- Federated Nested Learning: Collaborative Training of Self-Referential Memories for Test-Time Adaptation
- Federated Spectral-Guided Riemannian Graph Learning
- Federated Unlearning via Dual Distillation
- Federation of Experts: Communication Efficient Distributed Inference for Large Language Models
- FedGeo: Fine-Tuning-Free Federated Personalization via Geometric Stability
- FedLBA: Label-Semantic-Aware Classification Head Aggregation for Heterogeneous Federated Learning
- FedMFMS: Federated Robust Model Fitting for Motion Segmentation
- FedMoE-PFR: A Mixture-of-Experts Framework for Personalized Federated Recommendation
- FedPeel: Peeling Stabilized Layers for Robust Heterogeneous Federated Learning
- FedSmoothLoRA: Toward Smoother and Faster Convergence in Federated Low-Rank Adaptation
- FedSpec: A Communication-Efficient Federated Prompt Learning Method for Heterogeneous Clients via Spectral Disentanglement
- FedTLR: Federated Time-Series Learning via Temporal Latent Transition Regularization
- FedTTA-TS: Federated Test-Time Adaptation for Non-stationary Time-Series Forecasting
- FedZTA: A Federated Zero Trust Architecture for Securing Multi-Agent AI Systems
- Feedback World Model Enables Precise Guidance of Diffusion Policy
- Feed-Forward Gaussian Splatting from Sparse Aerial Views
- Feed-forward Neural Feature Compression for 3D Gaussian Splatting
- Feeling of Knowing in Large Language Models
- Feeling the Space: Egomotion-Aware Video Representation for Efficient and Accurate 3D Scene Understanding
- FELPS: Fair and Efficient Scheduling for Multi-LoRA Serving System
- Fenchel Dual Unlearning: Stable and Complete Machine Unlearning via Representation-Space Manipulation
- FEP-Agent: Grounding LLM Agent Self-Evolution in Active Inference with Semantic Memory
- FETTUCCINE: Fast and efficient brain-to-text decoding on mobile devices
- Few Channels Draw The Whole Picture: Revealing Massive Activations in Diffusion Transformers
- Fewer Tokens, Fewer Layers: Efficient Vision Token Pruning and On-Policy Distillation to Accelerate VLMs
- FFR: Forward-Forward Learning for Regression
- FGRPO: Federated GRPO with Adaptive Aggregation on Non-IID Data
- FHEvolve: An Evolutionary Framework for Automated Optimization of FHE Code
- FHIR-SetupBench: Evaluating AI Agent Capabilities in Baseline FHIR Environment Construction
- Fibration Compression in Deep Neural Networks
- Fidelity-Diversity Metrics for Text
- Fighting Fire with Fire: Assessing Test Set Contamination Through Deliberate Training on Test Data
- Fill the GAP: A Granular Alignment Paradigm for Visual Reasoning in Multimodal Large Language Models
- Filtered-Trace Online Variational Training for Probabilistic Spiking Neural Networks
- FinAl: Fine-grained Alignment for Detail-Preserving Medical Vision-Language Pretraining
- Final-Layer Hidden-State Jumps in Transformer LM: Factors, a Remedy, and Effect on Interpretability
- Finance LLMs in Stock Signal Prediction from SEC Form 8-K Filings
- Finding the Right Noise: A Convergence Theory of PPO Fine-Tuning for Diffusion Models
- FindIt: A Format-Informed Visual Detection Benchmark for Generalist Multimodal LLMs
- FinDocMRE: A Benchmark for Document-Level Financial Multimodal Reasoning Evaluation
- Fine-Detail Monocular Geometry Estimation with Self-Guided Sparse Volumetric Refinement
- Fine-T2I: An Open, Large-Scale, and Diverse Dataset for High-Quality T2I Fine-Tuning
- Finetuning and Post-training Camera-Controlled Video Generation at Scale
- Fine-Tuning Improves Information Conveyance in Language Models
- Fine-tuning language encoding models on slow fMRI improves prediction for fast ECoG
- Fine-Tuning Language Models to Know What They Know
- FineVLA: Fine-Grained Instruction Alignment for Steerable Vision-Language-Action Policies
- Finite-Particle Rates for Regularized Stein Variational Gradient Descent
- Finite-Rank Attention Metrics Reveal a Temporal-Change Mechanism in a Geophysical Flow Forecaster
- Finite Resources False Discovery Rate Control on Structured Hypothesis Spaces
- Finite-Sample Convergence in Networked Average Reward MARL: Decentralization Pitfalls and Entropy Remedies
- Finite Time Guarantees for Sample-Based Natural Policy Gradient in Finite-Horizon Markov Decision Processes
- FinReasoning: A Hierarchical Benchmark for Reliable Financial Research Reporting
- Finsler-Randers Metric Learning for Direction-Aware Latent Space Interpolation
- FinTSBridge: A Three-Layer Evaluation Framework for Financial Time Series Forecasting
- First 3D geological dataset for self-supervised representation learning and sedimentation environment characterisation
- First Impressions Last: Evaluating Anchoring and Evidence Insensitivity in Sequential Clinical Triage
- First-Order Regret for Online Convex Optimization with Memory and Online Nonstochastic Control
- FISC: Time-Series Forecasting via First-Layer Statistical Calibration Constraints
- Fisher-Guided Submodular Data Selection for Continual Pre-Training of Large Language Models
- Fisher-R1: Training LLM Agents for Reliable Hypothesis Testing
- FitLight: Federated Imitation Learning for Plug-and-Play Autonomous Traffic Signal Control
- FIT to Forget: Robust Continual Unlearning for Large Language Models
- FitVTON: Fit-aware Virtual Try-On via Body-Garment Size Control
- FIVE-VLA: Fast and EffectIVE Closed-Loop Autonomous Driving with Recurrent Action Memory
- Fixed Universal Transformers
- Fixel : A Plausible 2D-to-3D Top-View Floor Plan Generation Framework
- Fixing the Last Mile: Targeted Local Refinement for Perceptual Acceptability in Image Restoration
- Fix-Test Collapse: Diagnosing and Resolving Self-Confirming Verification in Code Agents
- FLAME: Flow Enhanced Legendre Memory Models for General Time Series Forecasting
- FLARE: Full-Modality Long-Video Audiovisual Retrieval Benchmark with User-Simulated Queries
- FLARE: Verifying MILP Reformulations with LLM-Based Formal Proof Synthesis
- FlashMol: High-Quality Molecule Generation in as Few as Four Steps
- Flash-WAM: Modality-Aware Distillation for World Action Models
- Flat-Corridor Quotients and the Computational Frontier of Exact Full \(Q^\pi\)-Realizability
- FLAT: Feedforward LAtent Triangle Splatting for Geometrically Accurate Scene Generation
- FlexGrad: A Root-Free Approach to Adaptive Step Sizes
- Flexible Doubly Robust Proximal Causal Inference for Censored Outcomes
- Flexible Intensities Matter: A comprehensive re-evaluation of Classical and Neural Temporal Point Processes
- FlexLAM: Resolving the Bottleneck Trade-off in Latent Action Learning
- FlexMoE: One-for-All Nested Intra-Expert Pruning for MoE Language Models
- FlexMoRE: A Flexible Mixture of Rank-heterogeneous Experts for Efficient Independently-trained Large Language Models
- FlexMS: Unified Public Benchmarks for Difficulty-Aware Tandem Mass Spectrometry Prediction
- Flipping Bits, Not Gradients: Sharpness-Aware Minimization Directly on the Boolean Hypercube
- Floating-Point Networks with Automatic Differentiation can Represent Floating-Point Functions up to Second-Order Derivative
- FLoRA-Chef: Making A Good LoRA Recipe in Federated Generalization
- Flow Annealing Posterior Sampling for Function-Space Regression and Inverse Problems
- FlowBanzhaf - A Graph-Level Generalisation Regression Dataset of Network-Flow Graphs
- FlowEdit: Information-Theoretic Control of LLM Reasoning Flows for Ill-posed Problems Involving Conflicts
- Flower: Conditional Representations With Information Constraints From Flow Matching
- Flower Hub: A Reproducible Benchmarking Platform for Federated Learning in Simulation and Deployment
- FLOWFORGE: A Staged Local Rollout Engine for Flow Field Prediction
- Flow Latent Stratification for Monte Carlo Variance Reduction
- FlowLong: Inference-time Long Video Generation via Manifold-constrained Tweedie Matching
- Flow Map Denoisers: Traversing the Distortion-Perception Plane for Inverse Problems
- Flow Map Language Models: One-step Language Modeling via Continuous Denoising
- Flow-Matching Based Refiner for Molecular Conformer Generation
- Flow Matching in Feature Space for Stochastic World Modeling
- Flow Matching is Adaptive to Manifold Structures
- Flow Matching on Symmetric Spaces
- Flow Matching Reinforcement Learning via SDE Inference
- Flow-of-Thought: A Framework for Visual Reasoning
- Flow Q-trace: Multi-step Critic Learning for Offline Reinforcement Learning
- FlowSteer: Prompt-Only Workflow Steering Exposes Planning-Time Vulnerabilities in Multi-Agent LLM Systems
- FlowSteer: Towards Agents Designing Agentic Workflows via Reinforced Progressive Canvas Editing
- Fluent Reasoning Is Not Auditable Reasoning: Temporal Constraint Sustainability Is a Required Evaluation Property for Delegation under Ambiguity
- FluidPipe: Eliminating the Pipeline Bubble through Local Learning
- FluoroBench: A Benchmark for Solvent Aware Fluorophore Photophysical Property Prediction
- FML-bench: A Controlled Study of AI Research Agent Strategies from the Perspective of Search Dynamics
- FMMI: Flow Matching Mutual Information Estimation
- FOAM: Factored One-sided Adam-Moment for Practical and Scalable SOAP
- Focal Point Convergence in LLM Pricing Games: Prior Entropy, Not Economic Reasoning, Drives Coordination
- Focusable Monocular Depth Estimation
- FocusBranch: Combinatorial Branch-and-Bound for $\ell_0$ Neural Network Robustness Verification
- Focus Calibration for Ultra-Low-Bit LLM Inference on Bandwidth-Limited Edge Devices
- Focused Forcing: Content-Aware Per-Frame KV Selection for Efficient Autoregressive Video Diffusion
- FocuSFT: Bilevel Optimization for Dilution-Aware Long-Context Fine-Tuning
- FogGS: Physics-Grounded 3D Foggy Effects
- FoldAct: Efficient and Stable Context Folding for Long-Horizon Agents
- Fold-CP: Context Parallelism for Structural Biology
- FoldUMM: Unlocking Efficient Inference in Unified Multimodal Models via Task-Specific Layer Folding
- FOMCBench: A Cross-Domain Evasion Detection Benchmark for Central Bank Communication
- FORAGE: Related Works Prediction as a Benchmark for Agentic Retrieval
- Forced Orders: What LLM Leaderboards Hide About Model Comparisons
- Forcing Clinical Foundation Models to Fail Safely: Event-Conditional Calibration for Rare-Event Survival Prediction
- Forecasting Downstream Performance of LLMs With Proxy Metrics
- Forecasting with Factor-Augmented Time Series Foundation Models
- Foresight: World Model-Guided Reinforcement Learning for Safe LLM Agent Planning
- FORGE: Business-Logic-Aware Agent Synthesis through Compositional Function Slices
- Forgetting in Language Models: Capacity, Optimization, and Self-Generated Replay
- Forgetting to Improve: Principled Data Removal in Active Learning
- Formal-AVS: a Lean benchmark for anytime-valid confidence-sequence theorem proving
- Formalizing Latent Thoughts: Four Axioms of Thought Representation in LLMs
- Formalizing User Preference Drift Tracking in Personalized Systems
- Form-Aware Routing for Structure-Aware KV Cache Eviction
- FORMBENCH: A Benchmark for Fine-Grained Retrieval in Formulation Patents
- For Questions of Ought, AI Could Use Some SAGE Advice
- Forward Shapley Scoring for Non-Myopic Active Feature Acquisition
- Foundation Models for Temporal Systems: From Forecasting to World Modeling
- Foundation Models for the Brain and Body
- Foundation Pareto Flow Policy for Multi-Objective Reinforcement Learning
- Foundations of Agentic Systems Theory (FAST)
- Foundations of Categorical Equivariant Deep Learning
- Foundations of Language Model Security: Theory, Practice, and Fundamental Limits
- Foundry: A Control Plane for Long-Horizon Agent Swarms
- Fourier Feature Pyramids for High-Precision Physics Informed Neural Networks
- Foveated Spatial Priors for Activation Quantization in Vision Transformers
- FP4 Explore, BF16 Train: Diffusion Reinforcement Learning via Efficient Rollout Scaling
- FracAug: Fractional Augmentation boost Graph-level Anomaly Detection under Limited Supervision
- Fractal Deep Equilibrium Models
- Fractal-G: Topology-Aware Heterogeneous Graphs for Medical Image Segmentation
- FractalLoss: Fractal Structure in Non-Dominant Frequencies Matters for Time Series Forecasting
- Fractional State Space Transition for Long Sequence Modeling
- FractiT: Differentiable Fractal Conditioning for Multivariate Time Series Forecasting
- Fragmentation is Efficiently Learnable by Quantum Neural Networks
- Fragment-Aware Graph Transformer with Label Prototypes for Multilabel Odour Prediction
- Frank-Wolfe Beyond 1/t Convergence
- FrED: External Data Influence Estimation via Domain Knowledge Graph Grounding
- Free Cumulants and the Depth–Width Phase Transition: A Non-Commutative Probability Theory for Deep Random Networks
- Freeing the Law with LOCUS: A Local Ordinance Corpus for the United States
- FreezeVLA: Action-Freezing Attacks against Vision-Language-Action Models
- FreqGS: Frequency-Guided Gaussian Splatting for High-Fidelity Novel View Synthesis from Monocular Street Scenes
- Frequency-Aware Graph Spectral Brain Network
- Frequency-aware Stable Ranking for Reliable Offline Model-Based Optimization
- Frequency-Guidance Diffusion Network for Noise Entanglement in Ultrasound Image Classification
- Frequency-Structured Field Learning for Light-Field Disparity Estimation
- Fre-Res: Spatial Anchors and Temporal-Frequency Residuals for Scalable Video MLLMs
- Freshness-Gated Imagination: Step-Level Trust for Latent World Models
- FRInGe: Distribution-Space Integrated Gradients with Fisher–Rao Geometry
- From Adapter Updates to Scale Updates: Rethinking Quantized Fine-Tuning
- From Alignment to Fusion in 3D Vision-Language
- From Anti-Forgetting to Fast Adaptation: Continual Reinforcement Learning with World Models
- From Approximation to Computation: Universal Power of Deep Narrow Networks at Constant Width
- From Assumption to Diagnostics: Rethinking Polyak–Lojasiewicz (PL) for Proxy Reliability in Deep Learning
- From Capability to Alignment: Repurposing MoE Routing for Safety
- From Chats to Markets: AgenticPay for LLM-Powered Negotiation in Multi-Agent Commerce
- From Classical to AI-Augmented: A Benchmark for Evolving Shortest Vector Problem Solvers
- From Clever Hans to Scientific Discovery: Interpreting EEG Foundational Transformers with LRP
- From Clips to Streams: A Unified Framework for Streaming Sign Language Translation
- From Comparison to Composition: Towards Understanding Machine Cognition of Unseen Categories
- From Competition to Collaboration: Designing Sustainable Mechanisms Between LLMs and Online Forums
- From Compression to Selection: Fine-Tuning Large Models with Sparse Adaptation
- From Critical Paths to Critical Cells: Learning Fine-Grained Timing Attribution for Chip Design
- From Cursed to Competitive: Closing the ZO–FO Gap via Input-to-State Stability
- From Data to Laws: Neural Discovery of Conservation Laws Without False Positives
- From Density Matrices to Phase Transitions in Deep Learning: Spectral Early Warnings and Interpretability
- From Descriptive to Functional: Rethinking Prototypes in Time Series Classification
- From Distance to Likelihood: The $k$-Likeliest Neighbor Classifier
- From Document Layout to Causal Topology: An End-to-End Architecture for Temporal Causal Reasoning
- From Empirical Fairness to Certified Fairness: Wasserstein Robust Audits Under Distribution Shift
- From End-to-End to Step-by-Step: Learning Composable Navigation Primitives for Vision-Language Navigation
- From Examples to Skills: A Self-Auditing Memory Harness for Frozen LLMs
- From Expert Knowledge to Optimization Modeling: Prototype-Based Data Synthesis and Logical Reinforcement Learning
- From Fixed to Learned: A Unified Theory of Architectural Innovation as Dimension-wise Selection
- From Frames to Sequences: Temporally Consistent Human-Centric Dense Prediction
- From Generalists to Specialists: Model Speciation via Interventional Sparsification
- From Generation to Restoration: Residual Diffusion for Neural Channel Decoding
- From Generic Correlation to Input-Specific Credit in On-Policy Self Distillation
- From Global to Local: Rethinking CLIP Feature Aggregation for Person Re-Identification
- From Hallucination to Deduction: Knowledge-Rule Preference Optimization for Structured Clinical Reasoning
- From History to State: Constant-Context Skill Learning for LLM Agents
- From Ideas to Code: Tree-structured Policy Optimization for Automated Algorithm Design with LLMs
- From Idea to User Journeys: RL Fine-Tuning of LLM-Based UI Code Generators with UI/UX Aesthetic, Usability and Scenario-Aware Rewards
- From Imitation to Understanding: Emergent Expert Modelling in Social Learning
- From Inexact Gradients to Byzantine Robustness: Acceleration and Optimization under Similarity
- From Intent to Evidence: A Categorical Approach for Structural Evaluation of Deep Research Agents
- From Isolated feature to Orbits:\\Discovering Music Concepts via Multi-SAE Alignment
- From Latent Signals to Reflection Behavior: Tracing Activation Trajectory in R1-Style LLMs
- From Lie Symmetry Orbits to Sparse PDE Discovery under Localized Observations
- From Matching to Reasoning: Query-Aware Long Video Summarization
- From Multi-Agent to Single-Agent: When Is Skill Distillation Beneficial?
- From Natural Language to Extensive-Form Game Representations: Defining and Automatically Verifying Consistency
- From Nodes to Pixels: Topological and Structural Two-View Graph Imaging
- From Operator Zoo to Möbius Family: Unified Nonlinear Lowering for Efficient LLM Inference
- From Optimization to Self-Regulation: Why AI Needs Emotion–Personality Engines
- From Order to Distribution: A Spectral Characterization of Forgetting in Continual Learning
- From Papers to Property Tables: A Priority-Based LLM Workflow for Materials Data Extraction
- From Persistence to Survival: Hypothesis Testing, Effect Sizes and Vectorisation for Topological Features
- From Pixels to Primitives: Scene Change Detection in 3D Gaussian Splatting
- From POMDP Theory to Deep RL with Particle Filters
- From Projection to Attention: Tracing the Connector Cascade in Vision-Language Models: A Closer Look
- From Proximity to Relevance: Learning Traffic-Aware Scene Graphs for Motion Forecasting
- From Random to Structured Perturbations: Cascaded Adversarial Noise Quantizer for Robust Medical Image Translation
- From Recognition to Understanding: Unlocking Cognitive Time Series Reasoning with LLMs
- From Retrieval to Re-Linkability: Identity Leakage from Released Vision-Language Person Embeddings
- From Sample to Subset Construction: Coverage-Aware Curation of Robot Demonstrations
- From Seeing to Foreseeing: Unleashing LVLM Thinking in Dynamic Latent Space
- From Sequential Nodes to GPU Batches: Parallel Branch and Bound for Optimal $k$-Sparse GLMs
- From Snapshots to Trajectories: Benchmarking Senescence-Conditioned Cell Morphology Generation
- From Snapshot to Trajectory: Flow-Constrained Distillation for Large Language Models
- From Sparse Sensors to Continuous Fields: STRIDE for Spatiotemporal Reconstruction
- From Stochasticity to Signal: A Bayesian Latent State Model for Reliable Measurement with LLMs
- From Storage to Steering: Memory Control Flow Attacks on LLM Agents
- From surveillance to signalling: escalation channels as environmental controls for agentic AI
- From Sweep to Seam: Interleaved Cross-Block Post-Training Quantization
- From Synapses to Softmax: A Biophysically Grounded Attention Mechanism Derived from Hippocampal Phase Precession
- From Table to Cell: Attention for Better Reasoning with TABALIGN
- From Task Affinities to Optimal Allocations: Probabilistic Graphical Modelling for Supervised Finetuning
- From Tokens to Numbers: Continuous Number Modeling for SVG Generation
- From Token to Structure: Efficient Agent Memory Retrieval via Layout-Aware Visual Anchoring
- From Tool Use to Tool Reasoning: Auditable Pixel Trajectories in Vision-Language Models
- From Topology to Retrieval: Decoding Embedding Spaces with Unified Signatures
- From Trace to Diagnosis and Text: TRACE for Unified ECG Understanding
- From Views to Worlds: Active Exploration over 3D Worlds for Vision-Language Models
- From Visibility to Visual Grounding: Understanding Disability Bias in Vision-Language Models
- From Weak Data to Strong Policy: Q-Targets Enable Provable In-Context Reinforcement Learning
- From “Weak” Signals to Strong Models: Preference Delta Aggregation with LoRA Merging
- From Words to Amino Acids: Does the Curse of Depth Persist?
- From Zero to Hero: Training-Free Custom Concept Spawning in World Models
- Frontier LLMs See Too Much: The Case for Trusted Execution Environments
- FrontierSmith: Synthesizing Open-Ended Coding Problems at Scale
- Frost Training and Cross-Entropy Games
- FSAN: Flow State Attention Network for Aerodynamic Prediction
- FTerViT: Fully Ternary Vision Transformer
- Full Attention Strikes Back: Transferring Full Attention into Sparse within Hundred Training Steps
- FullStackArena: Evaluating Browser Agents in Dynamic Full-Stack Environments
- Full-Stack FP4: Stable LLM Pretraining with Quantized Projections, Optimizers, and Attention
- Fully First-Order Algorithms for Online Non-Convex Bilevel Optimization
- Fully Kolmogorov-Arnold Deep Model in Medical Image Segmentation
- Fully-Marginalised Particle Gibbs with Ancestor Sampling for Conditionally Tractable State-Space Models
- FuncFormer: Circuit Representation Learning via the Flow of Functional Propagation
- Function-Coherent Bellman Equations: Log-Optimal Reinforcement Learning for Multiplicative Dynamics
- Function graph transformers universally approximate operators between function spaces
- Fundamental Limits and Utility-Privacy Tradeoffs in Machine Unlearning with Partial Information
- Fundamental Limits of Knowledge Discovery Through AI (Extended)
- FuRA: Full-Rank Parameter-Efficient Fine-Tuning with Spectral Preconditioning
- Fused Gromov–Wasserstein Distance with Feature Selection
- Fusion K-Means: Adaptive Cluster-Number Selection by Center Fusion
- FusionNeXt: Sequence-First 3D Multi-Modal Fusion in the Era of LLMs
- Fusion or Confusion? Multimodal Complexity Is Not All You Need
- FutureMap: Web-Scale Future Prediction Agents with Omnimodal Dynamic HyperGraph Speculation
- Future Validity is the Missing Statistic: From Impossibility to $\Phi$-Estimation for Grammar-Faithful Speculative Decoding
- fxBench: Evaluating and Understanding Formula Suggestions in Spreadsheets
- G2-RASE: Graph-Guided Reasoning with Analogical Structured Exemplars
- GACE: Guidance, Assessment, and Correction for EEG Foundation Models
- GAC: Stabilizing Asynchronous Off-Policy RL for LLMs via Gradient Alignment Control
- GAD in the Wild: Benchmarking Graph Anomaly Detection under Realistic Deployment Challenges
- GADMVP: Adaptive Few-Shot Graph-Level Anomaly Detection with Multi-View Structured Prompting
- GAIN: Multiplicative Modulation for Domain Adaptation
- Game Based Inference, Recovering Hidden Constraints from Optimal Play
- Gated DeltaNet-2: Decoupling Erase and Write in Linear Attention
- Gate the Filter, Not the Message: Node-Channel Mixtures for Pre-Propagation GNNs
- GAUGECAST++: A Physics-Informed Latent Forecasting System for Localized Flood Prediction
- Gauge Symmetries and Identifiability of Counterfactual Generators
- Gaussian Density Splatting Network
- Gaussian Sheaf Neural Networks
- GazeGGT: Geometry-Grounded Transformer for Uncalibrated Multi-View Gaze Estimation (Extended)
- GazeMind: A Gaze-Guided LLM Agent for Personalized Cognitive Load Assessment
- GCE-MIL: Faithful and Recoverable Evidence for Multiple Instance Learning in Whole-Slide Imaging
- GDAFormer: Geometry-Guided Deformable Attention for Robot-Centric Multimodal Video Panoptic Segmentation
- GD-FPS: Growth-Driven Feedforward Parameter Selection for Efficient Fine-Tuning
- GEAR: Bridging the Planner-Actor Gap via Gradient-Aligned Policy Extraction
- GEAR-FEN: Generalized Feature Representation for Human Activity Recognition
- GEAR: Generative Efficient Gap Amplification for Goal-Conditioned Reinforcement Learning
- GEAR: Granularity-Adaptive Advantage Reweighting for LLM Agents via Self-Distillation
- GEARS: Geometry-Enhanced Adaptive Reward Shaping for Parametric CAD Generation
- GEBench: Benchmarking Image Generation Models as GUI Environments
- GeLVR: Geometry-Consistent Latent Visual Reasoning in Multimodal LLMs
- GEM: A Dual-Scale Architecture for Graph-Level Hierarchical Representation Learning
- GEM: Graph-based Evolutionary Metacognition in Self-Improving Agentic Systems with Observational Causal Designs
- GEM: Interpretable Language Models via Geometric Embedding Alignment
- GEMS-3D: A Large-Scale 3D Gravity, Electrical, Magnetic, and Seismic Earth Simulation Dataset for Multimodal Geophysical Learning
- GEMS: Molecular Structure Identification via Geodesic Navigation of the Isomer Manifold
- GenAlphaBench: A Safety-Annotated Benchmark for Evaluating LLM Comprehension of Youth Digital Language
- GenC-Surv: Generative Causal Learning with Variational Inference for Dynamic Censored Data
- Generalising Travel Time Prediction To Varying Route Choices In Urban Networks
- Generalizable Next Location Prediction via Spatial Coordinates and LLM-Enhanced POI Semantics
- Generalization Analysis of Biased Stochastic Gradient Methods for Minimax Problems
- Generalization Bounds of Gradient Descent in Over-parameterized Vision Transformers
- Generalization Error Bounds for Picard-Type Operator Learning in Nonlinear Parabolic PDEs
- Generalization for Time Series in Tight Settings: Latency, Inference, Memory, prIvacy and Sustainability (TS-LIMITS)
- Generalization Guarantees for Uniform Random Interpolators in Over-parameterized Deep Neural Networks
- Generalization in Neural Networks Through the Lens of Magnitude Potential
- Generalization vs. Memorization in Partial Differential Equation Emulators: Or, Training Dynamics of Cross-Time and Cross-Class Gradient Alignment
- Generalization Without Compression Penalty: A Stability Analysis of Error Feedback
- Generalized Counteractive Reinforcement Learning
- Generalized Intention Modeling in Multi-Agent Reinforcement Learning
- Generalized Mean Absolute Directional Loss for Machine Learning Trading Models
- Generalized Peptide Design with Physically Inspired Diffusion
- Generalizing Preference-based Reinforcement Learning: a Rationality Model for Incomparability
- Generate in Reconstruction Space, Match in Semantic Space: Transport Geometry for One-Step Generation
- Generating in the Limit with Infinitely Many Hallucinations
- Generating Protein Thermodynamic Ensembles Using Distogram Flow Matching and a New Dataset
- Generating Symmetric Materials using Latent Flow Matching
- Generating the Wild: Individual-Consistent Image-to-Video Generation for Wildlife
- Generation Navigator: A State-Aware Agentic Framework for Image Generation
- Generation Order and Parallel Decoding in Masked Diffusion Models: An Information-Theoretic Perspective
- Generative Active Learning via Bayesian Acquisition for Improving the Efficiency of Synthetic Data
- Generative Actor-Critic with Soft Bridge Policies
- Generative Augmentation of Single Image Prior for Black Hole Imaging
- Generative collapse does not imply discriminative collapse
- Generative Conformal Prediction with Optimized Coverage Allocation
- Generative Conversational Recommender System
- Generative Modeling via Drifting
- Generative Multi-modal Gradient Matching
- Generative Semantic Communication Should Decouple Receivers from Generative Decoders
- Generator-based Graph Generation via Heat Diffusion
- GenShadow: Efficient Quantum Observable Estimation via Generative Learning of Adaptive Measurement Strategies
- GenSpan: Generation-Calibrated Motion Span Priors for Multi-Verb Video Corpus Moment Retrieval
- GeoBA: Geometry-Biased Attention for Molecular Property Prediction
- GeoBiaset: A Counterfactual Benchmark for Demographic Bias in World-Level Geolocalization
- GeoCore-9B: Towards Geo-Aware Generative Foundation Models in Earth Observation
- GeoDiffusion: Offset-Aware Diffusion for Irregular Spatio-Temporal Sequence Modeling
- Geofield-LoRA: Rendering Task-Specific Adapters from a Shared Gaussian Primitive Field
- GeoFlowVLM: Geometry-Aware Joint Uncertainty for Frozen Vision-Language Embeddings
- GeoG2U-Bench: When Does Generation Help Understanding in Ultra-High-Resolution Remote Sensing?
- GeoGR: Bridging Generation and Retrieval with Confidence-Aware Fusion for Image Geolocation
- GeoKAO: Geometry-Aware Kernel Adaptive Optimization for Diffusion-Based Satellite Image Inpainting in Remote Sensing
- Geometric Analysis of Neural Regression Collapse via Intrinsic Dimension
- Geometric and Topological Regularization for Robust Bioimage Segmentation
- Geometric Factual Recall in Transformers
- Geometric Flow Matching
- Geometric Gain Graph: Zero-Token Graph Construction for Multi-Hop RAG
- Geometric Inconsistency in EDL Regularization: FI-EDL Improves Calibration via Fisher-Informed Adaptive KL Weighting
- Geometric Inductive Biases for Semi-Supervised Equalization: The Constellation-Aware Transformer
- Geometric Latent Reasoning Induces Shorter Generations in LLMs
- Geometric Representation Training: Verifier-Labeled Subspace Contrastive Learning
- Geometry-Aware Bayesian Quantification via Compositional Data Analysis
- Geometry-Aware Directional Alignment for Coherent Model Merging
- Geometry-Aware Distributionally Robust Preference Optimization for LLM-based Recommendation
- Geometry-Aware Flow Matching for Sparse-View 3D Gaussian Splatting
- Geometry-Aware Gradient-Free Optimization via Curvature Estimation for Biophysical Models
- Geometry-Aware Local Control Under Recoverability Constraints for LLM Reasoning
- Geometry-Aware Representation Alignment Enhances Native Spatial Intelligence in MLLMs
- Geometry-Derived Hierarchical Conditional Generation for Protein Inverse Folding
- Geometry Guided Self-Consistency for Physical AI
- Geometry-Guided Semantic Reconstruction for 3D Instance Segmentation
- Geometry Without Coordinates: LiDAR Diffusion as a 3D Feature Bridge
- Geometry Without Geometry: Operational Collapse in Geometric Graph Neural Networks
- GeoR-Bench: Evaluating Geoscience Visual Reasoning
- GeoReason: Step-Level Hallucination Detection via Hidden-State Trajectory Geometry
- GeoRecon: Graph-Level Representation Learning for 3D Molecules via Reconstruction-Based Pretraining
- GeoReForm: Reflective Formalization Evolution for Multimodal Geometry Problem Solving: A Closer Look
- GeoSCoRe-R: Causal Spatial Verification for Video Spatial Reasoning
- Geospatial Latent Privileged Distillation for Remote Sensing Quantitative Reasoning
- GeoSym127K: Scalable Symbolically-verifiable Synthesis for Multimodal Geometric Reasoning
- GeoTransolver: Learning Physics on Irregular Domains using Multi-scale Geometry Aware Physics Attention Transformer
- GeoWorld-VLM: Geometry from World Models for Vision-Language Models
- GestureTraces: A Feature Selection Strategy for Controlling Identity Leakage in Sensor Data Collected from Human Touch
- gfnx: Fast and Scalable Library for Generative Flow Networks in JAX
- GGBound: A Genome-Grounded Agent for Microbial Life-Boundary Prediction
- GGT-100K: Generative Ground Truth for Generalizable Real-World Image Restoration
- GHOST: Gaussian Hypothesis Orientated Source Transfer
- GHOST: Geometry-Hierarchical Online Streaming Token Eviction for Efficient 3D Reconstruction
- Ghost Labels and Causal Validity: A Gate-Resolved Audit of Probe-Based Knowledge Claims
- GhostQuant: The Ghost that Appears Only After Quantization
- GIFT: Representation Geometry Matters for Single-Domain Generalized Object Detection
- GIM: Evaluating models via tasks that integrate multiple cognitive domains
- Gini-LoRA: Cost-Aware Rank Allocation via Importance Concentration
- GISA: A Benchmark for General Information-Seeking Assistant
- GLACIER: Bridging Chemical Perturbation Prediction and Transcriptomic-Based Drug Design via Latent Auto-Regressive Transformers: A Closer Look
- GLASS-LoRA: Serving-Aligned Structured LoRA Export via Hierarchical Sparsity
- GLNet: Hierarchical Global-Local Representation Learning for Non-Stationary Time Series Forecasting
- Global Convergence and Better Spectral Bias in Low-Rank Neural Networks
- Global Importance Estimation for KV Cache Eviction
- Globally Convergent Offline Reinforcement Learning with Smoothed Bellman Residual Minimization
- Global Metrics Hide Local Failures in Hyperbolic Hierarchy Embeddings
- GLOBE
- GlucoFM: A Dual-Stream Foundation Model for Continuous Glucose Monitoring
- GME-Init: Gamma-Moment Equalization Initialization for Data Awareness LoRA
- GNES: Neural-Guided Evolutionary Program Search for Interpretable Multi-Agent Control
- Goal-Directedness is in the Eye of the Beholder
- Goals as dynamical attractors: a momentum-based account of stable and flexible goal commitment
- GOAT-AL: Pseudo Neural Collapse Guides Adaptive Coverage for All-Budget Active Learning
- GoBOED: Goal-Driven Bayesian Optimal Experimental Design with Differentiable Convex Optimization
- Going Down Memory Lane: Scaling Tokens for Video Stream Understanding with Dynamic KV-Cache Memory
- Going Sparser with Test-Time Training
- Going the distance: substitution-based distances for structured evaluation of protein generative models
- Goldilocks RL: Tuning Task Difficulty to Escape Sparse Rewards for Reasoning
- Good Experience Maximization
- Governing the Post-Pretraining Era: H2E Geometric Governance as the Bounded-Error Safety Layer for Agentic AI
- G-PAC: A Game-theoretic framework for Privacy-aware Anomaly detection over Command-line sequences
- GPercept: Towards Human-Aligned Graphical Perception in Charts via Elementary Grounding
- GPT2MEG: Quantizing MEG for Autoregressive Generation
- GPT4D: Generative Pre-training Transformer with Next-Scale Spatio-temporal Token Prediction for 4D Human Action Recognition
- GRABS
- GRACE: Learning Editable Gravitational Environment Fields for Counterfactual Many-Body Astrophysics
- Grad Detect: Gradient-Based Hallucination Detection in LLMs
- Gradient-based Graph Structure Optimisation for Research Networks
- Gradient-based inference of task abstractions for generalization in neural networks
- Gradient Coloring: Expanding Gradient Diversity with Bayesian Color Mapping for Enhanced Adversarial Robustness
- Gradient-Informed Temporal Sampling Improves Rollout Accuracy in PDE Surrogate Training: A Closer Look
- Gradient Redirection: A Fourth Mechanism of Gradient Obfuscation via Fixed Encodings
- Gradient Routing Localizes and Removes Unintended Behaviors in RL
- Gradients through Logarithmic Scaling Formulations
- Gradient Surgery Backfires on Attention: A Decoupled-Query Fix
- GRAM: Group-wise Rank-Aware Modal Merging via Subspace Alignment
- GramStatTexNet: Efficient, Interpretable, and Neuro-Inspired Texture Analysis-by-Synthesis
- Granule-R1: Self-Adaptive Granularity Management for Agentic QA via Boundary-Aware Multi-LLM Routing
- Graph2TS: Structure-Conditioned Time Series Generation from Quantile Graphs
- Graph Anomaly Detection as Dynamical Transport: Training-Free Scoring via Empirical Bayes
- Graph-Based Stochastic-Power-UCT: Monte-Carlo Graph Search with Power Mean Estimation
- GraphBridge: Safe Graph-Mediated Transfer under Structural Modality Missingness
- Graph Cascades: Contagion-Based Mesoscopic Rewiring for Structure-Aware Graph Machine Learning
- Graphene: A Lightweight Transformer with Ising Mean-Field Refinement for Image Super-Resolution
- Graph Energy Matching: Transport-Aligned Energy-Based Modeling for Graph Generation
- Graph Federated Unlearning for Privacy Preservation
- Graph Foundation Models Are Cheap to Steal
- Graphical einops: bridging tensor networks and computation graphs
- GraphInfer-Bench: Benchmarking LLM's Inference Capability on Graphs
- GraphInstruct: A Progressive Benchmark for Diagnosing Capability Gaps in LLM Graph Generation
- GRAPHLCP: Structure-Aware Localized Conformal Prediction on Graphs
- GraphMD: A Multiscale Geometric Deep Learning Framework for Protein-Ligand Binding Affinity and Conformational Dynamics Prediction
- GraphMemRL: Action-Native Reinforcement Learning for Persistent Graph Memory Construction
- GraphNetz: Statistical Benchmarking of Graph Neural Networks with Paired Tests and Rank Aggregation
- Graph Neural Networks with Triangle-Based Messages for the Multicut Problem
- GraphPrompt-CLIP: Gromov-Wasserstein Structure-Conditioned Prompt Tuning for Frozen Vision-Language Models
- Graph Representation via Elements of Discrete Morse and Cobordism Theories
- GraphSolver: Graph-Constrained Receding-Horizon Control for Training-Free Interactive Agents
- GraphToken: A General Graph Vocabulary for Text-attributed Graph Foundation Models
- Graph Topology Augmentation for Prioritized Sweeping in Non-stationary Reinforcement Learning
- Graph Transductive Sharpening: Leveraging Unlabeled Predictions in Node Classification
- GraphUAT: Uncertainty Attribution in Graph Neural Networks
- GRASP: 2:4 activation sparsification for low-bit inference becomes near-lossless
- GraVa: Controlling Gradient Variance via Batch-Level Loss Variance for Stable Image Super-Resolution Training
- GRC: Unifying Reasoning-Driven Generation, Retrieval and Compression
- Greedy Multi-Path Block Verification for Faster Decoding in Speculative Sampling
- GridHunt: A Customizable Heterogeneous Predator-Prey Suite for Mixed-Motive MARL and Human-Agent Teaming
- GridProbe: Posterior-Probing for Adaptive Test-Time Compute in Long-Video VLMs
- GRIFDIR: Graph Resolution-Invariant FEM Diffusion Models in Function Spaces over Irregular Domains
- GRIK: Grouped Relevance-Integrated Key Pruning for Grouped-Query Attention LLMs
- GRL: Generalized R-learner for Estimating Potential Outcome and Heterogeneous Treatment Effects
- @GrokSet: multi-party Human-LLM Interactions in Social Media
- Grounded and Faithful Vision-Language Models for Real-World Deployment
- Grounded User Simulation for Model Evaluation and Training:\\ Diversity, Fidelity, and Validity
- Grounding Agent Reasoning with Structured Process Supervision for Multi-turn Reinforcement Learning
- Grounding Driving VLA via Inverse Kinematics
- Grounding Language to Structured 3D Motion in Complex 3D Scenes
- Grounding ML Data Exchange in Subjective Logic: Algebra, Interoperability, and Empirical Evaluation
- Ground It Before You Simulate It: The Case for Demographically Grounded LLM Simulations
- GroupAffect-4: A Multimodal Dataset of Four-Person Collaborative Interaction
- Group Perspective Matters: Regulating Debate Relationships Can Mitigate Blind Conformity in Multi-Agent Debate
- Group Preference Collapse in Personalized Multimodal Large Language Models
- Group selection promotes prosocial prompts in populations of LLM agents
- GROVE: Generative Spatial Deconvolution via Decoder Transfer and Cell Type Specific Spatial Modeling
- GRPO Does Not Close the Multi-Agent Coordination Gap
- GRPO-LLM: Group Relative Policy Optimization for Large Language Model Alignment
- GSMem: 3D Gaussian Splatting as Persistent Spatial Memory for Zero-Shot Embodied Exploration and Reasoning
- GSS: Compression and Extreme Parameter-Efficient Fine-Tuning of Diffusion Transformers via Globally-Sensitive Sketch
- GTSRB 15-Year Anniversary Edition: Traffic Sign Recognition with Same-Domain Unknowns
- GUI-AC: Enhancing Continual Learning in GUI Agents
- Guided Image Enhancement Path Diffusion Model
- Guided Trajectory Optimization with Sparse Scaling for Test-Time Diffusion
- Guiding Data Allocation for Robust Subpopulation Generalization
- Guiding Diffusion Samplers with Black-Box Optimization Solvers
- GUIGuard-Bench: Toward a General Evaluation for Privacy-Preserving GUI Agents
- GuiRouter: A Step-Wise Routing Framework for GUI Agents
- GVCC: Zero-Shot Video Compression via Codebook-Driven Stochastic Rectified Flow
- GVGAI-LLM: Evaluating Large Language Models with Infinite Games: A Closer Look
- GWScore: A structural diversity metric for consistent text-to-image generation
- GyroSafe: Outcome-Driven Closed-Loop Safety Alignment without Human Labels
- G-Zero: Self-Play for Open-Ended Generation from Zero Data
- H3D: Recovering Online Handwriting Dynamics from High-Resolution Offline Ink-on-Paper Traces
- Hack-Verifiable Environments: Towards Evaluating Reward Hacking at Scale
- Had It Not Been This One: Cross-Context Attribution for Entropy-Preserving LLM Post-Training
- Haiku to Opus in Just 10 bits: LLMs Unlock Large Compression Gains in Text
- Hallucination-Guided Unlearning: Using Hallucination Traces to Reveal Overfitted Memories
- Hallucination Mitigation via Directional Sample-Selective Contrastive Learning
- Hallucination Risk in Large Language Models: A Random Matrix Perspective
- HalluciText: Mitigating Text Hallucinations in Diffusion-Based Image Restoration
- HalluGuard: Decoupled Automatic Hallucination Detection and Mitigation in MLLMs: A Closer Look
- HalluText: Towards Benchmarking and Mitigating OCR Hallucination for Large Vision-Language Models
- HALMES: Knowing When to Intervene in LVLM Hallucination Mitigation
- HALO: Hierarchical Evidence Selection and Reinforcement Learning for Allocating Budgeted Label Optimization from Longitudinal Clinical Records
- HALO-VGGT: Heterogeneity-Aware Lightweight Online Compression Allocator for Efficient VGGT
- HandEdit: A Unified Benchmark for Egocentric Human-to-Robot Dexterous Hand Image Editing
- Handoff Fidelity: Diagnosing Evidence-Flow Failures at LLM Handoff Interfaces
- HAPS: Hierarchical LLM Routing with Joint Architecture and Parameter Search
- Hard Attention Transformers and BSS-Machines
- Hardening GraphRAG Against Poisoning Across the Knowledge Lifecycle
- Hardware-Aware Grouped Pruning for Efficient Vision Transformers
- HarmoGS: Robust 3D Gaussian Splatting in the Wild via Conflict-Aware Gradient Harmonization
- Harmony in Diversity: Multi-domain Contrastive Policy Optimization for Large Reasoning Models
- Harness as an Asset: Enforcing Determinism via the Convergent AI Agent Framework (CAAF)
- Hat-Muon: Communication-Efficient Polar Updates via Shared-Seed Random Projection and Local Error Feedback
- HazardArena: Evaluating Semantic Safety in Vision--Language--Action Models
- Hazard-Guided Generative Modeling for Sparse and Irreversible Transitions in Disease Trajectories
- HCDM: Controlled Diffusion Model for Hypergraph Representation Learning
- Head-Aware Token Selection in Latent Space for Scalable KV Cache Compression
- Head Similarity: Modeling Structured Whole-Head Appearance Beyond Face Recognition
- Heads That Write, Not Just Point: Image Retrieval Heads in Vision-Language Models
- Healthcare AI GYM for Medical Agents
- Healthcare Reasoning Quantization: Energy Decomposition and a Sparse-Activation Solution
- Hear, Localize, and Reason: Spatially Aware Scene Understanding for Audio-visual LLMs
- HearSayBench: Can LLMs Navigate from Abstract Human Rights to Lived Lives?
- Heartbeat-Level Masking for Lead-Agnostic Self-Supervised ECG Representation Learning
- HeatCache: Thermal-aware Energy-efficient LLM Inference Scheduling for Chassis-level Liquid Cooling in Sustainable Edge Server Rooms
- HEC-GNN: A Hierarchical Hardware-aware GNN for Chain Strength Selection in Quantum Annealing
- Hedging Memory Horizons for Non-Stationary Prediction via Online Aggregation
- HELM: Harness-Enhanced Long-horizon Memory for Vision-Language-Action Manipulation
- HELM - Hierarchy via Edge Learning and MST
- HEP-JEPA2: Competitive JEPA Pretraining for Jets
- Hessian-based Temperature for Knowledge Distillation
- Hessian-Dependent Sample Complexity in Zeroth-Order Stochastic Optimization: Suboptimality of Convex-Support Sampling and Optimal Sample Complexity
- HeST-LLM: Bridging Heterogeneous Temporal--Semantic Misalignment for Multivariate Long-term Time Series Forecasting
- HeteroAML: A Heterogeneous Benchmark for Realistic Single-Bank Anti-Money Laundering
- HeteroFuse: Missingness-Aware Multimodal Fusion for Robust Battery Prognostics under Dataset Heterogeneity
- Heterogeneity in Trajectories of Trust: Inferring and Clustering Dynamic Human–AI Reliance
- Heterogeneous Graph Reinforcement Learning for Lifted Multicut Problem
- Heterogeneous Scientific Foundation Model Collaboration
- Heterophily is a Data Problem, Not a Modeling One: Three Pillars, Path-Length Geometry, and the Case for ADR as an Optimization Objective
- Heteroscedastic Variational Last Layers
- Heuristics are Not Enough: Calibrated and Corrected Caching for Video Diffusion Acceleration
- HexiSeq: Accommodating Long Context Training of LLMs over Heterogeneous Hardware
- HGWM: Hierarchical Graph-guided World Model for Zero-shot Object Navigation via Graph Matching
- HIARR: Hierarchical Information Alignment and Reasoning-Intensive Retrieval for Scalable MCP Tool Selection
- Hidden in Memory: Sleeper Memory Poisoning in LLM Agents
- Hidden in Plain Sight: Benchmarking Agent Safety Against Decomposition Attacks with DECOMPBENCH: A Closer Look
- Hidden in Plain Sight: Improving Industrial Anomaly Detection via Camouflaged Object Detection Techniques
- HiddenMark: Analytic Hidden-State Watermarking with Closed-Form Guarantees
- Hidden Measurement Error in LLM Pipelines Distorts Annotation, Evaluation, and Benchmarking
- Hidden physics: geometric similarity, sparse observations and universal physics engine
- Hidden-State Safety Monitoring for Frozen Neural Networks
- Hidden States as States: Discretizing Hidden-State Geometry in Language Model Computation
- Hidden Tails: Certifying Tail-Risk Claims under Selective Labels
- Hidden Unimodality: Training-Time Modality Imbalance Induces Directional Modality Collapse
- Hide to Guide: Learning via Semantic Masking
- Hide to See: Reasoning-prefix Masking for Visual-anchored Thinking in VLM Distillation
- Hiding in Plain Sight: Detectability-Aware Antidistillation of Reasoning Models
- HIDRA: Hierarchical Dual-Routing Attention for Replay-Free Lifelong Imitation Learning
- HiDrive: A Closed-Loop Benchmark for High-Level Autonomous Driving
- Hierarchical Adaptive Frame Sampling For Video Understanding
- Hierarchical Agglomerative Clustering via Relaxed Representatives
- Hierarchical Concept Geometry in Language Representations Emerges from Word Co-occurrence
- Hierarchical Concept-Guided Tumor-Clone Multiple Instance Learning
- Hierarchical Conformal Classification
- Hierarchical diffusion for camera-controlled autoregressive video generation
- Hierarchical Domain Generalization
- Hierarchical Graph Alignment for Cross-Modal 3D Scene Grounding
- Hierarchical Graph Representation Learning with Pooling-Induced Substructures
- Hierarchical Lateral Inhibition Enables Robust Winner-Take-All Computation in Large-Scale Cortical Models
- Hierarchical manifold disentangling in the human ventral stream supports adversarially robust visual inference
- Hierarchical Verified Knowledge Curation for Multi-Agent Reasoning: A Closer Look
- Hierarchy of Discriminative Power and Complexity in Learning Quantum Ensembles
- HIERGROUND: Hierarchical Report-Guided Grounding in Multi-View Clinical MRI.
- HIFC-IQA: Train-Free Cross-Domain Image Quality Assessment via Dual-Process Cognition
- High-arity Sample Compression
- HighBuild-1M: A Multi-continental High-resolution Benchmark Dataset for Single-view Building Height Estimation
- High-Dimensional Dynamics of Mixture-of-Experts: Phase Transitions, Routing, and Specialization
- High-Dimensional Learning Dynamics of Attention-Indexed Models
- High-Dimensional Sparse Regression via Smooth Selective Regularization: Sharp Non-Asymptotic Guarantees
- Higher-order Persistence Diagrams
- High-Fidelity Boltzmann Samplingvia Physical Prior Lifted Continuous GFlowNets
- High Inter-Annotator Agreement Is Not Evidence of Accountability in AI-Assisted Annotation Pipelines
- Highly Data Parallelizable Estimation of the Sliced-Wasserstein Distance Using Cumulative Distribution Functions
- High-probability Convergence of Gradient Methods under Markovian Stochasticity
- High-Throughput Thompson Sampling for Large-Budget Bayesian Optimization
- Hilbert-Anchored Locality for Off-Policy Replay
- HiLL: LLM-Guided Recursive Semantic Refinement for Reference-Free Cell Type Annotation
- Hindi as a Stress Test: Probing Cross-Lingual Adaptation in Vision-Language Models
- Hindsight is Insight: Refining Temporal Structures for Online Temporal Action Segmentation
- Hi-Q: Hierarchical Evidence-guided Query Refinement for Multi-Hop Question Answering
- HIRaD: Hidden Interaction Inference from Predictive Radar Dynamics
- HiTokSR: A Coarse-to-Fine Tokenizer with Hierarchical Codebooks for High-Fidelity Real-World Image Super-Resolution
- HIVE: Hierarchical Induction of Visual Embeddings for Part–Whole Representation Learning
- HM-World: Evaluating Hybrid Memory in Dynamic Video World Models
- Hodge Decomposition of Diffusion Score Fields as a Failure-Mode Diagnostic (Extended)
- HoliPlace: Holistic Macro Placement via Global Planning and Entropy-Guided Ordering
- HoloGene: Learning to Lift Sliced Spatial Transcriptomics to Holistic 3D Gene Fields
- HOMIE: Human-object Centric Video Personalization via Multimodal Intelligent Enhancement
- Homogenization of $\ell_2$-Adversarial Training in High Dimensions: Exact Dynamics under Stochastic Gradient Descent
- HOPE: Hand-Object Pressure Estimation from Monocular Videos
- HoReN: Normalized Hopfield Retrieval for Large-Scale Sequential Model Editing
- HorizonComposer: Spatiotemporally Consistent Driving Video Editing with Enriched Traffic Semantics
- HoSA: Homotopy-Guided Simplex Optimization for Adversarial Suffix Attacks
- HouseDiffusion++: Constraint-Consistent Vector Floorplan Generation
- How Alignment Routes: Localizing, Scaling, and Controlling Policy Circuits in Language Models
- How Complete Should a Reference Be? A Benchmark Audit for Fluorescence Spot Detection
- How Data Imbalance Shapes Optimization in Deep Cognitive Diagnosis
- How Data Scales in Agentic Reinforcement Learning: Laws and Synthesis Strategies
- How Do Electrocardiogram Models Scale?
- How Does Personalized Memory Shape LLM Behavior? Benchmarking Rational Preference Utilization in Personalized Assistants
- How Do Language Models Compose Functions?
- How Do Multi-Agent Memory Systems Fail Under Regulatory Pressure? An Evaluation Framework and Benchmark
- How do Small Transformer Models Learn Hard Math Tasks?
- How Faithful Is Trajectory-Based Data Attribution? Error Sources, Remedies, and Practical Guidelines
- How Far Do Simple Transformations Translate Across Text Embedding Models?
- How Far Is Too Far? Object Recognition Declines Monotonically with Semantic Distance
- How Fine-Tuning Objectives Shape Layer-Wise Information in LLM Hallucination Detection
- How I learned to stop worrying and love StopGrads: Stationarity, Convergence, and a case study on Flow Map Learning
- How Input Encoding Determines Performance in Spiking Reservoir Computing
- How Language Models Compress and Compare: Understanding Selection with Token Covariance Maps
- How Many Iterations to Jailbreak? Dynamic Budget Allocation for Multi-Turn LLM Evaluation
- How Many Scientists? Epistemic Diversity in Expanding-Frontier Discovery
- How Mobile World Model Guides GUI Agents?
- How Much Can LLMs Hallucinate? An Upper Bound via Coupling and Ambiguity
- How much learnable information is shared across tasks?
- How Much Neural Computation Must Be Serial? A Case Study in Recurrent Models
- How Much of a Model Do We Need? Redundancy and Slimmability in Remote Sensing Foundation Models
- How Mutual Distillation Improves Generalization
- How, Not Just What or Where: Differential Kinematics on the Grassmannian for Unsupervised HOI Segmentation
- How Open Must Language Models be to Enable Reliable Scientific Inference?
- How Post-Training Shapes Biological Reasoning Models
- How Reasoning Breaks: A Variational Free Energy Perspective on Chain-of-Thought Failures
- How Sampling Shapes LLM Alignment: From One-Shot Optima to Iterative Dynamics
- How Scaling VLMs Trades Cognitive Biases
- How Selection Shapes Diversity in LLM Ecosystems
- How Server Feedback Shapes Regret in Online Mixture Optimization Over a Fixed Convex Dictionary
- How Should LLMs Listen While Speaking? A Study of User Stream Routing in Full-Duplex Spoken Dialogue
- How to Adapt SAM3 as a Strong Text-Prompted 3D Medical Segmentor?
- How to Build Long-Horizon Web Agents? (Effectively)
- How to make the most of your masked language model for protein engineering
- How to Preserve Effective Antigen-Conditioned Antibody Generation under Side-Chain Perturbations?
- How Well Can Models Follow Visual Instructions? VIBE: A Benchmark for Visual Instruction-Guided Image Editing
- HPC-Bench: A Comprehensive Benchmark for High Performance Computing Codes
- HRIL: Isolating Multimodal Synergy via Higher-Order Dependence
- HTJU: A Homotopy Trajectory-Unified Method for Deep Neural Network Training
- Hub for Outliers, Spokes for Inliers: Uniform Latent Space Construction for Dual-Mismatched Semi-Supervised Learning
- Hubs, Hedges, and Noise: A Spectral Framework for Sparse Representations in Large Language Models
- Human-AI Coevolution: Measuring Human-Agent Teams in the Agentic Era
- Human-AI Collaboration Needs Synchronization
- Human Creative Improvisation Is a Reverse Discrete Diffusion Process
- Human Decision-Making with AI Assistance under Correlated Features
- Human-Inspired, Task-Dimension-Guided Exploration for Efficient Learning and Transfer in High Dimensions
- HumanMoveVQA: Can Video MLLMs reason about human trajectory from videos?
- Humans Bend, LLMs Snap: Sharper Phase Transitions in Language Model Robustness to Character-Level Noise
- HumanStereo: A Benchmark for Metric Facial Depth Estimation and Its Evaluation
- Hunt Globally: Evaluating Search Completeness in Multilingual Drug Asset Scouting
- Hybrid Methods for Robust Tabular Data Imputation
- Hybrid XGBoost-PPO Framework for Battery Energy Storage Arbitrage Under Electricity Price Uncertainty
- HyCOP: Hybrid Composition Operators for Interpretable Learning of PDEs
- HyFAD: Hybrid Time-Frequency Diffusion with Frequency-Aware Embedding for Time Series Imputation
- HyPerAlign: Interpretable Personalized LLM Alignment via Hypothesis Generation
- Hyperbolic Concept Bottleneck Models
- HyperbolicGPT: Native Hyperbolic Language Modeling Improves Tail Predictions and Generation Quality
- Hyperbolic Gravitational Cone: A Physically-Inspired Approach for Partial Order Embeddings
- Hyperbolic Large Concept Models: Geometry-Aware Hierarchical Reasoning Beyond Euclidean Space
- Hyper Dynamic Convolution
- HyperGAM: A Generalized Additive Model for Hypergraphs with Closed-Form Distance-Level Attribution
- Hypergradient-based Bilevel Reinforcement Learning with Improved Sample Complexity
- Hypergraph Generation via Structured Stochastic Diffusion
- Hypergraph Generation with Latent Diffusion
- HyperMAS: Evolving Collaboration Primitives for Scalable Multi-Agent LLM Collaboration
- Hyperparameter Optimisation of Convex Portfolio Trajectories Using Large Language Models
- Hyperparameter Transfer for Dense Associative Memories
- HyperRestormer: Hierarchy-Aware All-in-One Image Restoration with Hyperbolic and Texture Priors
- HyperSkill: Training-Free Omnimodal GRPO via Hypergraph-Indexed Skill-Library Evolution
- Hyperspherical Autoencoder for High-Fidelity Image Reconstruction and Generation
- Hyperspherical Diffusion Consolidation for Continual Learning in Spiking Neural Networks
- HyperSym: Causal Verifier Feedback for Budgeted Neuro-Symbolic Algorithm Discovery
- HyperTree: Scalable Unsupervised Hierarchy Discovery in Hyperbolic Space
- HyperVAttention: Efficient Sparse Attention with Spatio-Temporal Clustering for Video Diffusion
- HypHOI: Hyperbolic-Enhanced Zero-Shot HOI Detection with Hierarchical Entailment
- HypMoE-ReID: Hyperspherical Mixture-of-Experts for Large Scale Person Re-Identification
- HYQK: A Hybrid QUBO K-Means Algorithm for Improved Clustering
- HyrCap: Hybrid Rank-Calibration of Action Proposals for Temporal Event Understanding
- HySparse: A Hybrid Sparse Attention Architecture with Oracle Token Selection and KV Cache Sharing
- I2V-DETACH: Source Grounding Detachment for Unauthorized Image-to-Video Generation
- iASIDE: A Framework for Identifiable, Adaptable, and Separable Interventional Dynamics
- I Can’t Believe It’s Not Better (ICBINB): Failure Modes of AI in Biology
- ICAT: Incident-Case–Grounded Adaptive Testing for Physical-Risk Prediction in Embodied World Models
- ICG-Guard: Structural Risk Reasoning via Intent-Constraint Graphs for Multi-Turn LLM Safety
- IDCFIQA: A Dual-Dimension Quality Assessment Benchmark Dataset for Identity-Consistent Face Image Generation
- IDEAL-Bench: Indoor Dataset for Evaluating Analysis by 3D Layout Reasoning
- IdealCache: Rethinking Cache Scheduling in Diffusion Transformers via Ideal Trajectories
- Ideas Have Genomes: A Comprehensive Benchmark for Scientific Lineage Reasoning and Lineage-Grounded Idea Generation
- Idempotency Exposes Consistency Problems in Sparse Autoencoders
- Identifiability of Real Concept Drift under Partial Observation
- Identifying LLM Hallucinations in Handwritten Exam Transcriptions using Logits
- Identifying Multiple Root Causes under Mean-Shift Interventions: Identifiability and Algorithm
- Identity-Prior Attribute Learning for Text-Based Person Retrieval
- IDOL: Inverse-Dynamics-Guided Future Prediction for End-to-End Autonomous Driving
- IDPruner: Harmonizing Importance and Diversity in Visual Token Pruning for MLLMs
- IFDECORATOR: Wrapping Instruction Following Reinforcement Learning with Verifiable Rewards
- IgGM2: An All-Atom Foundation Model for Adaptive Immune Receptor Design
- Ignored but Predictive: Projection-Head Null Spaces in Contrastive Learning under Spurious Correlations
- IGT-OMD
- I Have a Stream: Making Self-Supervised Learning Work on Continuous Video
- I-JEPAv2: Exploiting Emerging Properties in Joint Embedding Predictive Vision Transformers
- Image2Sim: Scaling Embodied Navigation via Generative Neural Simulator
- Image Backdoor Attack Detection and Trustworthy Aggregation for Federated Learning
- Image-Based Direct Prompting Attacks on AI Agents
- Image Keys and Values are Asymmetric in MLLM Text-to-Image Attention
- Image-to-Equation is the Missing Modality of Quantitative AI for Science
- Imbalanced Graph Classification via Graph Similarity Learning
- IMD-Blur: Implicit Motion and Defocus Blur Augmentation for Image Deblurring
- Immunogenicity Is Not a Binding Problem
- Impacts of Aggregation on Model Diversity and Consumer Utility
- Implementation Matters in Measuring Chain-of-Thought Monitorability
- Implicit Bias of Gradient Descent under Spurious Correlations
- Implicit Choices, Explicit Failures: Auditing EEG Foundation-Model Transfer to Passive Wearable Clinical Settings
- Implicit-Euler Value Iteration: Long-Horizon Planning via Stable Integration of the Bellman Residual Flow
- Implicit Graphon Neural Representations of Molecular Structure and Dynamics
- Implicit Kemeny Alignment: Resolving Non-Transitive Preferences without Inference-Time Aggregation
- Implicit Koopman Spectral Bounding in Transformers
- Implicit Reward Alignment For Training Causally-Coherent Tabular Data Generators
- Importance-Weighted Operator Learning Under Probability Measure Shifts
- Importance Weighting for Unlabeled-unlabeled Learning under Distribution Shift
- Impossibility of Distribution-Free Predictive Inference for Individual Treatment Effects
- IMPRINT: Rethinking Retrieval for Long-Term Conversational Agent Memory
- Improved Approximation Algorithms for Overlapping Clustering
- Improved Belief-Attention In Vision Tasks
- Improved Distributional Diffusion Models
- Improved Leverage Score Sampling for Constrained Active Linear Regression
- Improved Regret and Violation Guarantees for CMDPs with Stochastic and Adversarial Constraints
- Improved Sample Complexity of Learning Contextual Value Distributions
- Improved State Mixing in Higher-order and Block Diagonal Linear Recurrent Networks
- Improved techniques for fine-tuning flow models via adjoint matching: a deterministic control pipeline
- Improved Weakly Supervised Semantic Segmentation with A Relationally Optimized Prototype Memory Bank Framework
- Improving Chain-of-Thought Reasoning in RSVQA via Step-wise Direct Preference Optimization
- Improving Conditional Modeling via Inter-Class Likelihood-Ratio Maximization and Unifying Classifier-Free Guidance with Alignment Objectives
- Improving constraint-based discovery with robust propagation and LLM priors
- Improving Continual Video Instance Segmentation via Spatial-Temporal Balanced Mixture-of-Experts Adapters
- Improving Fine-Grained Control via Aggregation of Multiple Diffusion Models
- Improving Flexible Image Tokenizers for Autoregressive Image Generation
- Improving Full Waveform Inversion in Large Model Era
- Improving Generative Model Self-training with Geometrically Modified Outputs
- Improving LLM Perplexity via Adaptive Predictions
- Improving Minimax Estimation Rates for Contaminated Mixture of Multinomial Logistic Experts via Expert Heterogeneity
- Improving Neural Processes in the Low-Data Regime via Context-Subset Training and Self-Distillation
- Improving Out-of-Distribution Robustness in Humanoid Motion Imitation via Sparse Keyframe Supervision
- Improving Quantized Zeroth-Order Optimization through Reconstructed Low-Rank Structures
- Improving Recommendation Fairness without Sensitive Attributes Using Multi-Persona LLMs
- Improving Sample Diversity in Autoregressive Text-to-Image Generation via Cluster Truncation
- Improving the Optimization Landscape of Matrix Completion with $\epsilon$-close Surrogates
- Improving the Robustness/Accuracy Tradeoff Against Adversarial Attacks Using Information Bottleneck Distillation Through Dual Teachers
- Improvisational Reasoning in Vision-Language Models for Grounded Procedural Planning
- IMTS-Tokenizer: Time-Aware Tokenization for Irregular Multivariate Time Series Forecasting
- Incentive Aware AI Regulations: A Credal Characterisation
- In-context learning as a mitigation against catastrophic forgetting
- In-Context Learning Can Help Vision Language Models Overcome Training Prior
- In-Context Learning Meets RL Post-Training: A Directional Duality and Its Algorithmic Consequences
- In-Context Operator Learning for Robust PDE Solving
- In-Context Pure Exploration in Continuous Decision Spaces
- Incorporating Causal Structure and Direction into Data
- Incorporating Neural Network Structure in the Bayesian Learning Rule
- Increasing Computation Resolves Conflicts in Vision Language Models
- Incremental Multiple Oracle
- Indications of Belief-Guided Agency and Meta-Cognitive Monitoring in Large Language Models
- Indirect Intervention in Responsive Networks: Relief Certificates and Simulator-Supervised Audits
- Individuals Matter: Improving Deep Multi-View Clustering via Explicit Single-View Enhancement
- InduceKV: Fixed-Footprint Continual Adaptation of Multimodal LLMs via Inducing KV Memories
- Inducing Symbolic Planning Domains via Language- and Vision-Grounded Predicate Discovery
- Inductive Biases in Clinical Severity Prediction: Comparing Temporal, Structural, and Feature-Based AI Models on a Parkinson's Disease Cohort
- Inductive Deductive Synthesis: Enabling AI to Generate Formally Verified Systems
- IndusAgent: Reinforcing Open-Vocabulary Industrial Anomaly Detection with Agentic Tools
- Inertia-1: An Open Exploration of Wearable Motion Foundation Models
- InEX: Intrinsic and Extrinsic Evidence for Trustworthy Evaluation of Latently Verifiable Tasks
- Inference and estimation with unidentifiable latent treatment effects
- Inference and Uncertainty Quantification for Streaming $r$-PCA
- Inference for Newton Methods with Accelerated Sketch-and-Project via Random Scaling
- Inference-Time Guardrails: Reducing Hallucinations via Structured Invariants
- InferMatrix: Reproducible Benchmark, Verifier Protocol, and Reference Search Policies for LLM Serving Deployment
- Inferotemporal Cortex Collaborates Before It Codes: Non-Serial Inter-Area Synergy in the Macaque Ventral Stream
- Inferring response times of perceptual decisions with Poisson variational autoencoders
- INFILL: An Open-Vocabulary 3D Asset Database with Closed-Loop Expansion for Social Navigation
- InfoDP-LoRA: Heterogeneous Differentially Private LoRA via Information Bottleneck Regularization
- Information-Aligned Preference Optimization for Direct Preference Learning
- Information-Directed Offline-to-Online Reinforcement Learning
- Information Flow and Market Risk: Transformer-Based Evidence on Sentiment and Volatility Dynamics
- Information-Risk Bridges in Neural Credal Prediction
- Information-Theoretic Generalization for Set-Input Optimization-Valued Objectives
- Information-Time Proximal Policy Optimization
- Informative Data Reweighting for Image Classification
- Informative Model Validation for Anomaly Detection
- Informed Plastic Attractors for Adaptive Few-Shot Learning
- Informed Posterior Sampling: More Efficient Online Learning with Few Offline Demonstrations in Average-Reward MDPs
- IN-GSW: Identifiability-Restricted Gram-Schmidt Walk for Efficient Nonlinear Covariate Balancing in High-Dimensional Randomized Experiments
- Inherited-Support Control in Fixed-Evidence Adjudication: An Externally Replicated Comparator Study
- Initializing Multi-Objective Evolutionary Algorithms with the Optima of Each Objective
- InjectRBP: Steering Large Language Model Reasoning Behavior via Pattern Injection
- Inline Memory Meets Reusable Skills: Efficient Adaptation for Vision-Language-Action Model
- Inpainting physics: self-supervised learning for context-driven fluid simulation
- In-Parameter Learning: Why Lifelong AI Systems Need More Than Longer Context
- InQ-Attention: Toward Efficient On-Device LLM Inference via Incremental Quantized Attention
- InsHuman: Towards Natural and Identity-Preserving Human Insertion
- Inside the Address Book: Probing the Sparse Semantic Structure of LLM Key-Value Caches via Sparse Autoencoders
- Insights Generator: Systematic Corpus-Level Trace Diagnostics for LLM Agents
- InSpect: A Curated Natural History Collection Dataset for Insect Specimen Understanding
- INSPO : Unlocking Intrinsic Self-Reflection for LLM Preference Optimization
- Instance-Dependent Auto-Exploration for Tabular Online Discounted Reinforcement Learning
- Instance-Grounded Captions for Explainable Video Anomaly Detection
- InstantCarver: High-Fidelity 3D Geometry Enhancement within Seconds
- Instruction Anchor: Dissecting the Mechanistic Dynamics of Modality Arbitration
- Instruction-Evidence Contrastive Dual-Stream Decoding for Grounded Vision-Language Reasoning
- Instrumental Choices: Measuring the Propensity of LLM Agents to Pursue Instrumental Behaviors
- Integrating GCNII with KAN-Inspired Spline Activations for Nonlinear Learning and Efficient Graph Convolution
- Integrating Generative and Experimental Platforms for Biomolecular Design (GEM)
- IntegrityBench: Can LLMs Be Trusted as Co-Scientists? A Research Integrity Benchmark
- Intelligent Packet Tracer: A Hybrid Rule-Statistical Framework for Evidence-Grounded Network Fault Diagnosis
- Intend, Reflect, Refine: An Adaptive Multimodal Reflection Framework for Autonomous Driving
- Intent2Tx: Benchmarking LLMs for Translating Natural Language Intents into Ethereum Transactions
- Intent-Aware Re-ranking Agents On Mobile Device: Towards High-Fidelity Personalization with LLM Alignment and Compression
- INTENT-Edit-Bench: Benchmarking Scene-Level Intent Realization in Image Editing
- Intent Inference Produces the Retaliatory Cascades It Prevents
- Intentional Deception as Controllable Capability in LLM Agents
- Intention-Driven VLA for Task-oriented Dynamics Inference: from Reactive to Deductive Control
- IntentionNav: A Benchmark for Intent-Driven Object Navigation from Implicit Human Instruction
- Interaction-Aligned Robot Learning from Human Videos with Structured Graph Modeling
- Interaction Collapses Safety: On the Limits of Jailbreak-Robust Generation
- Interaction-Grounded Learning for Markov Decision Processes with Personalized Feedback
- Interactive Navigation for Open-Vocabulary Object Goals in Cluttered Environments
- Interactive Online Temporal Action Segmentation: Test-Time Learning from User Corrections
- Interference-Aware Multi-Task Unlearning
- Interleaving Compression and Training with INTRIM
- InterLV-Search: Benchmarking Interleaved Multimodal Agentic Search
- Intermediate Representations are Strong AI-Generated Image Detectors
- Internal Data Repetition Destroys Language Models
- Internal Tree Search Execution in Transformers
- Interpretability as a Science: Toward Rigorous Foundations for Understanding LLMs
- Interpretability for Discovery: Understanding and Discovering Novel Knowledge in AI Models
- Interpretable Client Contribution Evaluation in Federated Learning via Partial Information Decomposition
- Interpretable Discriminative Text Representations via Agreement and Label Disentanglement
- Interpretable Generalized Partially Linear Single-Index Models
- Interpretable Hierarchical Concept Reasoning through Graph Learning
- Interpretable Sparse-to-Sparse Feature Learning
- Interpretable Transformers by Condition Guided Self-Attention
- Interpreting Agent Behavior (IAB): Human-Centered Interpretation for Understanding Agents, Humans, and Interaction
- Interpreting Chain-of-thought Reasoning via Partial Information Decomposition
- Interpreting Neural Combinatorial Optimization via Evolving Programmatic Bottlenecks
- Interruptible Agent Planning: Matching Calibration Bounds for LLM Agent Routing
- InterTab: Interleaved Visual-Structure Alignment for Multi-Modal Table Reasoning
- Interval-Width Drift: Predictor-Coupled Shift Detection for Imputation-Augmented Predictors
- Intervening in Concept Bottleneck Model via Causal Effects from Pixel to Concept
- Intrinsic Flow Matching on Quantum Pure-State Manifolds with Phase-Aligned Transport
- Intrinsic Information Theoretic Analysis of ReLU Nets
- Intrinsic-Preserving Schrödinger Bridge for Direct Part-Aware 3D Generation
- Intrinsic rate limits of learned summaries in neural posterior estimation
- Intrinsic Reward Policy Optimization for Sparse-Reward Environments
- Invariant Hyperbolic Unfolding: Radial Canonicalization for Label-Free Cross-Graph Link Prediction
- Invertible Logits Transformation for Accuracy-Preserving Post-Hoc Uncertainty Calibration
- Inverting the Bioheat Equation: Volumetric Perfusion Recovery from 2D Thermograms via Neural Operators
- Investigation of Scaling Laws for Encoder-Decoder Protein Language Models
- Invisible Tremors: Inducing Action Deviation in VLA Models via Temporal-aware Vision Poisoning
- I-Perceive: A Foundation Model for Vision-Language Active Perception
- IPIBench: Evaluating Interactive Proactive Intelligence of MLLMs under Continuous Streams
- Iris: Empowering Video MLLMs with High-Frequency Pose Priors via Spatiotemporal Binding
- Irminsul: MLA-Native Position-Independent Caching for Agentic LLM Serving
- Is Agentic AI Ready for Real-World Hardware Engineering? A Deep Dive with Phoenix-bench
- Is a Linear Probe Evidence of a Linear Representation?
- Is Class Signal Clustered or Routed in Task-Induced Implicit Neural Representation Weight Spaces?
- Is Complex Training Necessary for Long-Tailed OOD Detection? A Re-think from Feature Geometry
- Is Constraint Diversity overstated? High-Precision Reward Generalizes to Robust Instruction Following
- Is Escalation Worth It? A Decision Theoretic Characterization of LLM Cascades
- Is Gradient Ascent Really Necessary? Memorize to Forget for Machine Unlearning
- Isharah-Selfie: Continuous Sign Language Recognition Dataset for One-handed Signing
- ISIMUD: A Unified Diffusion Model for Time Series Captioning and Generation
- Iso-7: Reliable Scaling Law Prediction with a 7-Parameter Loss Surface Surrogate
- ISO-Bench: Can Coding Agents Optimize Real-World Inference Workloads?
- ISO: Isospectral Optimization for RLVR
- Iso-LoRA: Spectral Isotropy LoRA via Delayed Scale-Invariant Orthogonality
- Is One Score Enough? Rethinking the Evaluation of Sequentially Evolving LLM Memory
- i-STAR: Influence-Shaped Team Adaptation via Reinforcement Learning for Human-AI Coordination
- Is the Importance Ratio Necessary for Stable Reinforcement Learning in LLMs?
- Iterative Amortized Inference: Unifying In-Context Learning and Learned Optimizers
- Iterative Critique-and-Routing Controller for Multi-Agent Systems with Heterogeneous LLMs
- Iterative Identification Closure: Amplifying Causal Identifiability in Linear SEMs
- Iterative Model-Based RL Beyond Global Lipschitz and Global Concentrability
- Iterative Scarcity-Guided Exploration: Bootstrapping Generative Auto-bidding from Narrow Support
- IterSafe: Decoder-Integrated Safety Shaping for Vision-Language-Action Autonomous Driving
- ITNet: A Learnable Integral Transform That Subsumes Convolution, Attention, and Recurrence
- It’s Not a Lottery, It’s a Race: Understanding How Gradient Descent Adapts the Network’s Capacity to the Task
- It’s the Thought That Counts: Stress-testing LLM Reasoning Using Altered Variations of Classical Psychology Paradigms
- It Takes Two: Coalition-Aware Exploration in Cooperative Multi-Agent Reinforcement Learning
- It Takes Two: Your GRPO Is Secretly Doing DPO
- IVPGC-Bench: Intention-driven Visual Perception across Geometry and Charts in Multimodal Large Language Models
- JailBound: A FOL-Guided Jailbreak Evaluation Framework for Revealing Safety Boundaries of LLMs
- Jailbreaking Large Language Models via Rebuttal Attack
- Janus-RL: Decoupling Exploration from Optimization in RLVR
- JAS-GS: Joint Appearance–Segmentation Optimization for 3D Scene Understanding with Gaussian Splatting
- JaxARC: A JAX-native RL environment for the Abstraction and Reasoning Corpus
- JEPA-DNA: Grounding Genomic Foundation Models through Joint-Embedding Predictive Architectures
- JEPA-Reasoner: Decoupling Latent Reasoning from Token Generation
- JEPAWG: Interpretable Hypernetworks for Weight-Space Physics
- Jet-Long: Efficient Long-Context Extension with Dynamic Bifocal RoPE
- JEWEL: O(N) Generator Discovery for Sequential Manifolds, a Scalable Alternative to Kernel PCA
- Jigsaw-CRL: Recovering Global Latent Causal Order from Fragmented Multi-Client Interventions
- JMed48k: A Multi-Profession Japanese Medical Licensing Benchmark for Vision-Language Model Evaluation
- JobBench: Aligning Agent Work With Human Will
- Joint Diffusion for Unconstrained Object Compositing with Learned Edit Budgets
- Joint EM Image Super-Resolution and Segmentation with Semantic and Structural Priors
- Joint implicit shape reconstruction and explicit correspondences from multi-view videos
- Joint Learning of Hierarchical Neural Options and Abstract World Model
- Jointly Reinforcing Diversity and Quality in Language Model Generations
- Joint Optimization of Multi-agent Memory System
- Joint protein, mRNA, DNA sequence design and optimization with nucleotide-level Potts models
- Joint Sequence--Vocabulary Selection for Efficient LLM Distillation
- JointsSpeak: Interpretable Joint-Level Motion Stylization with LLM-Guided Kinematic Priors
- Joint-Wisdom: Intelligent Scaling Arising through Focused Reward-Aware Value Optimization
- JOLT: Uncertainty-Selective Self-Ensembling for Geometric Consistency in Video Generation
- Journey Operators for Structured Multi-Axis Composition
- JPmHC Dynamical Isometry via Orthogonal Hyper-Connections
- JudgeArena: A Unified Framework for Reproducible LLM-Judge Evaluation
- JudgeRM: Distilling Open-Judge Ensembles into Calibrated Process Reward Models
- JudgeSense: A Benchmark for Prompt Sensitivity in LLM-as-a-Judge Systems
- Judgment Boundary Governance for Military AI: A Control Model and Full Benchmark
- Jump Start Your Policy Learning with Lessons from 145,000 Training Runs
- JumpTPP: Non-Autoregressive Diffusion Modeling for Multivariate Temporal Point Processes
- Justitia: Fair and Efficient Scheduling of Task-parallel LLM Agents with Selective Pampering
- KAIROS: Query-Aware Multimodal Trajectory Budgeting for Long-Video Reasoning
- KamonBench: A Grammar-Based Dataset for Evaluating Compositional Factor Recovery in Vision-Language Models
- KC-3DGS: Kurtosis-Constrained Gaussian Splatting for High-Fidelity View Synthesis
- KernelBrain: Coarse-to-Fine, Budget-Aware Search for Agentic GPU Kernel Optimization
- KernelDNA: Cross-Layer Kernel Sharing via Decoupled Neural Adapters
- KernelGate: Joint Weight Pruning and Activation Gating for Hardware-Efficient DNN Inference
- Kernel Granger Component Analysis for Nonlinear Directed Component Discovery
- Kernelized Affine Fisher Estimation (KAFE) for Natural Policy Gradients with Convergence Guarantees
- Kernel Selection is Model Selection: A Unified Complexity-Penalised Approach for MMD Two-Sample Tests
- KerONet: A Single Softmax Readout Suffices for Physics-Informed Operator Learning
- KeyframeFace: Language-Driven Facial Animation via Semantic Keyframes
- KG-Band: LLM-Guided Spectral Interface Learning for Model-Compatible Hyperspectral Representation
- KG-Guard: Graph-Based Hallucination Detection for Knowledge Base Question Answering
- KGHEdit: Knowledge-Graph-Guided Continual Model Editing via Heat Diffusion
- KGPFN: Unlocking the Potential of Knowledge Graph Foundation Model via In-Context Learning
- KG-QuerySeg: Reliability-Calibrated Anatomical Spatial Queries for 3D Lesion Medical Image Segmentation
- Killing the Pixels: Latent Imagination for Compositional Embeddings
- K-Myriad: Jump-starting reinforcement learning with unsupervised parallel agents
- KNN Implementation Details Can Dramatically Change Performance: An Example from Cover Trees
- Knocking-Heads Attention: Drop-in Shared Projections for Cross-Head Coordination
- Knot Forcing: Taming Autoregressive Video Diffusion Models for Real-time Infinite Portrait Animation
- Knot Polynomial Spaces: A Benchmark for Rare-Event Evaluation in AI for Math
- Know Before You Route: Dual-Signal Expert Prefetching with Conformal Coverage Guarantees for Sparse MoE LLMs
- Knowing When to Ask: Segment-Level Credit Assignment for LLM Tool Use
- Knowing When to \texttt{STOP}, \texttt{RECOVER}, and \texttt{SEARCH} for Reliable GUI Automation
- Knowledge-Enhanced Multimodal Alignment for Robust Reasoning Under Missing Modalities
- Knowledge-Level Consistency Reinforcement Learning: Dual-Fact Alignment for Long-Form Factuality
- Knowledge Localization in Mixture-of-Experts LLMs Using Cross-Lingual Inconsistency
- Knowledge or Reasoning: Do Large Language Models Truly Acquire Graph Reasoning Capabilities?
- Knowledge Propagation Across Embodied Experience
- Knowledge Transfer Scaling Laws for 3D Medical Imaging
- Know Thyself: Concept-Level Diagnosis for Financial Large Language Models
- KoFinSQL: A Deep Agent-Based Text-to-SQL Framework for Korean Financial Domains
- Koopman–Kalman State Estimation under Missing Observations: A Task-Sufficient and Cost-Aware Formulation
- Koopman Operator Bounds for Time-Series Joint-Embedding Predictive Architectures
- Kosmoi: Counterfactual World Modeling for Action-Aligned Autonomous Driving
- K-PWM: Control-Oriented Structured World Models under Partial Observation
- KS Assistant: A Simple General-Purpose AI Agent for Software Engineering
- Kurtosis-Guided Denoising Score Matching for Tabular Anomaly Detection
- KVBuffer: IO-aware Serving for Linear Attention
- KVEraser: Learning to Steer KV Cache for Efficient Localized Context Erasing
- L2R: Low-Rank and Lipschitz-Controlled Routing for Mixture-of-Experts
- LABSHIELD: A Multimodal Benchmark for Safety-Critical Reasoning and Planning in Scientific Laboratories
- LACE: Latent Alignment via Counterfactual Embeddings
- LACE: Lattice Attention for Cross-thread Exploration
- LACE: Lightweight Attribution-guided Concept Evolution for Continual Learning
- LAGO: Language-Guided Adaptive Object-Region Focus for Zero-Shot Visual-Text Alignment
- LakeAgents++: An LLM-Based Multi-Agent Framework for Tabular Data Augmentation
- Laminar Flow Matching: Geometry-Aware Multiscale Decoupling for Time-Series Forecasting
- LandlordBench: Evaluating the Legal Compliance of LLMs in Real Estate Disputes
- LandmarkMerge: Resolving Task Conflicts at Their Real Neural Interfaces
- LangBoltz: All-Atom Protein Ensemble Generation via Boltzmann-Aligned Protein Language Models
- Lang-SVG: Hierarchical Image Vectorization with Language Priors
- Language-aligned Visual Geometry Transformer
- Language as Control: A Unified Information-Theoretic Model of Trust, Cooperation, and Antagonism
- Language-Critique Imitation Learning from Suboptimal Demonstrations
- Language Models can Learn High-Capacity Secure Steganography
- Language Models Coupled with Metacognition Can Outperform Reasoning Models
- Language Model Shape Shapes Behavior
- Language models struggle with compartmentalization
- Language Models Without a Trainable Input Embedding Table: Learning from Fixed Minimal Binary Token Codes
- Language Unalignability: Why Some Concepts Resist Cross-Cultural Benchmark Evaluation
- LangZip: Lightweight Language Feature Compression for 3D Gaussian Splatting
- Laplace-Bridged Smoothing for Accelerated Randomized Smoothing Certification
- LAPLEX: The FFT of Learnable Laplace Kernels
- LAQuant: A Simple Overhead-free Large Reasoning Model Quantization by Layer-wise Lookahead Loss
- LaRA-WM: Test-Time Latent Action Refinement for Robotic Manipulation in Dynamic Environments
- Large Language Model Failures from Hallucination to Homogenization Are Different Facets of Miscalibration
- Large Language Models Approximate Theory of Mind But Lack Its Bayesian Structure
- Large Language Models as Amortized Pareto-Front Generators for Constrained Bi-Objective Convex Optimization
- Large language models can not and should not be banned from peer review
- Large Language Models Must Be Governed as Sociotechnical Attack Surfaces: The Case for Human-Adversarial Threat Modelling
- LARGER: Lexically Anchored Repository Graph Exploration and Retrieval
- Large-scale Score-based Variational Posterior Inference for Bayesian Deep Neural Networks
- LARGO: Low-Rank Hypernetwork for Handling Missing Modalities
- LaSA-Net: A Language-Guided Network for Outdoor Generalized 3D Referring Expression Segmentation
- LASER: Latent Space Adjoint Matching for Support Constrained Entropy Regularized Offline RL (Extended)
- LASS-ODE: Learning Scalable Representations Across Diverse Dynamical Systems
- Last-Iterate Analyses of FTRL with the 1/2-Tsallis Entropy in Stochastic Bandits
- LatencyBench: Evaluating LLM Robustness Under Delayed Information Availability
- Latent Algebraic Reasoning Operators
- Latent Algorithmic Structure Precedes Grokking: A Mechanistic Study of ReLU MLPs on Modular Arithmetic
- Latent-Augmented Discrete Diffusion Models
- Latent Bridge: Feature Delta Prediction for Efficient Dual-System Vision-Language-Action Model Inference
- Latent Corr: Monitoring Training Dynamics via Mini-Batch Representation Dependence
- LatentFold: Continuous Evolutionary Manifold Learning for Meaningful Pseudo-MSA Generation
- Latent Imagination Thinking (LIT): Beyond recursive models in the observation space
- Latent Information Sharing for Accelerating Federated Learning
- Latent Introspection: Models Can Detect Prior Concept Injections
- LatentMD: Benchmarking Markdown Boundary Failures in LLM-Generated Text
- Latent Phase Transitions as Signatures of Neural Reasoning Failure
- LatentPhyGS: Dynamic Physics-Grounded 3DGS with Material-Aware Latents
- Latent Poincar\'e Shaping for Agentic Reinforcement Learning
- LatentRec: Implicit Reasoning Internalization for LLM-Based Recommendation
- Latent Spatial Reasoning: Building Innate 3D Awareness via Latent-Space Distillation
- LatentTool: From Explicit Tool Calls to Implicit Latent Reasoning for MLLMs
- LatentUMM: Dual Latent Alignment for Unified Multimodal Models: A Closer Look
- Latent Visual Cache for Video Reasoning
- Late-Sequence Shortcuts Selectively Disrupt Sequential Architectures: A Controlled Study of Distributional Robustness in Learned Reasoning
- LAtte: Hyperbolic Lorentz Attention for Joint-Subject EEG Classification
- Lattice: Learning to Efficiently Compress the Memory
- Layer-Adaptive KV Eviction with Query-Anchored Importance for 1M-Token Inference
- Layer Normalization as Implicit Gain Control: Inducing Low-Frequency Bias in Vision Models
- Layer Precision Reduction for Deep Anomaly Detection
- Layer Skipping As Resampling
- Layerwise Convergence Fingerprints for Runtime Misbehavior Detection in Large Language Models
- Layer-Wise Token Dropping for Efficient Structured-Grid PDE Emulation
- LayoutBridge: Anisotropic Brownian Bridges for Public Indoor Floorplan Generation
- LayoutLattice: Pairwise Attention Biases for Contextual Layout Generation
- Lazy Ensemble for Tree Search under Fixed Compute Budgets
- Lazy Greedy Meets Bandits for Stochastic Submodular Maximization
- LCA-Flow: Optimal Transport to Latent Canonical Attractors for Cross-Subject EEG Emotion Recognition
- LDPCache: Locally Differentially Private Multi-Query Processing with Cache Optimization for Large Language Models
- LDS-Splat:Text-Guided Local Discriminative Smoothing for Referring 3D Gaussian Splatting Segmentation
- Leakage-Audited Distributional Event Studies with Conditional Flow Matching
- Leakage Thresholds for Sandwich Equilibria Under Partial Information
- Lean-GAP: A Dataset of Formalized Graduate Algebra Problems
- Lean Refactor: Multi-Objective Controllable Proof Optimization via Agentic Strategy Search
- LeanSearch v2: Global Premise Retrieval for Lean 4 Theorem Proving
- LeanTree: Expanding the Reasoning Frontier of LLMs via Lean-Guided Subproblem Synthesis
- LEAP: Latent Encoder Alignment for Physics Rewards in Video Generation
- LEAP: Layer-skipping Efficiency via Adaptive Progression for Vision Transformer Distillation
- LEAP: Library-driven Evolutionary Abstraction Paradigm for Large Language Models
- Learnable Diffusion-based Positional Encodings for Link Prediction
- Learnable Structured Noise Injection for Representation Stabilization in Low-Label Regimes
- Learned Dynamics Enable Optimality Regularizers for Data-Efficient Offline Inverse Constrained RL
- Learned Geospatial Priors for Malaria Forecasting
- Learned Relay Representations for Forward-Thinking Discrete Diffusion Models
- Learn from your own latents and not from tokens: A sample-complexity theory
- Learning a definition from data
- Learning and Dynamically Deploying Ordinal Event Schemas for Sequence Working Memory in RNNs
- Learning and Planning on Spatial Networks: A Topology-Driven Framework for Multi-Aircraft Taxiing
- Learning Annotator Behavior for Cost-Aware Active Learning
- Learning a Task-Adaptive Low-Dimensional Semantic Space for Improved Visual Classification
- Learning-Augmented Coordination Mechanisms
- Learning Augmented Exact Exponential Algorithms
- Learning Better Certified Models from Empirically-Robust Teachers
- Learning Better Predictions for Warm-Starting Algorithms
- Learning Bilevel Policies from Demonstrations for Long-Horizon Planning
- Learning Biological Hierarchies in Single-Cell Foundation Models
- Learning Causal Orderings for In-Context Tabular Prediction
- Learning Compact Boolean Networks
- Learning Data-free Universal Adversarial Perturbation with Hybrid Priors and Gradient-Guided Sharpness Regularization
- Learning Decoding Order in Diffusion for Secure Code Generation
- Learning Diffusion Policies with Sublinear Regret
- Learning Dynamic Belief Graphs for Theory-of-mind Reasoning
- Learning Dynamics for Zero-Sum Games with Heterogeneous Agents: An Optimal Transport Approach
- Learning efficient graph topology for stochastic matching
- Learning Event-to-Field Operators Without Interpolation
- Learning Explicit Chain-of-Thoughts can Induce Implicit Thinking
- Learning Faithful Mechanism Subgraphs from Partially Trusted Biological Priors for Genetic Perturbation Prediction
- Learning Fine-Grained Vision-Language Alignment from Discriminative Part Descriptions
- Learning-Forgetting Optimality in Supervised Finetuning: A Cliff Perspective
- Learning from Complexity: Exploring Dynamic Sample Pruning of Spatio-Temporal Training
- Learning From Future: Leveraging Time-Induced Privileged Information for Temporal Prediction
- Learning from Language Feedback via Variational Policy Distillation
- Learning from Unreliable Trajectories: Adversarially-Robust Federated Q-Learning
- Learning Galaxy Kinematics from Spatially Unresolved Spectra
- Learning Gaussian Graphical Models under Total Positivity via Spectral Graph Sparsification
- Learning Halfspaces from Perturbed Contrastive Examples
- Learning How to Cube
- Learning Human-Intention Priors from Large-Scale Human Demonstrations for Robotic Manipulation
- Learning in Context, Guided by Choice: A Reward-Free Paradigm for Reinforcement Learning with Transformers
- Learning Individual Dynamics from Sparse Cross-Sectional Snapshots
- Learning inexact alternating minimization
- Learning Instruction-Following Policies through Open-Ended Instruction Relabeling with Large Language Models
- Learning Invariances for Causal Abstraction Theory
- Learning Large-Scale Competitive Team Behaviors with Mean-Field Interactions
- Learning Maximally Informative Sectioning Patterns in Cumulative Electron Microscopy for Connectomics
- Learning Menu-Based Mechanisms for Truthful Budget-Feasible Procurement
- Learning Modular Addition with Auxiliary Modulus
- Learning Motion-Appearance Coupling Priors for Solving Video Inverse Problems
- Learning Neighbor Scoring Rules for Local Search in Probabilistic Graphical Models
- Learning Networked Dynamical Systems via Neural ODE-based Graph Structure Inference
- Learning Ordered Top-K Targeted Universal Adversarial Perturbations
- Learning Orthogonal Multi-Index Models Beyond Small Initialization: Incremental Learning, Competitive Dynamics and Symmetry
- Learning Paradigms Are Not Distinct and Learning Is a Dynamical Process
- Learning Patient Trajectories from the Electronic Health Records with World Models
- Learning Physics-Constrained Collocation Samplers for Dual-Cone Optimization in Physics Informed Neural Networks
- Learning Planning Budgets in Real-Time RL
- Learning Plug-and-play Memory for Guiding Video Diffusion Models
- Learning Provable Neural Network Observer for Uncertain Dynamical Systems
- Learning Representations for Multivariate Time Series Classification via Context Prediction
- Learning Reusable Options by Decomposing Neural Policies
- Learning Robotic Skills from Offline Preference-based RL
- Learning Robust Normalizing Flows Using Spectral Response Theory
- Learning Robust Representation for Missing-Modality RGB-D Semantic Segmentation
- Learning Robust Representations for Defending White-Box Adversarial Attacks in Continual Learning
- Learning Shared Latent Cellular States for Multi-Assay Prediction
- Learning Sparse Semantic-Cortical Atoms for Multisubject Naturalistic fMRI Encoding
- Learning Spatio-Temporal Foundation Models from Pure Synthetic Data
- Learning Spectral Compositional Koopman Operators for Global-to-Regional Weather Forecasting
- Learning Stable Concept Abstractions for Experimental Planning
- Learning State-Conditioned Feasible Action Manifolds for Physics-based Humanoid Control
- Learning Structured Neural Energies for Nonlinear Constitutive Modeling from Equilibrium Deformations
- Learning Symbolic Scoring Functions for Constrained UAV Coverage Planning
- Learning Task-Centric World Models from Visual Foundations
- Learning Temporal Cryptographic Risk in Post-Quantum Systems Under Uncertain Adversary Timelines
- Learning Temporal Evidence Selection for Video Anomaly Understanding
- Learning the Observation Function in POMDPs: Identifiability through State Consistency
- Learning the Signature of Memorization in Autoregressive Language Models
- Learning Through Experience: Episodic Memory for Long-Horizon Cognitive Agents
- Learning to Act under Noise: Enhancing Agent Robustness via Noisy Environments
- Learning to be Coherent: Data-Driven Discovery of Logical Submanifolds for Artificial Knowledge Systems
- Learning to Bid in Repeated Second-Price Auctions with Dynamic Values and Aggregated Feedback
- Learning to Communicate Under Alignment Uncertainty: A Bayesian Persuasion Bandit Approach
- Learning to Cut: Reinforcement Learning for Benders Decomposition
- Learning to Discover Iterative Spectral Algorithms
- Learning to Discriminate Scene Structures Makes Self-Supervised Depth Learning Scalable
- Learning to Evolve Scenes: Reasoning about Human Activities with Scene Graphs
- Learning to Explore by Predicting What Matters: Entity-Centric Exploration for Visual RL
- Learning to Focus for Person Re-Identification
- Learning to Generalize from the Generative Vicinity: Knowledge Expansion from a Single Teacher
- Learning to Generalize Recursively: Structural Insights from the Tower of Hanoi
- Learning to Generate Formally Verifiable Step-by-Step Logic Reasoning via Structured Formal Intermediaries
- Learning to Generate Multiple Objects from Dense and Occluded Layouts
- Learning to Ground Without Forgetting: Reinforcement Learning for Long-Video Temporal Understanding in Multimodal LLMs
- Learning to Hand Off: Provably Convergent Workflow Learning under Interface Constraints
- Learning to Inject: Automated Prompt Injection via Reinforcement Learning
- Learning to Learn from Multimodal Experience
- Learning to Optimize in Gradient Subspaces
- Learning to Persuade a Biased Receiver
- Learning to play with spikes. Characterizing, predicting, and engineering unsupervised plasticity rules for spiking reservoir computing
- Learning to Pose Problems: Reasoning-Driven and Solver-Adaptive Data Synthesis
- Learning to Rank Spreaders’ Influence via Physics‑Aware Graph Neural Networks (Extended)
- Learning to Route and Schedule LLMs from User Retrials via Contextual Queueing Bandits
- Learning to Test: Physics-Informed Representation for Dynamical Instability Detection
- Learning Transferable Representations from Operating System Entities via Provenance Graph Distillation: A Closer Look
- Learning Universal Motion Primitives for Domain Generalized Trajectory Prediction
- Learning What’s Missing: Failure-Driven Skill Discovery via Predicate Bridges
- Learning What to Predict: Downstream-Guided Task Design for Continued Pretraining
- Learning What to Recommend: Minimax Optimal Simple Regret in Logistic Bandits
- Learning What to Remember: Test-Time Training via Context Distillation
- Learning When to Adapt
- Learning When to Collaborate: Selective Multi-Agent Medical Reasoning via Uncertainty-Aware Routing
- Learning When to Stop: Selective Imitation Learning Under Arbitrary Dynamics Shift
- Learning When to Terminate: Deliberation Cost Strategies for Option-Critic Algorithms
- Learning When to Think: Adaptive Internal Computation for Reinforcement Learning
- Learning When to Trust LLM Priors: A Validated Framework for Semantic Prior Integration
- Learning When to Verify: Reduced-Information Routing for Surgical Safety
- Learning with Locally Private Examples by Inverse Weierstrass Private Stochastic Gradient Descent
- Learning with Structural Inductive Biases: A Semantic-Anchor Distillation Framework for Complex Classification Tasks
- Learn Locally, Recurse Globally: Neural Circuit Synthesis Beyond Training Depth
- Learn to Explore In-Context via Reinforcement Learning
- Learn Where Outcomes Diverge: Efficient VLA RL via Probabilistic Chunk Masking
- Learn where to Click from Yourself: On-Policy Self-Distillation for GUI Grounding
- Learn While Searching An Online Learning Framework for Dynamic Robot Visual Search
- Least but not Last: Fine-tuning Intermediate Principal Components for Better Performance-Forgetting Trade-Offs
- LEAT: LLM-Guided Experience-Aware Adaptive Tuning for MIP Solver Configuration
- LED: Latent Enhanced Detection in Adverse Degradations
- Leech Lattice Vector Quantization for Efficient LLM Compression
- LeFlur: A Biomolecular Design Model with Latent Structure Tokens
- LEGO-MATH: Stacking Reasoning Blocks for Reliable and Harder Math Problem Synthesis
- LensCT: Fine-Grained AI-Involved Text Detection via Temporal-Hierarchical Tomograms of LLM Internals
- LENS: Language-aligned Exploration for Budgeted Token Selection
- LENS: Low-Frequency Eigen Noise Shaping for Efficient Diffusion Sampling
- Lens: Rethinking Training Efficiency for Foundational Text-to-Image Models
- LensVLM: Selective Context Expansion for Compressed Visual Representation of Text
- LERTQA-RAG: Can Structured Validation Make Large Language Models for Reliable Multi-Paper Scientific Reasoning?
- Less Is More in Multi-Agent LLM Systems: The Case for Computation–Reasoning Separation and Structural Provenance Enforcement
- Less is More: Rethinking Pseudo-Labels in Federated Semi-Supervised Learning (Extended)
- Less Language, More Latents: Annotation-Efficient VLAs for Driving
- LessMimic: Versatile Humanoid-Object Interaction with Unified Distance Field Representations
- Less Tuning, Better Planning: Simplifying Offline Model-Based Planning
- Let's ask Gauss: Improved One Run Privacy Auditing
- Let the Target Select for Itself: Data Selection via Target-Aligned Paths
- Letting LLMs Speak for Themselves: Self-Expressive Fine-Tuning
- Leveraging hallucination and biophysics for unified biomolecular seq-structure co-design
- Leveraging LLM Priors for Hierarchical Semantic Disentanglement in Sequential Recommendation
- Leveraging Psychophysical Attentional Distribution for Gaze-Augmented Reward Modeling
- Leveraging Soft Prompts for Privacy Attacks in Federated Prompt Tuning
- Leviathan: Decoupling Input and Output Representations in Language Models
- L-FAME: Longitudinal Focused Attention Meditation EEG Dataset and Benchmark
- LFHE: Local-First Heuristic Evolution for Topology Optimization in Fully Decentralized Learning with Non-IID Data
- Liars' Bench: Evaluating Lie Detectors for Language Models
- LIASFormer: An Innovative Bio-Inspired Spiking Transformer for Feature Enhancement using Lateral Inhibition Attention
- LiBrA-Net: Lie-Algebraic Bilateral Affine Fields for Real-Time 4K Video Dehazing
- LiDAR Scene Synthesis with Triplane Diffusion Models via LiDAR Simulation-based Coarse-to-Fine Point Sampling
- Lie Generator Networks for Nonlinear Partial Differential Equations
- LieSolver: PDE-Constrained Learning for IBVPs via Lie Symmetries
- Lifted Geometry-Aligned GNNs for Min-k-Partition
- Lifted Hamiltonian dynamics for designing accelerated first-order methods
- LiFT: Likelihood-Free Tree-Structured Policy Optimization for Flow-Based VLAs
- LIGHT: Deployable Small Foundation Models
- LIGHT: Local Importance Guided High-order Trimming for Neural Networks
- LightNorm: A Lightweight Alternative to Fine-Grained Normalization
- Lights, Camera, Carbon: Architectural Scaling Laws for Video Generation Energy Consumption
- Lightweight Explainable Physics-informed Neural Networks by Learnable Activation Function and Contextual Modulation
- Lightweight Multi-Relational Graph Learning for Natural Language Service Composition
- Likelihood-Free Generative Policy Optimization
- LILA: Calibration-Free Structured Pruning of Large Language Models via Latent Spectral Geometry
- LIME: Link-based User-item Interaction Modeling with Decoupled XOR Attention for Efficient Test Time Scaling
- LIME: Making LLM Data More Efficient with Linguistic Metadata Embeddings
- LiMemFlow: Memory-Enhanced Flow Matching for Lithium-Ion Transport in Solid-State Electrolytes
- LinAlg-Bench: A Forensic Benchmark Revealing Structural Failure Modes in LLM Mathematical Reasoning
- LINC: Decoupling Local Consequence Scoring from Hidden Matching in Constructive Neural Routing
- Linear approximations to HMM filtering
- Linear Attention Scales Better with Shared Latent Memory Projection
- Linear Contextual Bandits with Quasi-Optimism
- Linear Ensemble Sampling with Fewer Ensembles
- LinguaMotion: Interpreting the Language of Motion for Generalizable Robot Manipulation
- Linguistic Cartography: Metric Spatial and Hierarchical Platial Representations in Language Models
- LinUCB Is (Almost) All You Need: A Regime Analysis of LLM Warm-Starts in Contextual Bandits
- LIPAR: Latent Inter-Frame Pruning with Attention Recovery
- Lipschitz-Guided Monte Carlo Tree Search with Knowledge Transfer across Sequential Tasks
- Lipschitz-inspired Amplified Learning for Noise-Insensitive Trajectory Generation
- Liquid Temporal Dynamics Enhances Generalization Ability in Architecture-Isolated Visual Control
- LiteNav: Lightweight Map-free Outdoor Visual Navigation
- LiveProteinBench: A Contamination-Free Benchmark for Assessing Models' Specialized Capabilities in Protein Science
- LLaVA-UHD v4: What Makes Efficient Visual Encoding in MLLMs?
- LLM Advertisement based on Neuron Auctions
- LLM Agent Collaboration Needs Trust Framing
- LLM Alignment Must Continuously Integrate Population-Representative Survey Data
- LLM Alignment--Utility Asymmetry under Semantic-Preserving Transformations
- LLM-Auction: Generative Auction towards LLM-Native Advertising
- LLM-based Semantic Enhancement and Policy Network Distillation for Fast Task Planning
- LLM Compression by Block Removal with Constrained Binary Optimization
- LLM-Enhanced Random Forests in Orthogonal Hyperbolic Subspaces for Tabular Learning
- LLM Expects Maturity from Older Children
- LLM Hypnosis: Characterizing the Fragility of RLHF Against Unprivileged Knowledge Injection
- LLM Is a Good Conditioner: End-to-End Sign Language Video Generation with VQ-Diffusion
- LLM Moral Reasoning Does Not Inherently Require Diversity-Preserving RL
- LLMs Show No Signs Of Individuated Metacognition
- LLMs Struggle to Rank Products Robustly
- LLM Wardens: Mitigating Adversarial Persuasion with Third-Party Conversational Oversight
- LMARL: LLM-Enhanced Multi-Agent Reinforcement Learning for Portfolio Management
- LMEB: Long-horizon Memory Embedding Benchmark
- LMR: Towards Expressive Fully Inductive Relation Prediction With Latent Mechanisms For Knowledge Graphs
- Local Causal Attribution of Chain-of-Thought Reasoning
- Local-Curvature-Aware Knowledge Graph Embedding: An Extended Ricci Flow Approach
- LOCALE: Local Alignment Embeddings for Noise-Robust DNA Search at SRA Scale
- Local Guidance, Global Impact: Gaussian-Reshaped Trust Region Unlocks Behavior Transitions
- Local Influence Inference: Identifiable Feature Rankings from Spatial Neighborhoods
- Local-Interaction Learning Dynamics: A Markov Random Field Framework for Convergence of Deep Neural Network Learning
- Local Intrinsic Dimension Unveils Hallucinations in Diffusion Models
- Locality Sensitive Hashing for p-Exponential Kernels with Applications to Density Estimation
- Localize Any Object in X-Ray Security Scans without Human Annotation
- Localized Dynamics-Aware Domain Adaption for Off-Dynamics Offline Reinforcement Learning: A Closer Look
- Localized, Unstable, and Entangled: Exploring ‘Us vs. Them’ Bias in Large Language Models
- Localizing and Repairing Sparse-Prompt Failure in SAM Decoders via Box-to-Point Counterfactual
- Localizing Memorization in Graph Neural Networks
- Local Linear Convergence of Projected Gradient Descent from an ODE Perspective
- Local LMO: Constrained Gradient Optimization via a Local Linear Minimization Oracle
- Local Truncation Error-Guided Neural ODEs for Large Scale Traffic Forecasting
- LOCI: A Locator-Critic with Refinement Loop
- Locking Pretrained Weights via Deep Low-Rank Residual Distillation
- LOCO: Local Light-Aware Object Compositing with Spatially Varying Illumination-Augmented Data
- LoFi: Low-Frequency RoPE Channels for Position-Invariant KV Cache Merging for LLMs
- Logarithmic-time Schedules for Scaling Language Models with Momentum
- LogiAgent: A Logic-Adaptive Multimodal LLM Agent for Trustworthy Decision-Making in High-Stakes Domains
- Logic-Aligned Multimodal Offline Reinforcement Learning for High-Stakes Decision-Making
- Logic Bus: When Operator Matters More Than Vector
- LogicSR: A Unified Benchmark for Logical Discovery from Data
- LogicTree-RAG: Logic Tree-guided Retrieval-Augmented Generation for Long-form Patent Drafting
- LogiFakeID: Exposing the Logical Manipulation Blind Spot in Identity Verification
- Logistic Smoothing for Sparse Generalized Additive Models
- Logo10M for Benchmarking Large-Scale Logo Retrieval
- Long Context Pre-Training with Lighthouse Attention: A Closer Look
- Long-Context Retrieval Through One Learnable Attention Temperature
- Long-Horizon Embodied Decision-Making via Multimodal Memory Compression
- Long-horizon Multi-Agent Exploration With Contrastive Frequency Prediction
- Long-Horizon Streaming Video Generation via Hybrid Attention with Decoupled Distillation
- LongHorn: Long Term Human Object Removal via Global Context Extraction
- Longitudinal-MedGemma: Learning Discontinuous Temporal Dynamics for Longitudinal Medical Imaging
- LongLive-2.0: An NVFP4 Parallel Infrastructure for Long Video Generation
- LongMemEval-V2: Evaluating Long-Term Agent Memory Toward Experienced Colleagues
- Long-Range Spatio-Temporal Graph Propagation Through Oscillations
- LongRM: Revealing and Unlocking the Context Boundary of Reward Modeling
- LongSpike: Fractional Order Spiking State Space Models for Efficient Long Sequence Learning
- Long-Term Risks of Risk-Based Allocation
- Long Video Instructional Editing in the Wild
- Look-Ahead from Following Context: A Cue for AI-Generated Text Detection
- Look-ahead Fusion Attack for Transferable Targeted Attacks on Vision-Language Models
- Look-ahead Variational Flow for Generative Online Reinforcement Learning
- Look Before You Act: Bridging Active Perception and Vision-Language-Action Models via Observation-Aligned Data Adaptation
- Looking Back to Move Forward: Temporal Verification for Generative Robot Policies
- LookThere! Sparse Vision by Reinforced Selection
- Loop alignment: Self-organized Weight Transpose in Predictive Coding through Independent Hebbian Plasticity.
- LoopCTR: Unlocking the Loop Scaling Power for Click-Through Rate Prediction
- LoopNav: Benchmarking Spatial Consistency in World Models
- LoopQ: Quantization for Recursive Transformers
- LoopUS: Recasting Pretrained LLMs into Looped Latent Refinement Models
- LoRAcles: Self-Supervised Weight-Space Interpretability at Scale
- LoRA-GC: Federated Low-Rank Adaptation with Factor-Space Gradient Centralization
- LoRA is All You Need for Safety Alignment of Reasoning LLMs
- LoRA-Muon: Spectral Steepest Descent on the Low-Rank Manifold
- LoRAtorio: An intrinsic approach to LoRA Skill Composition
- LoReC: Rethinking Large Language Models for Graph Data Analysis
- Loss is Not Behavior: A Unified Output-Space Analysis of Gradient-Based Machine Unlearning
- Loss-Tail SGD: Generalization-Aware Optimization via Batch Rejection Sampling
- Loss Transformation Invariance of the Damped Newton Method
- Lost in Distribution: SAEs Under Domain Shift
- Lost in Translation: Evaluating Multilingual Gaps in Vision-Language-Action Models
- Lost on Campus: Evaluating Embodied Spatial Reasoning of Vision-Language Models in the Wild
- Lost or Hidden? A Concept-Level Forgetting in Supervised Continual Learning
- LoTR: Logic-of-Thought Routing for Plug-and-Play Reasoning of LLMs
- LoVE: A Lego-Like Video Editing Framework
- Low-Cost Black-Box Detection of LLM Hallucinations via Dynamical System Prediction
- Lower Bounds and Proximally Anchored SGD Under Non-Convexity and Unbounded Variance
- Lower bounds for multi-group transductive learning
- Low-Rank Adaptation for Critic Learning in Off-Policy Reinforcement Learning
- Low-Rank Adapters Initialization via Gradient Surgery for Continual Learning
- LowRankArena: A Standardized Evaluation Platform for SVD-Based LLM Compression
- Low-Rank Class-Adaptive Federated Continual Learning for Continual Knowledge Retention
- Low-Rank Hierarchical Merging for Efficient Long-to-Short Reasoning
- Low-rank Interpretable Cell–Cell Hidden Interactions from Embeddings
- Low-Rank Pairwise Ranking with Sequence-Induced Dependence and Graph Sparsification
- Low-Resolution Key Region Generation for High-Resolution Data-Free Knowledge Distillation
- LoWR: LoRA Weight Rescaling for Effective Rank Utilization beyond Reduction
- LPDP: Inference-Time Reward Control for Variable-Length DNA Generation with Edit Flows
- LPE-Router: Extensible LLM Routing via Lightweight Probe-based Encoding
- LPS-Bench: Benchmarking Safety Awareness of Computer-Use Agents in Long-Horizon Planning under Benign and Adversarial Scenarios
- LUA: Dual-Global-Model Local Update Accumulation for Efficient Heterogeneous Distributed Training (Extended)
- LucidLens: Eliminating Memorization Leakage in Language Model Probes
- LUCID: Representation Learning for Neurophysiological Time Series via Diffusion Model Activations
- Lumen: Consistent Video Relighting and Harmonious Background Replacement with Video Generative Models
- LVC-World: Structured State Abstraction for Prediction under Perception Degradation
- LVS: Learning Latent Visual Segmentation via Hierarchical Semantic-to-Spatial Reasoning
- Lyapunov-Certified Direct Switching Theory for Q-Learning
- Lyra: Hierarchical Alignment for Autoregressive Audio Generation via Variational Policy Optimization
- M$^2$E-UAV: A Benchmark and Analysis for Onboard Motion-on-Motion Event-Based Tiny UAV Detection
- M$^2$TSF: LLM-empowered time series forecasting via multi-stream multi-scale modeling
- M3-HNTM: Hyperspherical Multimodal Topic Modeling with Symbolic and Contextual Evidence
- M3Risk: Assessing the Helpfulness of MLLMs for Malicious Intent-Driven Misuse
- Machine Learning-Based Variance Reduction for Kaplan-Meier Estimators in Survival Analysis
- Machine Learning for Simulations in Biology and Chemistry - The 2nd SIMBIOCHEM Workshop
- Machine Learning for Spatially Resolved High-dimensional Biology
- Machine Learning Research Conferences Should Establish an Open Source Research Track
- Machines See Better When They Imagine: Diffusion Imagination Reveals Better Visual Priors for Multimodal Reasoning
- Macroscopic Edge Representations for Transfer Learning on Dynamic Text-Attributed Graphs
- MaD-Mix: Multimodal Data Mixtures via Latent Space Coupling for Vision-Language Model Training
- MADSYN: Model-Aware Data Synthesis for Automated Bias Mitigation
- MAEB: Massive Audio Embedding Benchmark
- MAES: Modality-Aware Expert Slimming for Multimodal Mixture-of-Experts
- MAESTRO: Reinforcement Learning to Orchestrate Hierarchical Model-Skill Ensembles
- MafiaPersona: A Multi-Agent Adversarial Benchmark for Evaluating Persona Persistence in Large Language Models
- MAGE: Multi-Agent Self-Evolution with Co-Evolutionary Knowledge Graphs
- MAGIC: Material-Agnostic Gaussians for Identifying Heterogeneous Continuum
- MAGIC: Predictive Data Attribution at Scale via the Exact Influence Function
- MAGMA: Agentic Clinical AI Framework with Guideline-Constrained Reinforcement Learning
- MAGNET: Manifold-Aware Graph Diffusion Network for Connectome Generation
- Magnifying What Matters: Attention-Guided Adaptive Rendering for Visual Text Comprehension
- Making Complex Reasoning Student-Friendly: A Hybrid Distillation Framework for SLMs
- Making Generative Protein Language Models Faster and Less Repetitive with Post-Training
- Making Multi-Axis Models Robust to Multiplicative Noise: How, and Why?
- Making Open-Source Text LLM Watermarks Durable Against Merging
- Making Reconstruction FID Predictive of Diffusion Generation FID
- Making Transformers Stateful with Fast-Weight and Typed Episodic Memory
- MA-LoRA: Anchoring Low-Rank Updates to Data Manifolds via Implicit Projection and Energy Gating
- Mamba-Flow: Transferable Mixed-Precision Quantization via LLMs-Evolved Scoring Functions
- Mambura: A Mamba-Transformer Hybrid for Unified Multimodal Understanding and Generation
- M*: A Modular, Extensible, Serving System for Multimodal Models
- Managing Agents that Manage Agents: Workshop on Responsible Use of Meta-Agents that Build, Optimize, and Supervise Other Agents
- MANGO:Multi-Angle Neural Gated Operators for Chirp-Perturbed PDEs
- ManifoldAE - Mixture of Expert Autoencoders for Manifold Structured Data
- Manifold-Aware Dual Rejection Sampling for Diffusion Models with Limited Data
- ManifoldEdit: Geometry-Aware One-Step Image Editing on the Clean Latent Manifold
- Manifold Prior Guided Deep Unfolding for Hyperspectral Image Reconstruction
- ManipGPT: A Generative Action-Only Pre-trained Model for Robotic Manipulation
- ManipShield: A Unified Framework for Image Manipulation Detection, Localization and Explanation
- Manipulation Is Task-Dependent: A Multi-Axis, Multi-Environment Evaluation of Frontier LLMs
- Many Within One: Type Sampling for Practical LLM-Based Survey Synthesis
- MAPLE: Marker-Aware Protein Latent Embedding for Spatial Proteomics Reconstruction and Unseen Marker Prediction
- MAPLE: Mechanism-Aware Posterior LEarning for MNAR Correction in Tabular Imputation
- MAPPA: Scaling Multiagent Systems with Process Rewards
- Mapping Uncharted Symmetries: Machine Discovery in Combinatorics
- MARCEDES: Score-based causal discovery under non-Gaussianity with continuous optimization
- MARD: Multi-mode Anchored Residual Diffusion for Conservative Trajectory Planning
- Margin-Drop Coordinates for Cross-Budget Robustness Audits
- Marked Inducing-Point Cascaded SDEs for Interpretable Neural Trajectory Inference
- MarketMirror: A Physics-Informed Causal Potential Framework with Expert-Gated Alignment for Risk Inference under Concept Drift
- Marking the Wrong Symptoms: Evaluating LLM Watermarks in Medical Texts
- MaRK: Markov-adapted Recurrent Kernels for Dynamic Operator Conditioning in State Space Models
- Markov Proposals for Improving Vision-Language Alignment in Large Vision-Language Models
- Markov-Stitching: Data Augmentation for Efficient Transition Path Learning
- MARLHospital: A Benchmark for Evaluating Fairness in Heterogeneous Multi-Agent Teams
- Marrying Optimal Transport and ODEs for Unified Continuous-Time 4D Reconstruction and Tracking
- MARS: Margin-Adversarial Risk-controlled Stopping for Parallel LLM Test-time Scaling
- MARS-RL: Order-Invariant Learning for Autoregressive Map Generation
- Mask-Conditioned Gradient Masking for Fine-Tuning Mixture-of-Experts Diffusion Language Models
- MaskDD: Sculpting Discriminative Features for Dataset Distillation
- Masked Visual Actions for Unified World Modeling
- Masked Visual Fine-Tuning for Encoder-based Diffusion Models.
- MaskForge: Structure-Aware Adaptive Attacks for Jailbreaking Diffusion Large Language Models
- MaskMIL: Weakly Supervised Anomaly Detection in Physiological Time Series Through Mask Learning and Contrastive Mask Regularization
- MaskSense: Confronting the Visual Exploration Trap in Masked Image Generation
- Massively Parallel Exact Inference for Hawkes Processes
- MAST: A Multi-fidelity Augmented Surrogate model via Spatial Trust-weighting
- Matched Abstention Audits Native Operating-Point Wins in Sequential External-Memory Maintenance
- Matching Features, Not Tokens: Energy-Based Fine-Tuning of Language Models
- Matching Object or Relation? Tracing Abstract Reasoning Inside VLMs
- Match the Geometry, Skip the Surrogate: Extreme Low-Budget Optimization in High Dimensions
- MatGPTQ: Accurate and Efficient Post-Training Matryoshka Quantization
- MATH-AI: The 6th Workshop on Mathematical Reasoning and AI
- MathAtlas: A Benchmark for Autoformalization in the Wild
- MathCD: A Benchmark Dataset for Cognitive Diagnosis with Semantic Information
- Matrix Recovery Via Symmetric Rank-one Measurements With Random Unit-modulus Vectors
- Matryoshka Concept Bottleneck Models
- Matter to Mechanism: A Benchmark for AI Co-Scientists in Materials and Battery Research
- Maturing Markov Decision Processes: Decision Making under Increasing Information and Shrinking Action Sets
- MaxSketch: Robust Distinct Counting in Streams via Random Projections
- MCLM: Mixed-Curvature Space for Language Modeling
- MCP-Atlas: A Large-Scale Benchmark for Tool-Use Competency with Real MCP Servers
- MCPShield: A Security Cognition Layer for Adaptive Trust Calibration in Model Context Protocol Agents
- MCTS-QAG: Pareto-Optimized QA Generation over Real-World Evidence Chains
- MD-ACE: MLLM-guided Semantic Distillation and Adaptive Core Embeddings for Visual Grounding
- MDFE-DVLO: Multi-Dimensional Feature Enhancement for Deep Visual-LiDAR Odometry
- MDK-MoE: Multi-view Decomposed Kalman Mixture of Experts Framework for Non-stationary Time Series Forecasting
- MDoom: A Controlled First-Person Benchmark for Agentic Vision-Language Models
- Mean-Field Control on Sparse Graphs: From Local Limits to GNNs via Neighborhood Distributions
- Mean-Independent Fair Representation Learning with Excess-Risk Guarantees
- Mean-Shifted Intermediate Sampling: Training-Free Acceleration for Few-Step Diffusion Generation
- Mean Testing under Truncation beyond Gaussian
- Measure Less, Know More: Self-Supervised Test-Time Feature Acquisition
- Measure Transformation via Score Correction in Diffusion Models
- Measuring AI Agents' Progress on Multi-Step Cyber Attack Scenarios
- Measuring and Strengthening Behavioral Suppression in Language Models
- Measuring Layer-wise Intrinsic Dimensionality of FFNs in LLMs via PCA
- Measuring Orthogonality as the Blind-Spot of Uncertainty Disentanglement
- Measuring Preferred Representation in AI-generated Images with Community Scorers
- Measuring Unsupported-Belief Reinforcement in Multi-Turn LLM Dialogue
- Mechanism Design for AI Overviews: Creator Incentives and Long-Term Profit
- Mechanism Design for Generative Engine: From Exploitation to Win-Win Equilibrium
- Mechanism Parameter Optimization under Behavioral Uncertainty via Sequential Experimental Design
- Mechanisms and Limits of Sycophancy Mitigation in Instruction-Tuned LLMs
- Mechanisms Matter: Transportability of Cellular Perturbation Effects
- Mechanistic Attention Guidance for Agent Memory Refinement
- Mechanistic Interpretability of EEG Foundation Models via Sparse Autoencoders
- Mecha-nudges for Machines
- MedAgentWorld: Medical Agentic Reinforcement Learning with World Model
- MedCache: Training-Free Spatially Aware Caching for Accelerated Medical Video Generation
- MedEvoEval: Evaluating Continual Evolution of Doctor Agents through Simulated Clinical Episodes
- MedGuards: Multi-Agent System for Reliable Medical Error Detection and Correction
- Medical Reasoning with Multimodal Foundation Models
- MedMisBench: Measuring Epistemic Resilience of LLMs Under Misleading Medical Context
- Med-RewardBench: Benchmarking Reward Models and Judges for Medical Multimodal Large Language Models
- MemArena: An Ego-Centric Benchmark for On-Device Agentic Personal Memory Assistants at Scale
- MemCompiler: Compile, Don’t Inject — State-Conditioned Memory for Embodied Agents
- MemContract: Contract-Sensitive Evaluation for Mutable Agent Memory
- MemCoRe: Recovering Evidence from Progressively Compressed Factual Knowledge for Agent Memory
- MemDirector: Memory-augmented Director for Long-form Video Storytelling
- MemEdit: Fine-Grained Memory Surgery for Expandable Parametric Agent Memory
- MEME: Multi-Entity & Evolving Memory Evaluation
- MEMENTO: Leveraging Web as a Learning Signal for Low-Data Domains
- MEMEVO: A Memory-Evolved Video Agent for Long Video Understanding
- MemEye: A Visual-Centric Evaluation Framework for Multimodal Agent Memory
- MeMix: Selective State Preservation for Streaming 3D Reconstruction
- MeMo: Memory as a Model
- MemoReason: Evaluating the Effect of Parametric Memory on Contextual Reasoning in LLMs
- Memorization Detection in Diffusion Models via Text Embedding Interpolation
- Memorization Is Folding: Topological Signatures of Noisy-Label Learning
- Memorize-and-Generate: Towards Long-Term Consistency in Real-Time Video Generation
- Memory by Design: Probabilistic Sequence Layers
- Memory-Driven Contrastive Embedding Enhancement for Fine-Grained Open-Set Semi-Supervised Learning
- Memory-Efficient Continuous Adjoint Sensitivity for Neural SDEs via Signature Kernels
- Memory-Efficient Federated Fine-Tuning of LLMs via Block-wise Progressive Training
- Memory-Efficient Self-Supervised Incremental Video Hashing via Latent Replay
- Memory flows: geometry and dynamics of sequential retrieval in input-driven Hopfield networks
- MemoryFusion: Cross-Temporal Memory Learning for Multimodal Video Fusion
- Memory is Not Search: Towards Proactive, Lifelong Memory in AI
- Memory Is Not the Bottleneck: Disentangling Memorization and Agentic Linkage in LLM Re-identification of Public Records
- Memory Optimized Structured Backward Dropping
- Memory Residuals: An O(1)-Storage, O(1)-Routing Recurrent Memory Primitive for Frozen Pretrained Language Models
- Memory Retrieval for Changing Preferences
- Memory Type Varies: Empowering LLM Agents for Long-Term Memory with Diverse Strategies
- MemReg: Streaming Outdoor LiDAR Point Cloud Registration with Hybrid Memory Buffers
- MemSkate: Trace, Assess, and Adjudicate with Memory-Evolving Multi-Agent Framework for Figure Skating Video Reasoning
- Meno: Interpretable Genomic Foundation Modeling via Region-based Memory Retrieval
- MERA: Matrix-Entropy Reranking Actor-Critic for Sequential Recommendation
- MERIT: A Sentence-Level Benchmark for Earnings Call Language Trajectories and Sub-Minute Market Microstructure
- MermaidSeqBench: An Evaluation Benchmark for NL-to-Mermaid Sequence Diagram Generation
- MESA: Selection-Optimized Metadata for Agent and Skill Routing in Multi-Agent LLM Systems
- Mesh BDF: Barycentric Dominance Field for 3D Native Mesh Generation
- MeshCrafter: Boundary-Aware Autoregressive Mesh Generation with Structural Stability Control
- MetaboNet-Bench: A Multimodal Benchmark for Glucose Forecasting in Type 1 Diabetes
- Metacognition as Reward: Reinforcing LLM Reasoning via Knowledge and Regulation Signals
- MetaColloc: Optimization-Free PDE Solving via Meta-Learned Basis Functions
- MetaGauge: A Framework for Quantification of Metacognitive Skills in Large Language Models
- Meta-incremental Learning
- Meta Inverse Prompting for Video Generative Models
- Meta-Learned CMA-ES Prompt Optimization for Test-Time Adaptation in Zero-Shot Video Object Segmentation
- Meta-Learned Exposure Correction for Rapid Adaptation Across Devices
- Meta-Learning for AI-Generated Image Detection
- META-PAP: Meta-learning for Prompt-aware Preference Pairing in LLM Alignment (Extended)
- Metaphor Is Not All Attention Needs
- Meta-Reinforcement Learning with Zero-Shot Reinforcement Learning
- Meta-SA-BCP: Parameter-Free State-Adaptive Conformal Prediction via Online Expert Aggregation
- Meta-TTRL: A Metacognitive Framework for Self-Improving Test-Time Reinforcement Learning for T2I Generation in Unified Multimodal Models
- METIS: Multi-Source Egocentric Training for Integrated Dexterous Vision-Language-Action Model
- MetricHand: Grounding 3D Hands in Metric Camera Space via Depth Priors
- Metric Illusion: How Evaluation Choices Diverge Conclusions in Graph Learning
- Metric Shift: Predicting Expensive Scientific Measurements from Cheap Ones
- Metric Unreliability in Multimodal Machine Unlearning: A Systematic Analysis and Principled Unified Score
- METRO: Metric-Enhanced Token Routing Operator
- Metropolis Matching: Promoting Mode Coverage in Normalizing Flows
- MEV: A Multi-Event Video Dataset for Long-Take Generation
- MFS: A Saliency-Driven Multimodal Interactive Learning Framework for Robust Semantic Segmentation using Infrared-Visible Images
- MGHF: Multi-Granular High-Frequency Perceptual Loss for Image Super-Resolution: A Closer Look
- MGPO: Manifold-Guided Diffusion Alignment for Task-Aware Dataset Distillation
- MH-CRL: Nonlinear Unpaired Multi-Domain Causal Representation Learning for Heterogeneous Medical Imaging
- MIAAM: a Semantically Rich Multimodal Education Dataset of 7 Million Maths Learning Interaction that Challenges Frontier AI Models
- Miami: Missingness Imputation via Adversarial Multichoice Iterations
- MIBE: Multi-subject Interaction Benchmark and Evaluator for Personalized Image Generation
- MICA: Activation Checkpointing for Double-Backward Training of Machine Learning Interatomic Potentials
- MICA: Multi-granularity Intertemporal Credit Assignment for Long-Horizon Emotional Support Dialogue
- Microstructure Descriptor Fields as Supervision for Scientific Images
- Micro-to-Macro Evidence Grounding for Mechanism-Faithful Simulator Construction
- MicroWorld: Empowering Multimodal Large Language Models to Bridge the Microscopic Domain Gap with Multimodal Attribute Graph
- Mid- (Not Late-) Layer Representations Drive Brain Alignment in Vision Models
- Midpoint Generative Models
- MIITA: Memory-Induced Inference-Time Adaptation for Continual Learning with Small Language Models
- MIND: Automated Optimization Modeling via a Localizable Error-Driven Perspective
- MIND-DDI: Multi-Omics Interpretable Drug-Drug Interaction Prediction with Joint Optimization of Graph Structure, Neural Architecture, and Symbolic Rules
- MindGames:A Multi-Agent Benchmark and Trajectory Dataset for Evaluating Social and Strategic Reasoning in LLMs
- MindLoom: Composing Thought Modes for Frontier-Level Reasoning Data Synthesis
- MindStrata: Learning Robust Multi-Subject Cortical Visual Representations via Stratified Layout-Semantics Decoding
- Mind the Delay: Balancing Accuracy, Cost, and Latency in LLM Query Routing
- Mind the Gap: Bridging Benchmarks and the Real World in Face Anti-Spoofing
- Mind the Gap: Constrained Reconstruction for EEG Foundation Models
- Mind the Gap: Phase-Aware Temporal Contrastive Learning Bridges Perception and Imagination in EEG
- Mind the Gap: The Divergent Rebound Dynamics of Diffusion and Autoregressive Model
- Mind the Loop: Feedback-Safe LLM Priors for Causal Discovery
- Mind the Tool Failures: Achieving Synergistic Tool Gains for Medical Agents
- MindVoice: Reconstructing Intelligible Speech from Non-invasive Neural Signals with Pretrained Priors
- MINER: Mining Multi-Modal Internal Representation for Efficient Retrieval
- MineXplore: An Open-Source Reinforcement Learning Exploration Benchmark for GNSS-Denied Underground Environment
- Minimal-Action Discrete Schrödinger Bridge Matching for Peptide Sequence Design: A Closer Look
- Minimal-Overhead Bit-Plane Multiplexing of Neural Network Parameters
- Minimal Sufficient Benchmarks: Score Imputation for Cost-Efficient LLM Leaderboards
- Minimax Optimal Kernel Two-sample Testing in Sub-quadratic Time
- Minimax Regret Laws for Learning-Augmented Online Adaptation
- Minimax Round Complexity of Distributed Offline Dynamic Programming: Discounted Locality, Bit--Radius Tradeoffs, and Gossip Baselines
- Minimum-Sufficient Selective Control for Instruction-Based Image Editing
- Mining the Shadows: Distilling Dark Ranking Knowledge for OOD Detection.
- Minionese: Comprehensive Benchmark and Mechanistic Study of Multilingual LLM Safety
- mini-vec2vec: Scaling Universal Geometry Alignment with Linear Transformations
- MIRACL: Diversity-Aware Hierarchical Meta-Reinforcement Learning for Sustainable Multi-Echelon Supply Chain Optimisation
- MIRAGE: Adaptive Multimodal Gating for Whole-Brain fMRI Encoding
- MIRAGE: Medical Image Retrieval Augmented Generation Engine
- MIRAGE: Motion Identity Removal via Autoregressive Generative Encoding for Privacy-Preserving Skeleton-based Motion Data
- MIRAGE: Mutual Information-guided Reconstruction and Adaptation for Graph Domain Adaptation with Edge Incompleteness
- MIRA: Reinforcing Multimodal Reasoning via Deceptive Contextual Augmentation
- Mirror Descent-Ascent for mean-field min-max problems
- Mirror Learning
- MIRROR: Manifold Ideal Reference ReconstructOR for AI-Generated Image Detection
- MIRROR: Multi-Ratio Mixing and Region-Aware Refinement for Training-Free Zero-Shot Composed Person Retrieval
- MISB: A Benchmark for Metacognitive Identity Stability in Large Language Models
- Misled by the Few: A Mechanistic Analysis of Demonstration Conflicts in In-Context Rule Inference
- MISP-Bench: Decomposing User-Provided False Priors into Answer, Rationale, and Guard Effects
- Mistake-Bounded Language Generation
- Mitigating Adaptive Attacks against Reasoning Models with Activation Consistency Training
- Mitigating Content Shift and Hallucination in GenAI Image Editing via Structural Refinement
- Mitigating Many-shot Jailbreak Attacks with One Single Demonstration
- Mitigating Overgeneralization in RND via Spectral Target Design
- Mitigating Safety Tax via Distribution-Grounded Refinement in Large Reasoning Models
- Mitigating Saliency Collapse: Robust Saliency-Aware Long-Text Image-Text Alignment
- Mixed-Curvature Latent Representation Learning for Structured Perturbation Prediction
- Mixed-Integer Relaxation Activation Enables Channel-Adaptive Nonlinear Feature Learning in Deep Neural Networks
- Mixing Makes Markovian Contexts Cheap for Linear Bandits
- Mixture of Activations: Token-Adaptive Mixing for Expressive Feedforward Layers
- Mixture of Attribute-Aware Attention Experts for Fine-grained E-Commerce Composed Image Retrieval
- Mixture-of-Control: State-Aware Fine-Tuning for Transformer-based Models
- Mixture of Domain Experts (MoDE): Specializing Domains for Fast LLM Inference
- Mixture-of-Rotations: Soft Expert Routing for Rotation-Based Feed-Forward Networks
- Mixture of Sketches: Learning Distributed Sketch Aggregation with Tight $\Omega(\sqrt{K})$ Bounds
- Mixtures of Neural Operators Reduce Active Complexity in Operator Learning
- ML4D: Scalable Feed-forward 4D Gaussians for Minute-long World Modeling
- MLAIRE: Multilingual Language-Aware Information Retrieval Evaluation Protocol
- MLAuditBench: An Interactive Reinforcement Learning Environment for Evaluating LLM Agents on ML Experiment Integrity Auditing
- ML-Bench&Guard: Policy-Grounded Multilingual Safety Benchmark and Guardrail for Large Language Models
- ML for Systems
- MLLMEraser: Achieving Test-Time Unlearning in Multimodal Large Language Models through Activation Steering
- MLLM Makes Strong Backbone for Multi-Modal Object Detection
- MLLMs Fail to Refuse when Using Tools Agentically
- ML-PatchGCD: Multi-Label Generalized Category Discovery via Patch-level Semantic Residual Decomposition
- MMAgentic: Benchmarking Agentic Capability of Multimodal Large Language Models
- MMFineReason: Closing the Multimodal Reasoning Gap via Open Data-Centric Methods
- MMInduction: Towards Inductive Multimodal In-Context Learning
- mmLIP: mmWave Radar-Language Interactive Pretraining via Point Confidence
- MM-OptBench: A Solver-Grounded Benchmark for Multimodal Optimization Modeling
- MM-TS: Channel-Structured Vision-Language Modeling for Multivariate Time Series Forecasting
- MOA Dreamer: A Compartmentalized Model Architecture for Continuous Memorization
- MobiGraph: Towards a Practical Evaluation Paradigm for Mobile Agents via Trajectory-Fused State Graphs
- MobileWan: Closing the Quality Gap for Mobile Video Diffusion
- Modality-Adaptive Depth Up-Scaling for Multimodal LLM Adaptation
- Modality-Aware Expert Pruning for MoE-Based Multimodal Large Language Models
- Modality Partition in Multimodal Large Language Models: A Layer-wise Explanation of Language Degradation
- Model and Predict Time-Inconsistent Agents in Contextual Choice Tasks
- Model-Based Globalization of Quasi-Newton Methods with Explicit Stepsizes
- Model-Based Meta-Learning for Algorithm Discovery
- Model Collapse is a Singular Complexity Trajectory
- Model Compression with Exact Budget Constraints via Riemannian Manifolds
- Model Distribution-Aware Multimodal Dataset Distillation
- Model Evaluation Should Treat Information Access as a First-Class Variable
- Model-Free Neural Filtering: A Comparison with Classical Filters in Nonlinear Systems
- Model generalization is information accounting
- Modeling Local, Global, and Cross-Modal Context in Multimodal 3D MRI
- Model Merging with Functional Dual Anchors
- Model Parallelism With Subnetwork Data Parallelism
- Model Spec Midtraining: Improving How Alignment Training Generalizes
- Model Understanding and Reasoning with Transformers: A Theoretical Perspective
- MODEPrompt: Taxonomy-guided Pareto Evolution for Prompt Optimization
- MoDiTac: Modality-Decoupled Diffusion Transformer for Controllable Visual-to-Tactile Generation
- Module-Aware Optimization for Graph Neural Networks
- MoE-ESM++: Robust TCR–Epitope Binding Prediction via Adversarial Refinement and Mixture-of-Experts Adaptation
- MoE-SpAc: Efficient MoE Inference Based on Speculative Activation Utility in Heterogeneous Edge Scenario
- MoGeoNet: A Multi-Order Geometric Neural Network for Modeling 3D Glycan Conformations
- MoG: Mixture of Experts for Graph-based Retrieval-Augmented Generation
- MoG-Splat: Structured Per-Pixel Gaussian Mixtures for Generalizable Novel View Synthesis
- MolDeTox: Evaluating Language Model’s Stepwise Fragment Editing for Molecular Detoxification
- Molecular Representations in Implicit Functional Space via Hyper-Networks
- Moment kernels: a simple and scalable approach for equivariance to rotations and reflections in deep convolutional networks
- Momentum-Accelerated Adaptive Federated Multi-objective Optimization
- Momentum as Residual-Driven Multiplier Correction for Deep Learning Optimization
- MonarchRT: Efficient Attention for Real-Time Video Generation
- MONET: A Massive, Open, Non-redundant and Enriched Text-to-image dataset (Extended)
- MonoMAFA: Multi-Attention Feature Aggregation Framework for Robust Monocular 3D Object Detection in Adverse Weather
- MonoSVS: Generate Street View Synthesis with Monocular Videos
- Monte Carlo Random Walk Kernel
- Moral Intelligence Requires Structural Preconditions That LLMs Lack
- Morality Is (Mostly) Normativity: A Cross-Architecture Feature-Level Analysis of Moral-Stance Circuits via Sparse Autoencoders
- Moral Orientation and Calibration: Coupled in Human Annotators, Separable in Judge LLMs
- More Force Is Not the Answer for Adversarial Transferability on Vision Transformers: A Closer Look
- More is not better: Clinical performace Saturarion in LLMs
- MORE: Rethinking Model Merging for Reinforcement Learning Enhancement in Large Language Models
- More Than Can Be Said: A Benchmark and Framework for Pre-Question Scientific Ideation
- More Value per Key: Asymmetric Sparse Attention for Faster LLM Decoding
- Mori-Zwanzig Memory for Stable Autoregressive Generative PDE Rollouts
- MorphCL: Morphological Contrastive Learning for Inertial-based Human Activity Recognition
- MorphGen: Controllable Cell-Image Generation with Biological Representation Alignment
- MorphoHELM: A Comprehensive Benchmark for Evaluating Representations for Microscopy-Based Morphology Assays
- Morphology-Regularized Tokenization for Low-Resource Indic Languages
- Morpho-Temporal Decoupling: How Primate Neurons Expand Dendrites Without Losing Speed
- MoRSE: Task-Oriented Multi-Agent System with Mixture of Role-Subtask Experts
- Mosaic: A Benchmark Suite for Differentiable Physics Solvers
- MOSAIC: A Multimodal, Multi-Agent Orchestrated System with Adaptive Memory for Tri-Horizon Portfolio Construction
- MoSE3: Learning World-Space SE(3) at Every Pixel
- MOSES: Memory-Efficient LLM Optimization via Geometric Preconditioning and Hybrid Quantization
- MoT3DVG: A Benchmark for Outdoor 3D Visual Grounding with Motion-Aware Descriptions and Temporal Cues
- MOTIF: LLM-Defined Material Taxonomies for OOD Property Prediction via Invariant Feature Selection
- Motif-Mamba: network motif improved mamba for long-range sequence modeling: A Closer Look
- Motif Matching with Structure-Semantic Signatures for Robust CLIP Test-Time Adaptation
- MotionBench-XAI: A Benchmark for Manifold-Aware Shapley Attribution on Temporal Data
- MotionCFG: Boosting Motion Dynamics via Semantic Motion Sharpening
- Motion-Conditioned Prompt Guidance for 3D Human Pose Estimation
- Motion Deblurring via Trajectory-Aware Dynamic Weighting
- MotionFirst: An Autonomous Driving World Model Driven by Residual Motion Planning and Guidance
- Motion Forcing: Decoupling Ego and Object Motion via Sparse Inputs for Structured Video Generation
- MotionHalluc: Diagnosing Kinematic Hallucinations in Fine-Grained Motion Reasoning
- Motion is Structure: ZORA for Zero-Shot 3D Rigging from In-the-Wild 4D Dynamics
- MotionVLA: Vision-Language-Action Model for Humanoid Motion
- MoTok: Learning Structured Tokenization for Human Motion Representation
- M-plicits: Neural Implicit Surfaces via Nested Multiscale Residuals
- MPQ: A Message-Passing View of Post-Training Quantization
- MPSENet: Multi-Period Mean and Sample Entropy Guided Residual Connection Network for Time Series Forecasting
- MR-SRIF: A Scalable, Neural-Augmented Multi-Robot Square-Root Information Filter for Vision-aided Inertial Navigation
- MSC-Mol: Modality-Synergy Contrasting for Multimodal Molecular Representation Learning
- MSCR: Jointly Balancing Modality Utilization and Discovering Synergistic Information
- MSR-3D: Multi-Mode Semantic Representation for Open-Vocabulary 3D Scene Understanding
- MT-CC: Multi-Group Temperature Scaling for Asymmetric Calibration Behavior in Class-Incremental Learning
- MulGeCa: Anchoring Scattered Semantic Representations to Physical Entities for Open-Vocabulary Segmentation
- Multi-Agent AI Control
- Multi-agent Collaboration with State Management
- Multi-Agent Coordination via Support-Preserving Distillation
- Multi-Agent Intelligent Traffic Signal Control with Emergency-Aware Cooperative AI
- Multi-agent systems have some more to learn from human multi-agent systems
- Multi-Bridge Denoising Diffusion Probabilistic Models
- Multigroup Fairness and Omniprediction: Separations and Equivalences
- Multi-Hop Question-Answering via Triple Placeholder Resolution with Entity-Centric Summaries
- Multi-Hop Resolvent Attention: A Principled Generalization of Transformers for Compositional Reasoning
- Multi-Level Game Theory Framework for Automated Chest X-Ray Report Generation
- Multilingual Emotion Neurons in Large Audio-Language Models
- Multi-Marginal Inverse Optimal Transport for Contrastive Learning Via Explicit Anchor-Positive-Negative Coupling
- Multimodal-CL-Bench: A Benchmark for Multimodal Context Learning
- Multimodal Context-Aware Human Motion Generation with Language, Vision, and Object
- Multimodal Continuous Reasoning via Asymmetric Mutual Variational Learning
- Multimodal DeepFake Detection via Domain-Incremental Multi-Adapter Learning
- Multimodal Fine-tuning with Synthetic Captions
- Multimodal Irregular Time-series Analysis on Clinical Databases
- Multi-Modal Point Cloud Interpolation with Camera BEV Guidance
- Multimodal Prediction under Systematic Modality Missingness: Day-Ahead Bleeding Risk in ECMO Patients
- Multimodal Thinking with Renderable Programs
- Multi-Nonsmooth-Nonconvex-Objective Optimization
- Multi-Objective Causal Bandits: Minimal Intervention Space and Policy-Level Learning
- Multi-objective Hyperparameter Optimization with Expert Priors
- Multi-Objective {\it min-max} Online Convex Optimization
- Multi-Objective Reinforcement Learning for Large-Scale Tote Allocation in Human-Robot Collaborative Fulfillment Centers
- Multiphysics Bench: Benchmarking and Investigating Scientific Machine Learning for Multiphysics PDEs
- Multiple Importance Sampling for Policy Optimization in Reinforcement Learning
- Multi-Quantile Regression for Extreme Precipitation Downscaling
- Multi-Relaxation Initialization for Hybrid Search in Sparse Mixed-Integer Optimization
- Multi-Rollout On-Policy Distillation via Peer Successes and Failures
- Multiscale Euclidean Network Trajectories: Second-Moment Geometry, Attribution, and Change Points
- Multi-Scale Exposure Correction Network with Adaptive Feature Fusion
- Multi-Scale Graph Learning for Epilepsy Surgery Planning using Stereo-EEG
- Multiscale Log-Euclidean Diffusion for Geometry-Preserving EEG Covariance Generation
- Multi-Scale Structural Supervision for Time-Series Forecasting
- Multi-site PPG
- Multi-Source Multi-View (MSMV): A robust SSL framework for imbalanced data
- Multi-Stage Planning from Single-Stage Data: Reinforcement Learning Helps Composition but Requires Anchoring
- Multi-step Heterogeneous Causal Policy Learning for Personalized Education
- Multi-Stream Video Anomaly Detection with Semantic Threat Intelligence System
- Multi-Token Residual Prediction
- Multi-Turn RL Makes Small Language Model Competitive for Optimization Modeling
- Multi-View Multi-Level Prototypical Contrastive Heterophilic Hypergraph Learning
- Multi-View Vision-Language Reasoning with 3D Cognitive Maps
- Muon Dynamics as a Spectral Wasserstein Flow
- Muon is Not That Special: Random or Inverted Spectra Work Just as Well
- Muon+: Towards More Effective Muon via One Additional Normalization Step for LLM Pre-training
- MuscleMimic: A Physiological-Fidelity Benchmark for Full-Body Neuromuscular Imitation Learning
- MUSE: Memory-Guided Uncertainty-Aware Semantic Belief Optimization for Zero-Shot Embodied Search
- Muskie: Learning Geometrically Consistent Representations via Multi-view Masked Modeling
- Mutation Flows: Discrete Flow Matching over Edit Graphs
- MuteBench: Modality Unavailability Tolerance Evaluation for Incomplete Multimodal Fusion
- MUTE: Multi-Level Alignment Uncoupling Against Talking-Head Exploitation for Voice Protection
- Mutual Colormap Dependence: An Evaluation Protocol for Visualization Robustness in Scientific Vision-Language Models
- Mutual Information-Guided Corruption for Improved Self-Supervised Representation Learning in Tabular Data
- Mutual Information Guided Diffusion Model for Partial Label Learning
- MvFFN: Multi-view Floor-Plan Feed-Forward Network for Unposed Wide-Baseline Panorama Layout Reconstruction
- MVRA: Mixture-enhanced Vector-based Random Matrix Adaptation
- MV-WAM: Manifold-Aware World Action Model with Value Augmentation
- Myla: Maximizing Yield on Lean Hardware for Resource-Constrained MoE Inference
- MyoChallenge 2025: A New Benchmark for Human Athletic Intelligence
- MZ-Rain: Moisture-Budget-Informed Zero-Inflated Model for Precipitation Nowcasting
- NAMI: A Heuristic Forward Transfer Approach for Continual Reinforcement Learning
- NanoCIF: Autoregressive Generation of Alloy Nanoparticle Structures via Text-Based Representation
- Narrow Fine-Tuning Beyond LLMs: Cross-Modal Behavioral and Representational Drift
- Nash without Numbers: A Social Choice Approach to Mixed Equilibria in Context-Ordinal Games
- Native Symbolic Emergence Should Replace the Translation Layer Paradigm in Neuro-Symbolic AI
- Natural Language Actor-Critic: Scalable Off-Policy Learning in Language Space
- Natural-Language-Guided Protein Generation for Ligand-Binding Design
- Natural Language Is Flatland for Machine Intelligence: AI Systems Should Decouple Their Reasoning Substrate from Human Language
- Nature3D-AD: Geometry-Aware Feature Learning for Natural-Growth 3D Anomaly Detection
- Navigate the Unknown: Enhancing LLM Reasoning with Intrinsic Motivation Guided Exploration
- Near Cost-Optimal Best-Arm Identification with LLM Judges
- Nearest-Neighbor Radii under Dependent Sampling
- Near-Linear Time Generalized Sinkhorn Algorithms for Bounded Genus Graphs
- Nearly-Optimal Algorithm for Adversarial Kernelized Bandits
- Nearly Optimal Attention Coresets
- Nearly Optimal Bounds for Orthogonal Trace-Sum Maximization
- Nearly Optimal Fixed-Confidence Best-Arm Identification with 1-Bit Feedback
- Near-Optimal $\varepsilon$-Approximate Machine Unlearning for Smooth Strongly Convex Losses
- Near-Optimal Learning in Parametric Bandits with Action-Dependent Coarsened Feedback
- Near-Optimal Online Inventory Optimization on Downward-Closed Sets via Reduction to Delayed-Feedback Online Convex Optimization
- Near-Optimal Regret in Adversarial Kernel Bandits
- Near-Optimal Swap Agnostic Learning for all Proper Losses
- Neat Cross-Domain Representational Convergence via Frozen Embedding-Anchored Training
- Necessary and Sufficient Conditions for Autoencoder-based Style Transfer
- Negative Momentum for Convex-Concave Optimization
- Neglected Free Lunch from Post-training: Progress Advantage for LLM Agents
- Neighbor-Aware Snapshot-Based Temporal Graph Learning
- NEmo: Neuro-Symbolic Embodied Intelligence
- NeoWorld: Neural Simulation of Virtual Worlds via Progressive Object-Centric 3D Unfolding
- NeRD: Neural Replicator Dynamics for Maximum Clique Extraction
- Nereus: A Large-Scale Underwater Dataset for Fine-Grained Attribute Understanding and Grounded Counting Perception
- NestedVLA: Learning to Consolidate and Generate Skills for Vision-Language-Action Model
- NEST: Nascent Encoded Steganographic Thoughts
- NeSyGeo: A Neuro-Symbolic Framework for Multimodal Geometric Reasoning Data Generation (Extended)
- NeSyKC: Neurosymbolic Knowledge Compilation For Lifelong Learning Embodied Agents
- NetIG: Contrastive Information Gain for Scalable Step-Level Supervision
- NeuRA: Attribution-Aware Reasoning Agent for Trace-Guided Neuromorphic Architecture Design
- Neural Activation Functions via Kuhn–Tucker Stationarity: An Optimization-Theoretic Framework
- Neural Algorithmic Reasoning for Graph Saddle-Point Problems
- Neural Algorithmic Reasoning Must Explain When Neuralization Adds Value
- Neural Bridge Processes
- Neural Certificate Pricing for Combinatorial Optimization Problems
- Neural Chameleons: Language Models Can Learn to Hide Their Thoughts from Unseen Activation Monitors
- Neural Cluster First, Route Second: One-Shot Capacitated Vehicle Routing via Differentiable Optimal Transport
- Neural Collapse Curriculum for Imbalanced Time-Series Classification
- Neural Compression of Long ADMM Trajectory for Multiparametric Quadratic Program
- Neural Continuous-Time Markov Chain: Discrete Diffusion via Decoupled Jump Timing and Direction
- Neural Differential Utility: Costly Internal Reconfiguration in Adaptive Behavior and Neuroimaging
- Neural Equilibria for Long-Term Prediction of Nonlinear Conservation Laws
- Neural Fields for NV-Center Inverse Sensing
- Neural Garbage Collection: Learning to Forget while Learning to Reason
- Neural Mechanics: Model-Specific Response Atlases for Non-Destructive Multi-Track Steering in Large Language Models
- Neural Network Artifacts as a New Data Modality
- Neural Network Decomposition Using Discrete Cosine Transform
- Neural Network Memorization and Globajkl NTK Analysis
- Neural networks are more modular than single neurons suggest
- Neural Networks on Bounded Domains of Groups over Involutive Algebras
- Neural Operators via Geometric Laplace Bases
- Neural Quantum Spectral Operator Learning for Solving Partial Differential Equations
- Neural Refraction Fields for Image Verification
- Neural Scaling Laws for Order Flow Generation
- Neural-Schwarz Tiling for Geometry-Universal PDE Solving at Scale
- Neural Slack Variables for Shape Constraints
- Neural SOC: A Scalable Neural Framework for Stochastic Optimal Control Problems
- Neural Spectral Capacity: An Architectural Quantity from Network Specification Alone
- Neural Statistical Functions
- NeurIPS 2026 Workshop on Dynamic Alignment in Human-AI Coupled Systems
- NeurIPS 2026 Workshop on SaTQuML: Secure and Trustworthy Quantum Machine Learning
- NeurIPS 2026 Workshop on Tackling Climate Change with Machine Learning
- NeurIPS’26 Workshop on AI-Native Academia: Authorship, Peer Review, and Conference Governance under AI
- NeurIPS Should Mandate Interactive Sessions Between AI Researchers and Domain Experts
- NeuroAtlas: Benchmarking Foundation Models for Clinical EEG and Brain-Computer Interfaces
- NeuroAxisNet: Confound-Aware Multi-Program Representation Learning for Single-Cell Disease States
- NeuroCast: Neural Decoding Benchmarks for Naturalistic Speech in Human ECoG
- NeuroFuse: Complementarity-Guided Visual Supervision for EEG-to-Image Decoding
- Neuro-KE: Knowledge-Guided Interfaces for Semantically Grounded EEG Foundation Models
- Neuro-Mechanistic Latent Dynamics: Cognitive Load as a Generalizable Neural Stability Margin
- Neuromorphic Online Adaptation: Local Eligibility Traces for Test-Time Learning on Event-Driven Hardware
- Neuronal Identity as an Organizational Basis for Analyzing Neural Population Dynamics
- NeuronEye: Query-Guided Visual Concept Activation for Vision-Language Reasoning
- NeuroNTP - A Generalizable Multimodal Foundation Model for Epilepsy
- Neuroprobe: Evaluating Intracranial Brain Responses to Naturalistic Stimuli
- NeuroQA: A Large-Scale Image-Grounded Benchmark for 3D Brain MRI Understanding
- NeuroSentry: Monitoring Neuron Dormancy in Deep Neural Networks
- Neuro-Symbolic Convolutional Networks for Real-Time Object Detection
- Neuro-Symbolic Drive: Rule-Grounded Faithful Reasoning for Driving VLAs
- Neuro-Symbolic Forensic Reasoning for Open-Generator AI-Generated Video Detection
- Neuro-symbolic Syntactic Parsing: Shaping a Neural Network with the CYK Algorithm
- Neutral-atom quantum features as complementary structural encodings for graph learning
- New York Smells: A Large Multimodal Dataset for Olfaction
- NEX:Neuron Explore–Exploit Scoring for Label-Free Chain-of-Thought Selection and Model Ranking
- Next Embedding Prediction Makes World Models Stronger
- Next-Latent Prediction Transformers Learn Compact World Models
- Next-Token Prediction Enables Scalable Learning of Sleep Physiology
- NICE FACT: Diagnosing and Calibrating VLMs in Quantitative Reasoning for Kinematic Physics
- Nice Fold or Hero Call: Learning Budget-Efficient Thinking for Adaptive Reasoning
- NLE-Bench: Feasibility-Aware Tool Use in Symbolic Non-Linear Video Editing
- NOCE-Net: Representation Learning for Battery Operational Context via Nested Sequence Modelling
- No Contextualization Without Representation: In-Context Learning is Hindered by Covariate Shift to Low-Support Inputs
- No-Free-Fairness: Fundamental Limits and Trade-offs in Learning Systems
- No Free Pixels: Sharp Risk Bounds for Heavy-Tailed Residual Fields
- Noise-Regularized Training for Learned Image Compression
- Noisy2Latent: Nonparametric Deconvolution and Denoising via Maximum Mean Discrepancy
- Noisy Data is Destructive to Reinforcement Learning with Verifiable Rewards
- No More Guessing: a Verifiable Gradient Inversion Attack in Federated Learning
- No More, No Less: Task Alignment in Terminal Agents
- Non-asymptotic Convergence of Average-reward Q-learning with Options
- Non-Colliding Biometric Identities for Digital Entities: Geometry, Capacity, and Million-Scale Virtual Identity Provisioning
- Non-Conservative Sinkhorn: Optimal Transport with Dissipation
- Non-differentiable Regularization for Heavy-tailed Differentially Private Stochastic Convex Optimization
- Nonlinear SVD: A new approach to analyzing high dimensional nonlinear representations
- Non-Replacement Function Space Sampling for Bayesian Optimization
- No Patch Is an Island: Structure-Aware Test-Time Adaptation for Open-Vocabulary Semantic Segmentation
- Normalized Architectures are Natively 4-Bit
- NORMA: Norm-Guided Explanation Subgraph Discovery
- Norm or Direction? Decoding Vision Mambas for High-Resolution Vision
- NorSA: Accelerating LLM Decoding via Normalized Sparse Activation
- No Scale Left Behind: Multi-Scale Autoencoders with Bidirectional Attention for Time Series Anomaly Detection
- Not a Blank Slate: The Imprinted Moral Profile of Large Language Models
- Not All Attention Is Created Equal: Robustness of In-Context Learning Across Attention Mechanisms
- Not All Channels Are Equal: Perturbation-Invariant Channel Selection for Robust AI-Generated Image Detection
- Not All Layers Denoise Alike: Rethinking Adaptation for Continuous Diffusion Language Models
- Not All Nodes Are Created Equal: Homophily-Aware Stratification for Stable GNN Evaluation
- Not All Noise Is Harmful: Towards Perception Aware and Controllable RAW Image Joint Denoising and Demosaicing
- Not All Observations Are Equal: Guiding Derivative-Free Ensembles via Uncertainty Contraction
- Not All Positions Are Equal: Ordinal Asymmetry in Efficient Visual Autoregressive Model Training
- Not All Proofs Are Equal: Evaluating LLM Proof Quality Beyond Correctness
- Not All Refusals Are Equal: How Safety Alignment Fails Cybersecurity at Scale
- Not All Routes Should Be Equally Constrained: Token- and Layer-Aware MoE Fine-Tuning
- Not All Slots Are Equal: Non-Co-Progressive Markov Bridge for Bundle Construction
- Not All Tasks Quantize Equally: Fisher-Guided Quantization for Visual Geometry Transformer
- Not All Tokens Are Equal: Importance-Aware Masking for Discrete Diffusion Language Models
- Not All Tokens Are Equally Useful for Steering: Robust Directions and Prefix Steering
- Not All Turns Matter: Credit Assignment for Multi-Turn Jailbreaking
- Not Every Image Teaches Vision: Visual-Necessity-Gated Continual Learning for Multimodal Large Language Models
- Not Just Oversmoothing: Detecting Echo Chamber Effect in Graph Neural Networks
- Not Just RLHF: Why Alignment Alone Won't Fix Multi-Agent Sycophancy
- NoTVLA: Semantics-Preserving Robot Adaptation via Narrative Action Interfaces
- NoW-Net: Phase-Coherent Spectral–Structural Diffusion Learning enable Stealing Screen Content only from Wall Reflections
- npMCRL: A Deep Reinforcement Learning Based Solution for Non-Preemptive Makespan Multicoloring
- NREMP Removes Accidental Reflection Symmetry for Chiral Non-Reciprocal Dynamics
- NSDG-EEG: Nuisance-Stable Domain Generalization for MI-EEG Decoding
- NSW-EPNews: A News-Aligned Benchmark for Diagnosing Multimodal Electricity Price Forecasting
- NTILC: Neural Tool Invocation via Learned Compression
- NucEval: A Robust Evaluation Framework for Nuclear Instance Segmentation
- Nudging Beyond the Comfort Zone: Efficient Strategy-Guided Exploration for RLVR
- Nuisance-Induced Confounding in LLM Output Sensitivity Analysis: A Semantic Normalization Framework
- NV-Rep: Gene Embeddings and Contrastive Learning for Neuronal Vulnerability Modeling from Limited Single-Cell Data
- OASIS: Observation-Aware Simulation-Based Inference via Distributional Matching
- OASIS: Occupancy-Aware Selection of Informative Samples for KL-Regularized Alignment
- OASIS: Online Adaptive Steering for In-Training Safety of LLMs
- Object-Agnostic Contextual Domain Discovery for Repurposing Object-Centric Datasets
- Object-Based Curiosity for Reinforcement Learning
- Object Detection Benchmarks are Incomplete: The Role of Label Errors and Annotation Uncertainty
- ObsConDA: Observability-Constrained Data Assimilation with Control-Space Inference
- Observable- and Positional-Encoding-Dependent Symmetry Readout from Neural Network Weights
- OccamToken: Efficient VLM Inference with Training-Free and Budget-Adaptive Token Pruning
- OCR Recognition and Sensitive Information Intelligent Desensitization Method for Court Documents
- OCTOPUS: Optimized KV Cache for Transformers via Octahedral Parametrization Under optimal Squared error quantization
- ODDR: One-Step Deshadow Diffusion via Reward Guidance
- ODRPO: Ordinal Decompositions of Discrete Rewards for Robust Policy Optimization
- OenoBench: A Wine-Domain Benchmark for Knowledge-Grounded Evaluation of Large Language Models
- Offline and Online KL-Regularized RLHF under Differential Privacy
- Offline Constrained Reinforcement Learning under Partial Data Coverage
- Offline Inverse Reinforcement Learning with Unified Diffusion Planning
- Offline-Online Reinforcement Learning for Linear Mixture MDPs
- Offline Policy Evaluation via Mixed Bellman Residuals and Adaptive Critic Representations
- Offline Policy Optimization with Posterior Sampling
- Offloading Score: Measuring AI Reliance through Counterfactual Workflows
- Off-policy Learning with Excursion Policies
- OffQ: Taming Structured Outliers in LLM Quantization by Offsetting
- OGTDM: Operator-Guided Tri-Scale Dependency Modeling for Compressive Hyperspectral Imaging
- OISE: Orthogonal Intra-Tile Spatial Embedding for Grounded Document Understanding
- OLIVE: View-Augmented Latent Prediction with Waveform Reconstruction for Speech SSL
- OmicsBench: A Realism-First Benchmark for AI Coding Agents in Computational Biology
- OmicsLM: A Multimodal Large Language Model for Multi-Sample Omics Reasoning
- Omission Constraints Decay While Commission Constraints Persist in Long-Context LLM Agents
- OmniCap: Omnidirectional Ego-Exo Motion Capture in the Wild
- OmniDex: Scaling Dexterous Hand Grasping to Diverse Cluttered Scenes
- OmniDrop: Layer-wise Token Pruning for Omni-modal LLMs via Query-Guidance
- Omni-ECD: Omni Emotion Cause Derivation in Conversation
- OmniEgoCap: Camera-Agnostic Sequence-Level Egocentric Motion Reconstruction
- Omni-Geo: Full-Domain Geometry Benchmark with Multimodal Diagram Generation
- OmniGF: A Dual-Branch Vision-Language Framework for Unified Gaze Following
- OmniMemBench: Towards Scalable Evaluation of Long-Term Omni-Modal Agent Memory
- OmniMem: Scalable and Adaptive Memory Retrieval for Long Video Generation
- OmniMM: Biomechanically Aligned Language Motion Model
- OmniRAG-Agent: Agentic Omnimodal Reasoning for Low-Resource Long Audio-Video Question Answering
- OmniRefine: Alignment-Aware Cooperative Compression for Efficient Omnimodal Large Language Models
- OmniRefiner: Reinforcement-Guided Local Diffusion Refinement
- Omni-Safety under Cross-Modality Conflict: Vulnerabilities, Dynamic Mechanisms and Efficient Alignment
- OmniSimulator: Aligning Small Language Models for Authentic Heterogeneous Behavior Modeling
- OmniSpace: Efficient Geometry Awareness for Autonomous Vehicles MLLMs
- OmniToM: Benchmarking Theory of Mind in LLMs via Explicit Belief Modeling
- OmniVAL: Validated Chain-of-Thought Distillation for True Native Omni Models
- Once a Response, Always a Response: Detecting LLM-generated Text via Latent Prompt Restoration
- OncoJarvAIs: Trustworthy Agentic Orchestration of AutoML for Oncology
- On Concentration Inequalities for Sampling without Replacement
- On Contractive Compression in Federated Learning with Arbitrary Client Sampling
- On-Device Intelligence: Foundation Models under Real-World Constraints
- On Differentially Private Mechanisms for Linear Regression
- One Adapter, Many Tasks: Task-Conditioned Feature Transformations for Continual Learning
- One Algorithm, Every Input, Every Level: SANC — An Axiomatic Approach to Self-organizing Active Network of Concepts for Emulating Language Acquisition by Infant Intelligence
- One Editor, Many Edits: A Unified Training-free Framework for Diverse Video Edits
- One-Layer Transformers Provably Learn In-Context K-Nearest Neighbor Prediction with Chain-of-Thought
- One Model, Two Modes: Weight-Shared Small VLMs via Attention Mode Switching
- One Pass Is Not Enough: Recursive Latent Refinement for Generative Models
- One-Point Contraction: Erasing Representational Separability toward Irreversible Deep Forgetting
- One Question, Many Voices: Consistency Perturbation for Reasoning Stability in RLVR
- One Rotation to Align Them All: Representation Alignment in Geospatial Foundation Models
- One Scan is Enough: Demonstration-Free Adaptation for Language-Guided Navigation
- ONE-SHOT: Compositional Human-Environment Video Synthesis via Spatial-Decoupled Motion Injection and Hybrid Context Integration
- One-Shot Split Federated Learning under Class Imbalance
- One-Stage Fine-Grained Mask Generation from the Diffusion Denoising Process
- One-Step Distillation of Discrete Diffusion Image Generators via Fixed-Point Iteration
- One Step is Enough: Multi-Agent Reinforcement Learning Based on One-Step Policy Optimization for Order Dispatch on Ride-Sharing Platforms
- One Temperature to Rule Them All?
- One Token Per Frame: Reconsidering Visual Bandwidth in World Models for VLA Policy
- On High-Frequency Collapse in Spectral Vision Transformers
- On Inherent Privacy of Posterior Sampling: A Unified R\'enyi-Divergence Framework
- On Length Bias in EEG-to-Text Decoding
- Online Active Testing: Adaptive Importance Sampling for Unbiased Risk Estimation in Data Streams
- Online Algorithms with Unreliable Guidance
- Online Allocation with Differential Privacy
- Online Balanced Partitioning with Predictions for Ring Demands
- Online Bayesian Calibration under Gradual and Abrupt System Changes
- Online Bidding for Contextual First-Price Auctions with Budgets under One-Sided Information Feedback
- Online Clustering with Stochastic Arrivals
- Online Evaluation of LLMs via Dyadic Designs
- Online Finetuning Decision Transformers with Pure RL Gradients
- Online Learning-guided Learning Rate Adaptation via Gradient Alignment
- Online Learning via Learned Latent Bayesian Tracking
- Online Learning with Improving Agents: Multiclass, Budgeted Agents and Bandit Learners
- Online Planning for Continuous Robust POMDPs with Formal Guarantees
- Online Preference Learning via Generalized Linear Models: Convergence and Monotonicity of SGD
- Online Self-Training for Co-Adaptation in Hierarchical Diffusion Policies
- Online Semi-Infinite Linear Programming via Nonnegative Function Approximation
- On Lipschitz Explosion in Deep Neural Networks with Normalization: Consequences for Optimization and Robustness
- Only Humans Can See: Dynamic Visual-Persistence CAPTCHAs against Vision-Language Models
- Only Say What You Know: Calibration-Aware Generation for Long-Form Factuality
- On Monotonicity in AI Alignment
- On Normalization-Based Unconstraining
- On Observation Time for Recovering Latent Hawkes Networks
- On-Policy Hindsight Distillation for Early Risk Prediction
- On-Policy Self-Distillation with Sampled Demonstrations Reduces Output Diversity
- On the Adversarial Robustness of Discrete Image Tokenizers
- On the Blessing of Pre-training in Weak-to-Strong Generalization
- On the Burden of Achieving Fairness in Conformal Prediction
- On the Capability and Limitation of Hard Prompt
- On the Complexity of Discounted Robust MDPs with $L_p$ Uncertainty Sets
- On the Complexity of Preference-Based Bandits
- On the Conditional Validity of Conformal Prediction: a Loose but Universal Bound
- On the Construction and Implications of Low-Loss Valleys in LoRA-based Bayesian Inference
- On the Convergence of Split Federated Learning: A Masked Compositional Optimization Perspective
- On the Decomposition of Differentiable Games
- On the Design Space of Discrete Diffusion Online Adaptation for Molecular Optimization
- On the Efficiency of Structured Pruning in Small Language Model Pretraining
- On the Fragility of the Privileged-Signal Assumption in On-Policy Self-Distillation
- On the Generalization Gap of Streaming Federated Learning with Markovian Data
- On the Geometric Structure of Token--Token Interactions in Deep Language Models
- On the (Im)possibility of Sufficient Explanations
- On the Information Loss of Multi-Token Prediction: Origin and Solution
- On the Inherent Privacy Amplification of Missing Data
- On the Intrinsic Limited Robustness of Latent-Based Watermarking
- On the Learnability of Messages for Image Watermarking
- On the Limits of Candidate Generation in Biomedical Entity Linking
- On the Linear Representation of Graph Properties in Large Language Models
- On the Optimization of Minimum MMD Estimators
- On the Promise and Limits of Training LLM Search Agents in Fictional Worlds
- On the Quantum Advantage in Black-Box Adversarial Attacks
- On the Ranking of Human Pose Confidence
- On the Recoverability of Causal Relations from Bulk Gene Expression Data
- On the Relaxation of Conditional Independence Assumption for Image Segmentation
- On the Robustness of Multilingual Text Embedding Rankings Across Learning Tasks, Languages, and Benchmark Datasets
- On the Robustness of Watermarking for Autoregressive Image Generation
- On the Role of Intermediate Representations in Knowledge Distillation for Robust Generalization
- On the Role of Preference Variance in Preference Optimization
- On the Stability of Spherical Hellinger-Kantorovich Flows and Their Implications for Differential Privacy
- On the Superlinear Relationship between SGD Noise Covariance and Loss Landscape Curvature
- On the Token Value Inequality in Efficient Reasoning
- OntoPlan: An Ontology-Grounded Scene Representation and Agentic Framework for Scalable Robot Task Planning
- On What We Can Learn from Low-Resolution Data
- OpenBrain: An Auditable Generated-Label Release for Whole-Brain MRI Parcellation
- OpenClaw-RL: Train Any Agent Simply by Talking
- Open, Collaborative, and Decentralized Training of Foundation Models
- Open-Ended Task Discovery via Bayesian Optimization
- Opening-Price Shortcut in Stock Prediction: A Conditional Opening Prior-and-Evidence Framework
- Opening the Black Box of Classifier-Free Guidance via Information Bottleneck
- OpenInterconnect: An Open Dataset and Benchmark for Surrogate Modeling of Die-to-Die Interconnect S-Parameters
- OpenPocket3D: Open-Ended 3D Instance Segmentation via Instance-Wise Pocket Vocabulary
- Open-Set Cross-Network Node Classification via OOD-Guided Selective Alignment
- Open-Set Source-Free Object Detection via Spectrum Aware Dual Teacher
- Open-Weight LLM Fine-Tuning Defenses are Susceptible to Simple Attacks
- Open-World Evaluations for Measuring Frontier AI Capabilities
- OPERA: Accelerating Private Inference for Transformers by Exploiting Oblivious Early-Exit
- OPERA: An Agent for Image Restoration with End-to-End Joint Planning–Execution Optimization
- Operation-Guided Progressive Human-to-AI Text Transformation Benchmark for Multi-Granularity AI-Text Detection
- Operator Projection for Depth-Robust Optimization in Factored Networks
- OPT 2026: Optimization for Machine Learning
- Optimal Bidding via Loss Function Design: A Generalized Value Maximization Framework
- Optimal Brain Decomposition for Accurate LLM Low-Rank Approximation
- Optimal Byzantine-resilient Federated Learning with User-level Differential Privacy
- Optimal Convergence Analysis of DDPM for General Distributions
- Optimal Counterfactual Search in Tree Ensembles: A Study Across Modeling and Solution Paradigms
- Optimal Crystal Flow for Fast and Reliable Crystal Generation
- Optimal Data Injection: A Stability-Based Framework for Resource-Efficient Augmentation
- Optimal Explicit and Implicit Regularization Strengths in High-Dimensional Continual Linear Regression
- Optimal hypersurface decision trees
- Optimality in Decentralized Optimization under Bandwidth Constraints
- Optimality of Sub-network Laplace Approximations: New Results and Methods
- Optimal Learned Bloom Filter Design: One Backup Filter to Rule Them All
- Optimal Learning-Augmented Algorithm for Online Bidding
- Optimal Post-Training Quantization Scales and Where to Find Them
- Optimal Projection-Free Adaptive SGD for Matrix Optimization
- Optimal Rates for Pure $\varepsilon$-Differentially Private Stochastic Convex Optimization with Heavy Tails: A Closer Look
- Optimal Risk Bounds of Stochastic Gradient Descent for Shallow ReLU Networks
- Optimal Subgroup Discovery at Every Support Threshold
- Optimal Transport and Nuclear-Norm Minimization
- Optimistic Multi-Step Value Estimation via Implicit Values($\tau,\lambda$)
- Optimistic Q-value Adaptation for Offline-to-Online Reinforcement Learning
- Optimization Dynamics in Pruned Networks with Sharpness-Aware Minimization
- Optimization via Greedy Oriented Progression
- Optimized Minimal 4D Gaussian Splatting for Efficient Dynamic Scene Representation
- Optimize Once, Execute Fast: Latency-Aware Multi-Agent Workflow Learning for Recurrent Queries
- Optimizer-Induced Mode Connectivity: From AdamW to Muon
- Optimizer-Model Consistency: Full Finetuning with the Same Optimizer as Pretraining Forgets Less
- Optimizer selection based on function-proxy fidelity
- Optimizing Computational-Statistical Runtime for Wasserstein Distance Estimation
- Optimizing Curricula for Human Visual Category Learning Via Surrogate Learners
- Optimizing Diagonal Rescaling for Joint Weight–Activation Quantization: A Closer Look
- Optimizing Expected Utility in Multi Objective Reinforcement Learning
- Optimizing Mixtures of Public Datasets can Significantly Improve Private Learning
- OptimThink: Measuring Optimization Reasoning Capability with Extremal Problems
- Opt-Judger: Rigorous Evaluation of LLM-Based Optimization Modeling via Feasibility Verification
- OPT-Zero: Data-Free Fine-Tuning of Language Models for Optimization Modeling
- ORACLE: Anticipating Scams from Partial Trajectories in Streaming App Usage
- ORBIT: A Framework for Multi-Agent Security Evaluations
- ORBIT-OFT: Bottleneck-Aware Latent Reasoning and Optional Reinforcement Learning for Vision-Language-Action Models
- ORCA: Hunting Compositional Failures in Text-to-Image Diffusion
- Order-Agnostic Autoregressive Modelling with Missing Data
- Order-based structure learning for zero-inflated count data under data heterogeneity
- Ordering Asymmetry in Splitting Methods via Commutator Flow Transport
- Order-Optimal Baselines for Noise-Limited Model Selection
- Ordinal Geometry Complements Reconstruction: Diagnosing Planning with Compressed Value Functions
- Ordinal Preference Learning for Language Models
- OrScale: Orthogonalized Optimization with Layer-wise Trust Ratio Scaling
- Orth-Dion: Eliminating Geometric Mismatch in Distributed Low-Rank Spectral Optimization
- Orthogonal Ensembles in Prediction Space for Improved Uncertainty and Calibration
- Orthogonal Inference for Local Pairwise Interactions of Continuous Exposures
- Orthogonalized Activation Steering
- Orthogonal Origin Parking: Decoupling Lorentz Manifolds for Robust OOD Generalization
- OrthoPhys: Physically Plausible Video Generation with Orthogonal-View Geometry Guidance
- OrthoReg: Orthogonal Regularization for Hybrid Symbolic-Neural Dynamical Systems
- ORZO: Oracle-guided Reasoning via Zeroth-order Optimization
- OSAM-DGS: Operational, Structured and Annotated Multimodal German Sign Language Dataset for Machine Learning
- OSDN: Improving Delta Rule with Provable Online Preconditioning in Linear Attention
- OTDQ: A Physics-Guided Framework for Blind Dehazing Quality Assessment via Optimal Transport Regularity
- OTformer: Non-Stationarity-Aware Adaptive Optimal Transport Attention for Time Series Forecasting
- OT-MIMIC: Scaling Medical Image Segmentation with Missing Modalities via Optimal Transport
- OTROPE: Optimal Transport-based Robust Off-policy Evaluation for Large Language Models
- OTSS: Output-Targeted Soft Segmentation for Contextual Decision-Weight Learning
- Outcome-Based RL Provably Leads Transformers to Reason, but Only With the Right Data
- Outcome Feedback Calibrates LLMs for Individual Behavioral Prediction
- Outcome-Guided Attention Exploration for Weakly Supervised Whole-Slide Image Analysis
- Outlier-Robust Variational State Estimation of Model-free Process
- Out of Context: Reliability in Multimodal Anomaly Detection Requires Contextual Inference
- Out-of-Distribution Detection via Channelwise Feature Aggregation in Neural Network–Based Receivers
- Out-of-Distribution Generalization of Risk Aversion in Language Models (Extended)
- Out-of-Distribution Spatiotemporal Forecasting: A Federated Causal Meta-Hypernetwork Approach
- Out-of-Sample Generalization for Empirical Risk Minimization over Fixed Designs via Integral Probability Metrics
- Overcoming Imbalanced Bias in Neural Predictor with Subsampled Distribution-Aware Contrastive Regularizer
- Overcoming Kernel Redundancy for Scaling Logic Gate Networks
- Overcoming Rank Collapse in Feedback Alignment
- Overcoming the Communication-Performance Tradeoff in LLM Pretraining
- Overconfidence in Crisis: Trajectory-Aware Risk Gating for Safe Sequential Recommendation
- OvOSkills
- P$^{3}$: Joint Program-and-Proof Planning\\ for Verified Code Generation
- P$^3$-VLM: A Point-based Alternative for Grounded 3D Vision-Language Models
- PAAC: Privacy-Aware Agentic Device-Cloud Collaboration
- PACE-dLLM: Elastic Block Decoding via Confidence Cliff Estimation for Diffusion Language Models
- PACE: Pareto-Adaptive Compression for Efficient Native MLLMs
- PACE: Phase-Aware Chunk Execution for Robot Policies with Action Chunking
- PACE: Preserving Attention via Clustered Evictions for Long-Context LLM Inference
- Pacing Branch Parallelism in LLM Serving
- PACR: Progressively Ascending Confidence Reward for LLM Reasoning
- Paint Anything: Toward Any-Color Controllable Image Generation and Editing
- PAINT: Partial-Solution Adaptive Interpolated Training for Self-Distilled Reasoners
- PAI: Plasticity-Anchored Initialization for Improved Stability-Plasticity Trade-offs in Continual Learning
- PAIR: Pair-Aware Inference-Time Refinement for Fast RNA Inverse Folding
- Pair Recurrence: Rethinking Negative Sampling for Temporal Link Prediction
- PALACE: Adaptive-Landmark Kernels for Certified Persistence Classification
- Paleoinspired Vision: From Exploring Colour Vision Evolution to Inspiring Camera Design
- PALIN: Pixel-Aligned LiDAR--NIR Multimodal 3D Object Detection with Benchmark
- PALMU: Correcting the Hidden Cost of Forgetting in Self-Training Models
- PAMFormer: Propagation-Aware Multimodal Pretraining for Radio Map Estimation
- PAMNet: Cycle-aware Phase-Amplitude Modulation Network for Multivariate Time Series Forecasting
- PAMod: Modeling Cyclical Shifts via Phase-Amplitude Modulation for Non-stationary Time Series Forecasting
- PANDO: Efficient Multimodal AI Agents via Online Skill Distillation
- Pandora-RAG: Optimal Stopping for Multi-Hop Retrieval-Augmented Generation
- PanelTS: A Multi-Domain Benchmark Dataset for Panel-Based Time Series Forecasting
- PanoHK360: A Large-Scale 8K Urban Panoramic Dataset and Benchmark for Depth Estimation
- Panoptic Scene Program Diffusion Transformer
- PanoWorld: Towards Spatial Supersensing in 360◦ Panorama World
- PaperFit: Vision-in-the-Loop Typesetting Optimization for Scientific Documents
- PapoDrive: Rethinking RL Post-Training for Autonomous Driving VLA Models From Reasoning Pattern-Aware Policy Optimization
- PAPO-VLA: Planning-Aware Policy Optimization for Vision-Language-Action Models
- ParadigmCoT: Paradigm-Guided Chain-of-Thought in Large Language Models via Policy Optimization
- Paradoxes of Game Theoretic Equilibria and Price of Anarchy
- Parallax: Parameterized Local Linear Attention for Language Modeling
- PARALLAX: Separating Genuine Progress from Benchmark Artifacts in Hallucination Detection
- Parallel Computation Algorithms and Convergence Guarantees for Mean-Field Langevin Dynamics
- Parallelism Is Not Free: The Parallel–Sequential Contradiction in Diffusion Language Model Reasoning
- ParallelKernelBench: Can LLMs Write Fast Multi-GPU Kernels?
- Parallel Reconstruction for Large-scale Network based on Compressive Sensing
- Parallel Recursive LSTM
- Parallel Rollout Approximation for Pixel-Space Autoregressive Image Generation
- Param2Task: Parameterized Task-Graph Generation for Search Agents
- Parameter-Efficient Subspace Optimization for LLM Fine-Tuning
- Parameters as Agentic Memory: Internalizing Long-Horizon Memories for Efficient LLM Agents
- Parameter symmetries determine representational geometry in overparameterized nonlinear networks
- Parametric Memory Decoding for Zero-Shot Routing in LoRA-Based External Parametric Memory
- PARA: Paraphrase Agreement as a Recovery Signal for Verifier-Free Reasoning
- Parasite Features: Causal Abstraction Through Spurious Pathways
- Parasitic-Light-Aware Low-Light RAW Image Enhancement
- ParBench: A Benchmark for Reliable Evaluation of LLM Parallel Code Translation
- Pareto-GRPO: Multi-Objective Group-Relative Optimization for Efficient Search-Augmented Reasoning
- ParetoM$^3$: Learning on the Pareto Set under Preference Guidance via Min-Max-Min Optimization
- ParetoSlider: Diffusion Models Post-Training for Continuous Reward Control
- PARI: Policy-Driven Active Residual Intervention for Weakly Supervised Point Cloud Segmentation
- Parity, Sensitivity, and Transformers
- ParliaBench: A Topology-Controlled Benchmark for Multi-Agent Scientific Drafting
- PAR-Net: Periodic Autoregressive Network for Cyclostationary Time Series Forecasting
- Partial AUC Maximization from Positive-unlabeled Data
- Partition Invariance in Generalized Spectral Clustering: Measure-Parametrized Stability Theory
- Partition-then-Prune: Accelerating VLLMs via Anisotropic Octree Partitioning and Sub-Supermodular Pruning
- Passing the Audit, Failing the Person: Digital Dignity Requires Non-Compensable Evaluation
- PaSTel: Anchoring Histology in Spatial Transcriptomics via Multi-Scale Hierarchical Bio-Prior Contrastive Pretraining
- Patch4Patch: Restoring Structural Connectivity in Patch-based Vision Encoders
- PatchBench: Measuring Collateral Damage in Activation Patching
- Patch-Conditioned Modality Routing: Beyond Scalar Gates for Cross-Modal Image Fusion
- PATCH-GS: Localized Post-hoc Correction for Dynamic Gaussian Splatting
- Patch Hierarchical Attention Transformers for Efficient Particle Jet Tagging
- Patching LLMs Like Software: A Lightweight Method for Improving Safety Policies in Large Language Models
- PatchKV: Weight Space Compensation of KV Cache
- PATCH: Preference-Adaptive TPOT-Constrained Path Learning for LLM Pipeline Placement over Decentralized GPU Networks
- Patch Rebirth: Fast and Transferable Model Inversion of Vision Transformers
- Path Dependence under Adaptive AI Delegation
- PathISE: Learning Informative Path Supervision for Knowledge Graph Question Answering
- PathMem: Toward Cognition-Aligned Memory Transformation for Pathology MLLMs
- PathNavigate: A Training-Free Pathology Agent with Surprise-Guided Scan and Shared Slide Memory for Whole-Slide VQA
- PathoSage: Towards Multi-Source Evidence Adjudication in Pathology via Experience-Aware Agentic Workflow
- Path-R1: Reinforcement Learning over OrderedKnowledge Paths for Retrieval-Augmented Generation
- PathReportEval: A Systematic Benchmark for Pathology Report Generation
- Path-Space Mirror Descent for On-Policy Reinforcement Learning under the Generalized Schrödinger Bridge
- PathSpeed: Efficient Flow Matching via Path-Aware Time Sampling
- Patients With Personality: Realistic Patient Simulation through Controlled Diversity and Selective Disclosure
- PATROL: Defending LLMs Against Jailbreak Attacks via Periodic Safety Monitoring and Token Rollback
- Pause and Reflect: Conformal Aggregation for Chain-of-Thought Reasoning
- PAWS: Perception of Articulation in the Wild at Scale from Egocentric Videos
- Paying More Attention to Competing Keys for Binary KV-Cache Attention
- PCBInnoBench: Benchmarking LLM Agents on Real-World PCB Design
- PCEval: A Benchmark for Evaluating Physical Computing Capabilities of Large Language Models
- PCS-FNO: A PDE-Coupled Fourier Neural Operator for Small-Data Multiphysics Simulation
- PDAGENT-BENCH
- PDEAgent-Bench: A Multi-Metric, Multi-Library Benchmark for PDE Solver Generation
- PDE-JEPA: Semigroup-Consistent Latent World Models with Soft Physical Regime Decomposition
- PDF-HR: Pose Distance Fields for Humanoid Robots
- PDHFormer: Progressive Dual-Head Transformer for Behavioral Choice Prediction
- PEARL: Prediction Error-Adaptive Regularized Learning for Selective Forgetting under Partial Preference Drift
- PEARL: Probabilistic Evidential Attention for Reliable Long-term Time Series Forecasting
- PEARL: Solver-in-the-Loop Interactive Optimization Modeling from Natural Language
- PEBS: Per-rater Empirical-Bayes Shrinkage for RLHF Reward-Model Calibration
- PEEK: Context Map as an Orientation Cache for Long-Context LLM Agents
- PEEK: Predictive Queue-Informed KV Cache Management for LLM Serving
- PEEK Variance: An Entropy-Inspired Signal for Representation Analysis, Pruning, and Error Detection
- Peer Review of Applied Machine Learning Papers Must Include Code Execution
- Peer Review Should Be at a Price
- Peer review should constrain evaluative authority
- PE-GRPO: Aligning Video Generation for Physical and Embodied Reasoning via World Model
- PEML: Parameter-efficient Multi-Task Learning with Optimized Continuous Prompts
- Penalty-Based Online Convex Optimization with Adversarial Constraints and Adaptive Dynamic Regret
- PEP-CI: Pre-Emptive Prediction of Image Quality & Compositional Failure from Attention Dynamics
- PepDDG: Peptide–Protein Binding ΔΔ𝐺 Prediction via Information Channel Decomposition
- PepSpecBench: A Unified Evaluation Benchmark for Peptide Tandem Mass Spectrometry Prediction
- Perceive, Route and Modulate: Dynamic Pattern Recalibration for Time Series Forecasting
- Perception, Not Reasoning, Limits Visual Theory of Mind
- PercepT: Model-Side Perception Tests for Risk-Aware Quality Estimation under Real-World Covariate Shifts
- Perceptrons That Attend: Transformers Inside MLPs
- Performance-Driven Policy Optimization for Speculative Decoding with Adaptive Windowing: A Closer Look
- Performative Prediction with Selective Labels
- Period-Conditioned Residual Eventification for Spiking Time-Series Forecasting
- Permanent and Transient Representations for Continual Reinforcement Learning
- PermaVid: Consistent Video Generation Across Edits via Disentangled Context Memory
- PermuQuant: Lowering Per-Group Quantization Error by Reordering Channels for Diffusion Models
- Permutation Equivariant Neural Networks for Antisymmetric Tensors
- Permutation-preserving Functions and Neural Vecchia Covariance Kernels
- Permutation Sensitivity in t-SVD-based Multi-view Clustering
- Per-Sample Routing of Pre-trained Encoders via Embedding Typicality
- Persistent Exploration Theory: A Finite-State Control Theory of Discovery
- Persistent-Transient Policy Evaluation for Markov Chains via Minimal Peripheral Quotients
- Persistent Visual Memory: Sustaining Perception for Deep Generation in LVLMs
- PersonaBEAM: A Controlled Benchmark and Dataset for Persona Prompting in Vision-Language-Action Agents
- Personal Camera Roll Visual Question Answering
- Personalized, Aligned, Long-Term Memory for AI Systems (PALM) Workshop
- Personalized Safety with Continuous Monitoring for Agentic Systems
- Personalize-then-Store: Benchmarking and Learning Personalized Memory for Long-Horizon Agents
- Personalizing GUI Agents with Human-like Contrastive Interaction Trace
- Persona-Model Collapse in Emergent Misalignment
- Persona Topology: Latent Persona-Vector Projection Can Misreport Behavioral Persona Expression
- PerSPL: Personalized Semi-supervised Preference Learning
- Persuasion Attacks Can Decrease Effectiveness of CoT Monitoring
- Persuasive Prediction via Decision Calibration
- Perturbation Prediction Models Fail Out-of-Distribution: A Rigorous Benchmark and Prescriptive Framework
- PE-SHAP: Causally Interpretable Path-Wise Shapley Explanations
- Pessimistic Latent Task-aware Optimization for Robust Offline Meta-Reinforcement Learning
- PETS: Inference-Time Differentially Private Synthetic Time Series Generation
- PFE: Personalised Flexible Encryption for Privacy-Preserving Multi-Party Collaborative Inference
- PHANTOM: A Large-Scale Dataset of Multimodal Adversarial Attacks for Vision-Language Models
- PHANTOM: Phase Aware Neural Network Quantization Using Random Hadamard Transform
- Phase-Coded World Model for Spatial Memory, Imagination, and Localization
- Phase-DGS: Phase-Guided Dynamic Gaussian Splatting from Unsynchronized Multi-view Video
- PhaseLoRA: Control-Regime-Conditioned Low-Rank Adaptation for Continuous-Action Vision-Language-Action Policies
- Phase Transitions in Large Language Models: A Scaling Behavior for Inference-Time Stability
- Phase-wise MLLM Tuning for Multi-framework WebUI Code Generation
- PHIONet: Port Hamiltonian Inertial Odometry Network
- PhishNChips: A Benchmark for Evaluating LLM Email Agent Security Under Deployment Configuration Variation
- PhLEx: A graph-based pharmacogenomic framework via latent protein embeddings
- Phoenix: Checkpoint-less Failure Recovery for Auto-parallelism
- PhyGround: Benchmarking Physical Reasoning in Generative World Models
- PhyLatent: Physics-Aware Latent Dynamics with Neural Assimilation for Reconstruction and State Estimation
- Phylogenetically-Guided Data Augmentation for Data Efficient Fine-Tuning of RNA Language Models
- Phy-MamPose: Bridging the Gap between Mamba and 3D Human Pose Estimation
- PhyMo: Learning Physical Dynamics with Accurate and Continuous Motion from Multi-View Videos
- PhyProbe: Rethinking Physical Consistency Evaluation in Generated Videos
- PhysCons: Physics-Aware Consistency Modeling for Robust AI-Generated Image Detection
- PhysEditBench: A Protocol-Conditioned Benchmark for Dense Physical-Map Prediction with Image Editors
- PhysEval: Quantifying the Gap Between Video Generation and World Physical Laws
- PhysFlow: Physics-Intrinsic Velocity Regularization for Motion-Intensive Video Generation
- PhysGraphNet: Physical-State Scene Graphs via Latent Graph Reasoning and Counterfactual Supervision
- PhysGraph: Point Graphs for Physical Scene Understanding
- Physical Fidelity Reconstruction via Improved Consistency-Distilled Flow Matching for Dynamical Systems
- Physical Information Bottleneck: A New Paradigm for Extremely Robust Cross-Modal Topological Representation Learning
- Physically Viable World Models: A Case for Query-Conditioned Embodied AI
- Physical System Understanding Requires Locally Adapted World Models
- Physical Understanding for Decision-Making: Bridging Foundation Models and Reliable Agents
- Physical World Model
- Physics-Aligned Neural Modeling of Ocean Dynamics
- Physics-Aware Test-Time Scaling for Video Generation via Motion-Decomposed Rewards
- Physics-Constrained Underwater Image Enhancement via Polarization-Based Scattering Disentanglement
- Physics Derived Plausibility Metrics for 3D Human-Human Motion Generation
- Physics-Guided Refractive Ray Optimization for Transparent Object Reconstruction
- Physics-Induced Koopman Learning for Mesh-Based Dynamics
- Physics-Informed Latent Forcing for MRI Super-Resolution via Differentiable Bloch Equation Constraints (Extended)
- Physics-Informed Neural PDE Solvers via Spatio-Temporal MeanFlow
- Physics-informed operator priors universally improve scientific Bayesian optimization
- Physics-Informed Tracking (PIT)
- Physics-Informed Video Generation via Mixture-of-Experts Latent Alignment
- Physics-Mediated Diffusion: Leveraging Intermediate Field Representations for Sparse PDE Inversion
- Physics-R1: An Audited Olympiad Corpus and Recipe for Visual Physics Reasoning
- Physics Unrolled Neural Operator for Wireless Field Modeling
- Physiology-Constrained Learning: Physiological Laws as Pretraining Constraints for Invariant Representations
- PhysioSeq2Seq: A Hybrid Physiological Digital Twin and Sequence-to-Sequence LSTM for Long-Horizon Glucose Forecasting in Type 1 Diabetes
- PhysProg: Physics-Guided Progress Learning for Dexterous Garment Manipulation
- PhysRemover: A Unified Framework for Physically Realistic Object Removal
- PhysTC: A Physics-Enhanced Dataset and Architecture for High-Precision Tropical Cyclone Forecasting
- PiD: Fast and High-Resolution Latent Decoding with Pixel Diffusion
- PI-NAIM: Path-Integrated Neural Adaptive Imputation Model
- PipeFSDP: Efficient Pipeline Parallel under Fully Sharded Data Parallel for Large Language Model Training
- \(\pi\)-StepNFT: Wider Space Needs Finer Steps in Online RL for Flow-based VLAs
- PitchBench: Measuring Pitch Hearing in Audio-Language Models
- PIT-GCL: Protein Interaction using Topological Graph Contrasitive Learning
- PIVOT: Bridging Planning and Execution in LLM Agents via Trajectory Refinement
- PIVOT: State-Prior Visual World Models for Sample-Efficient Control
- PixelGen: Improving Pixel Diffusion with Perceptual Loss
- PixelPrune: Pixel-Level Adaptive Visual Token Reduction via Predictive Coding
- Pixels to Tokens: Token Space Efficient Active Learning for Low-Budget Semantic Segmentation
- PixFoundation 2.0: Do Video Multi-Modal LLMs Use Motion in Visual Grounding?
- PixSearch: Region-Grounded Retrieval for Knowledgeable Large Multimodal Models
- PixVerve: Advancing Native UHR Image Generation to 100MP with A Large-Scale High-Quality Dataset
- PLACE: Patch-Level Agnostic Concept Extraction
- Plan4D: Generative Plannable 4D Worlds
- Plan, Decompose, and Judge: Reasoning Agents for Zero-Shot Composed Image Retrieval
- Planner-Calibrated Diagnostics for Latent World-Model Benchmarks
- Planner-in-the-Loop Repair of LLM-Generated Heuristics for Classical Planning
- PlanningBench: Generating Scalable and Verifiable Planning Data for Evaluating and Training Large Language Models
- Planning-Grounded Risk Prediction with Structured Counterfactual Risk Explanations for Autonomous Driving
- PlatoLTL: Learning to Generalize Across Symbols in LTL Instructions for Multi-Task RL
- Platonic Representations in the Human Brain: Unsupervised Recovery of Universal Geometry
- Playing games with knowledge: AI-Induced delusions need game theoretic interventions
- Playing Markov Games Without Observing Payoffs
- Playing with Fire : What Transfers When RL Trains a Language Agent?
- PLDesign: Steerable Target-Binder Co-denoising in a Unified Complex Latent Space
- PM-LoRA: Scalable Continual Learning via Progressive Merging of Low Rank Adapters
- PointGS: Open-Vocabulary 3D Scene Understanding from Pure Point Clouds
- Pointwise Convergence in Games with Conflicting Interests
- Pointwise Detection of Well-Trained Regions in Physics-Informed Neural Networks
- Pointwise Metrics Mislead: An Evaluation Protocol for Multimodal Inverse Problems
- PointZero: 3D Point Track Completion for Learning Transferable 3D Dynamics
- POISE-Bench: A Benchmark for Evaluating Information-Seeking Strategies Under Partial Observability and Budget Constraints
- POISE: Instance-Specific Prompt Tuning under Latent Mixture Target Distributions
- Poisoning Attacks on LLMs Require a Near-constant Number of Poison Samples
- Poisoning What Models Already Know: Belief Corruption under Continual Pre-Training
- Poison-then-Hide: Finetuning-Activated Backdoor Attack on Pretrained Vision Encoders
- Poisson Empirical Bayes via Gamma-Smoothed Nonparametric Maximum Likelihood
- Poker-AD: Fast Opponent Exploitation by Distillation in Imperfect-Information Games
- POLAR-Bench: A Diagnostic Benchmark for Privacy-Utility Trade-offs in LLM Agents
- PolarVLM: Bridging the Semantic-Physical Gap in Vision-Language Models
- PolicyAlign: Direct Policy-Based Safety Alignment for Large Language Models
- Policy as Data: Replay Based Dual Averaging via Advantage Regression
- Policy as Generative Verifier: A Single-Stream RL Framework for LLM Self-Correction
- Policy-Calibrated Conformal Prediction for Adaptive Test-Time Compute
- Policy Learning under Distributional Shift: Robustness Guarantees via Invariant Reward Models
- Policy Regret Minimization in Partially Observable Markov Games
- POLLINATOR: Frugal Model Routing for Cost-Efficient Completion with Quality Guarantee
- PolyMind: Exploring Width Scaling for Reflective Reasoning in Language Agents
- Polynomial-Time Algorithm for Thiele Voting Rules with Voter Interval Preferences
- PolySplat: Workload-Regime-Aware Rasterization for 3D Gaussian Splatting
- PoMA: Reading Hidden Screens from a Single Wall Spot via Photometrically Marginalized vASM Looping
- POME: Post Optimization Model Edit via Muon-Style Projection
- PooFit: Confidential Auditing of Model Overfitting
- PO-PDDL: Learning Symbolic POMDPs from Visual Demonstrations for Robot Planning Under Uncertainty
- PopuLoRA: Co-Evolving LLM Populations for Reasoning Self-Play
- PoseBridge: Bridging the Skeletonization Gap for Zero-Shot Skeleton-Based Action Recognition
- Pose-Free Feed-Forward 3D Inpainting via Learnable Mask Attention and Support Token Refinement
- PosePlaner: Denoising Any Feedforward Pose Predictor from Pairwise Planar Geometry
- Position: Agentic AI Will Make the Gig Economy Less Fair Unless Audited and Regulated
- Position: Agent Verification Requires Typed Workflows and Bounded Commitments
- Position: AI-Agent Pricing Should Become More Outcome-Dependent: An Economic Perspective
- Position: AI Conferences Are Under Strain, Unbundling as a Testable Reform
- Position: AI Development Should Prioritize Cognitive Security
- Position: AI Should Verify, Not Judge, Scientific Work
- Position: Beyond Model-Centric Prediction—Agentic Time Series Forecasting
- Position: Decoupling Semantic Reasoning and Physical Grounding for Controllable and Versatile 3D Assets Generation
- Position-Dependent Forward Processes are CFG-Amplified Inductive Biases for Conditional Discrete Diffusion
- Position: Researchers Should Help Redirect Power Over AI Decision Support and Facial Recognition Systems Used by State Institutions Back to the Public
- Position: Task-Triggered Memory Evolution Destroys Memories with High Long-Term Utility
- Position: The Goal is Proactive Personalization, NOT User Replication.
- Position: We Need Greater Transparency to Maintain Research Pipeline Reliability Despite GenAI
- Posterior Alternative Calibration in Ambiguous Inverse Problems
- PosteriorBench: From Point Estimates to Posterior Matching in Evaluating Generative Inverse Solvers
- Posterior Continuation with Noise-Conditioned Frequency Exposure for Diffusion Inverse Problems
- Posterior Inference in Latent Space for Scalable Constrained Black-box Optimization
- Post-Outcome Failure Triage in World-Model RL: Auditing a Realized-Shortfall Selector Under Matched Late-Update Controls
- POST: Progressive Object-Slot Tokenization for Multimodal Large Language Models
- Post-Training Architectural Robustness via High-Rank Random Lifting
- Post-Training Quantization with Gradient-Projected Fisher Approximation for Vision Transformers
- Pot of Gold: Efficient and Performant Evolutionary Policy Mixtures for Deep Reinforcement Learning
- Power in Liquid Democracy: A Network Centrality Approach
- PPC-Flow: Paired Posterior Correction for Compute-Efficient Flow-Matching Inverse Problems
- PPI2Text: Captioning Protein-Protein Interactions with Coordinate-Aligned Pair-Map Decoding
- Practical Dynamic Pricing with Optimistic Likelihood Estimation
- Practical Performative Policy Learning with Strategic Agents
- Practical Scaling Laws: Converting Compute into Performance in a Data-Constrained World
- PreCoMem: Predictive Cognitive Memory for Self-Evolving Long-Term Dialogue Agents
- Precorrecting Structured Kernel Approximations for Gaussian Process Regression
- Predicting and Explaining Goal Failure in Sparse Goal-Conditioned Reinforcement Learning via Policy-Induced Functional Graphs
- Predicting Before Perceiving: A Dispositional Prior Framework for Emotion Prediction in Conversation
- Predicting Nothing Beats SAM 3: Revisiting Evaluation in Video Object Segmentation
- Prediction and Empowerment: A Theory of Agency through Bridge Interfaces
- Prediction Sets on the Plausibility Spectrum: A Geometric Framework for Conformal Prediction
- Predictive Geometry Shapes Spatiotemporal Representations of Constrained Random Walks
- Predictive Representation Learning for Partially Observed Neural Dynamics
- Preference-Based Self-Distillation: Beyond KL Matching via Reward Regularization
- Preference Optimization with Residual Anchoring for Underconstrained Physics Reconstruction
- Preference Summary Optimization for Popularity Bias Mitigation in LLM-based Recommendation
- Preferential dynamic modeling with forward-backward smoothing
- PrefillShare: A Shared Prefill Module for KV Reuse in Multi-LLM Disaggregated Serving
- Prefix Likelihood-Ratio Control: Tail-Stable Training for Long-Horizon Language Generation
- Prefix Semantic Dynamics for Measuring and Steering Long-Form Text Generation
- Pre-Policy Intervention for Bias-Aware Human–AI Decision Support
- Preregistered Belief Revision Contracts for Evidence-Gated Multi-Agent AI Deliberation
- Prescriptive Tree-based Learning for Contextual Scheduling under Duration Uncertainty
- Preserving Directional Correspondence with Ego-Centric Alignment for Cross-Modal Geo-Localization
- PressureBench: Stress-Testing the Parse-Execute Gap in Planner-Executor Systems
- Preventing Rank Collapse in Federated Low-Rank Adaptation with Client Heterogeneity
- Primal-Dual Flow Matching for Sample-Wise Constrained Generation
- PRIME: Poincaré return induced measure for learning partially observed dynamical systems
- Principled Policy Optimization for LLMs via Self-Normalized Importance Sampling
- Principled Top-$k$ Selection for Language Models with Hybrid Gradients
- Principles at the Back of Mind: Gated K ′ V ′ Memory for Durable Normative Control
- Prior-Calibrated Preference Optimization for Popularity-Debiased Recommendation
- Prioritize the Use of Simple Models to Diagnose Data
- PRISM : A Diagnostic Study of Support-Limited Budgeted Retrieval on One Shared RAG Stack
- PRISMat: Policy-Driven, Permutation-Invariant Autoregressive Material Generation
- PRISM-Bench: A Benchmark of Puzzle-Based Visual Tasks with CoT Error Detection
- PRISM-Bench: Measuring Value, Evidence, and Source Hierarchies in Frontier AI Systems
- PrismFlow: Residual Dynamics for Flow Matching in Time-Series Generation
- Prism: Generation-Time Detection and Mitigation of Secret Leakage in Multi-Agent LLM Pipelines
- Prism: Memory-Safe Parallel Reconfiguration for Elastic Long-Context MoE Training
- PRISM: Personalized Routing over Item Sparse-decomposition Modules
- PRISM: Prior Rectification and Uncertainty-Aware Structure Modeling for Diffusion-Based Text Image Super-Resolution
- PRISM: Progressive Reasoning through Iterative Slot Memory for Vision
- PRISMR: Overcoming Parse Collapse in Multimodal Listwise Ranking via Parameterized Representation Internalization
- PRISM: Spectral Analysis Aided Nested Manifold Learning for Efficient Multimodal Large Language Model Adaptation on Edge
- PRISM-Zero: Progressive Reasoning via Imagined Scene Manufacturing from Zero Data
- PrisonBreak: Jailbreaking Large Language Models with at Most Twenty-Five Targeted Bit-flips
- Privacy Amplification for BandMF via $b$-Min-Sep Subsampling
- Privacy–Harmonization Trade-offs in Data Space: Privacy-Budgeted Pooling under Domain Shift
- Privacy in the Era of Large Opaque Models: Theoretical, Legal, and Practical Perspectives
- Privacy Risk Scales with Effective Dimension in Federated Learning
- Private Incentive Learning with Adaptive Privacy Scheduling
- Privately Aligning Large Language Models via Rank-Based Group Relative Policy Optimization
- Private Online Prediction from Experts with Small Losses
- PrivateSeal: Low-Sensitivity Latent Directions for Diffusion-Resilient User-Specific Watermarking
- PrivGemo: Privacy-Preserving Graph Retrieval with Memory-Guided Control for LLM Reasoning
- PrivMark: Pixel-Space Steering against Privacy Leakage in Vision-Language Models
- PROACT-Agent: Progressive Runtime Oversight and Active Circuit-breaking for Real-Time Safety
- ProactBench: Beyond What The User Asked For
- Probabilistic Calibration Is a Trainable Capability in Language Models
- Probabilistic Guarantees for Adversarial Linear Contextual Bandits
- Probabilistic Recurrent Intention Switching Model
- Probabilistic Robustness-driven Universal Adversarial Perturbations with Explainability against Deep Reinforcement Learning-based Intrusion Detection System
- Probabilistic Sparse Auto-Encoders
- Probability-Conserving Flow Guidance
- Probe-and-Exit: Adaptive Early Termination for Chain-of-Thought Reasoning via KV Cache Rollback
- Probe-Geometry Alignment: Erasing the Cross-Sequence Memorization Signature Below Chance
- PROBE: Learning to Audit Policy Compliance in Tool-Using LLM Agents
- PROBE: When Prompt Learning Meets Graph Unlearning with Propagation-Aware Erasure
- Probing for Representation Manifolds in Superposition
- Probing Graph Neural Network Activation Patterns Through Graph Topology
- Probing the Hallucination Gap: Benchmarking Prompting Strategies and Mechanistic Interpretability in Large Language Models
- Problem-Conditioned and Relation-Aware Recommendation of Domain-Specific Symbols for Efficient Input Assistance via Neuro-Symbolic Graph Framework
- Problem Reductions at Scale: Agentic Integration of Computationally Hard Problems
- ProbMedTOD: A Bayesian Network Guided Task-Oriented Dialogue System for Patient History Taking
- Procedural Refinement by LLM-driven Algorithmic Debugging for ARC-AGI-2
- Procedure-Level Auditing of Symbolic ARC Systems: A Self-Audit with External Replication on ARC-AGI-2
- Process-conditioned Pretraining with Topographic Spatial Retrieval for Large EEG Models
- ProcFunc: Function-Oriented Abstractions for Procedural 3D Generation in Python
- ProCrit: Self-Elicited Multi-Perspective Reasoning for Multimodal Sarcasm Detection
- ProCTI: Prototype-Refined Global Conditioning for Diffusion-Based Time Series Imputation
- Procurement Auctions for Generative AI Search
- Profit Maximization in Bilateral Trade against a Smooth Adversary
- Prof-K: Probabilistic One-Pass Filtering for Efficient Top-$k$ Selection
- ProgramBench: Can Language Models Rebuild Programs From Scratch?
- Programmatic Reasoning with Structural Schema: A Unified Framework for Multi-Table Inference
- Progressive Denoising Regulation for Accelerating Diffusion Language Model Decoding
- Progressive Pseudo-label Self-balancing Towards Unsupervised Vision-Language Models Adaptation
- Progressive Residual Warmup for Language Model Pretraining
- Progressive Risk Estimation for Accident Anticipation
- Progressive Signal Calibration for Medical Image Segmentation: Diagnosing and Correcting Structural Misalignment in Training Signals
- ProjDevBench: Benchmarking AI Coding Agents on End-to-End Project Development
- Project Auto-World: Towards Automated Benchmarking of Neural Relational Reasoners
- Projected-Distortion Guided Quantization
- Projection-Free Algorithms for Nonsmooth Stochastic Convex-Concave Saddle-Point Problems
- Projection Learning: A Principled Way to Overcome Memorization in Distribution Learning
- ProLoRA: Bidirectional Communication-Efficient Federated Fine-Tuning of Large Language Models via Random Projection
- Promise-Driven Reinforcement Learning
- PROMISE: Provably Convergent Mutual Information Maximization for Robust Set-to-Set Matching
- Prompt-Collaborated Heterogeneous Graph Completion for Incomplete Multi-modal Learning
- Prompt-Driven Simulation with Feature Perturbation for Cross-Domain Few-Shot Object Detection
- PromptEcho: Annotation-Free Reward from Vision-Language Models for Text-to-Image Reinforcement Learning
- PromptExpert: Prompt-Guided Mixture-of-Experts Collaboration for Cross-Domain Few-Shot Object Detection
- Prompt Manifold Interpolation via Anchor-Residual Decomposition for Class-Incremental Learning
- Prompt Prototype Learning (PPL): Rethinking Text Supervision for AI-Generated Image Detection
- PROOF: Mixed-Traffic Jailbreak Defense via Profile Routing in Frozen LLMs
- Propagate to Discover: Graph-Structured Propagation for Generalized Category Discovery
- Propagation of Chaos in Contextual Flow Maps
- Proper Agnostic Learning of Functions of Halfspaces
- Property-Guided LLM Program Synthesis for Planning
- Property Memorization in Text-to-Image Generative Models
- Proportional Integral Control for Graph Convolutional Networks
- Proportional Representation from Pairwise Approval Data
- Proposal: 2nd Workshop on Agentic AI Benchmark and Application for Enterprise Tasks
- ProSearch: Process-Supervised Agentic Search with Dual-Granularity Advantage Shaping
- PROSH: Probabilistic Shielding via Risk Augmentation for Safe Model-free Reinforcement Learning
- Prosocial Readiness Bench: Safety Refusal, Cooperation, and Commons Restraint in LLM Agents
- Prospective Compression in Human Abstraction Learning
- PROTAX-SSL: Calibrated Probabilistic Taxonomic Classification with Self-Supervised Representations
- Protecting Shared Agent Workflows from Invalidation Attacks with Signed Trajectories and Runtime Verification
- ProteinArena: An Interactive Benchmark Suite for Evaluating General Protein Encoders
- ProteinOPD: Towards Effective and Efficient Preference Alignment for Protein Design
- Protenix-mini+: efficient structure prediction model with scalable pairformer
- Proteus: Incremental Memory Activation for Long-Context Sequence Modeling
- PROTEUS: Provable Self-Evolution for Skill-Composing Language Agents
- ProtGenBench: A Multi-task Benchmark for Evaluating State-of-the-Art Generative Models in de novo Protein Design
- ProtGPT3: An Open-source family of Promptable and Aligned Protein Language Models
- ProtoCDisco: Unsupervised Concept Discovery using Pretrained Representations (Extended)
- Protocol Sensitivity in Financial Numerical Reasoning Benchmarks
- ProtoMM: Prototype-based Self-Supervised Multimodal Learning for PPG and Accelerometry Signals
- ProTon: Parallel Logic Tree Neurons for Secure Hardware-Software Learning in AI Accelerator
- Prototype-Guided Latent Alignment for Data-Efficient Fine-Tuning of Molecular Foundation Models
- Prototype Self-optimization and Alignment Network for Few-Shot Medical Image Segmentation
- ProtoVis-Nav: Prototypical Visual Imagination for Visual-Spatial Aligned UAV Navigation
- Provable Anytime Ensemble Sampling Algorithms in Nonlinear Contextual Bandits
- Provable Data Scaling Law for Meta Learning via Complexity Minimization
- Provable Explanations for Any-Order Neural Additive Models
- Provable Generalization in Sub-1M Parameter Regimes via Near-Linear Time Tensor Sketching
- Provable Hallucination-Aware Video Sampling
- Provable Quantization with Randomized Hadamard Transform
- Provable Speedups From Dynamic Population Sizes in Evolutionary Algorithms for Multiobjective Optimization
- Provable Unlearning with Gradient Ascent on Two-Layer ReLU Neural Networks
- Provable Warm-Start Reinforcement Learning for Causal Structure Learning with Finite-Sample Guarantees
- Provably Communication-Efficient and Privacy-Preserving Federated Graph Neural Networks
- Provably Efficient Offline-to-Online Value Adaptation with General Function Approximation
- Provably Efficient Sample Complexity for Robust CMDP
- Provably Learning Diffusion Models under the Manifold Hypothesis: Collapse and Refine
- Provably Optimal Policy Learning over a Distribution of MDPs
- Provably Reduced Sample Cost in Prior-Guided Hyperparameter Optimization
- ProvenAI: Auditing Citation Faithfulness and Evidence Influence in Retrieval-Grounded Multi-Hop Question Answering
- PROWL: Prioritized Regret-Driven Optimization for World Model Learning
- Proximal Causal Learning of Optimal Individualized Dose Rule
- Proximal Conformal Prediction: Distribution-Free Counterfactual Inference under Unmeasured Confounding
- Proximal Multiple Policy Evaluation
- ProxyPose: 6-DoF Pose Tracking via Video-to-Video Translation
- ProxyUp: Training-Free Proxy-Conditioned Video Generation for Controllable Dynamics
- Prysma: Efficient Modality Adaptation for SLO-aware LLM-based Video Question Answering
- Pseudo-MDPs: A Novel Framework for Efficiently Optimizing Last Revealer Seed Manipulations in Blockchains
- PSRFQI: Purified and Smoothed Robust Offline Reinforcement Learning against Dynamics and Observation Perturbations
- Psytrax: flexible inference of learning trajectories in sequential decision-making experiments
- PTA: From Pretrained Representations to Acting Agents -- Bridging Pretraining, Planning, and Test-Time Decision Making
- Public Discourse Analysis Is Asking the Wrong Question
- Publishing Below-Threshold Triangle Counts under Local Weight Differential Privacy
- PulseCol: Periodically Refreshed Column-Sparse Attention for Accelerating Diffusion Language Models
- PULSE: Identifying Demonstration-Utility Features with Sparse Autoencoders
- PURE: Budget-Aware Update Inference from Historical Model Outputs
- PVMMoE: Robust Multimodal Photovoltaic Forecasting under Dynamic Lag and Missing Observations
- PV-WSI: Rethinking Pseudo-Video Learning for Whole Slide Image Segmentation via Spatial Discontinuity Modeling
- PyINE: A Framework for Scalable Elicitation and Oversight via Code Execution
- Q3.6: Metadata-Conditioned Lossless Compression for Quantized LLMs
- QAC: Data Quality-Aware Uncertainty Calibration for Trustworthy AI
- QASM-Eval: A Dataset to Train and Evaluate LLMs on OpenQASM-3 Beyond Quantum Circuits
- QEC Model Zoo: Democratizing AI-enhanced Quantum Error Correction
- QGR: A Query-Grounded Reasoning Transformer for Visual Reasoning
- QGround: Condition-Wise Evidence Aggregation for 3D Grounding with 2D VLMs
- Q-MMR: Off-Policy Evaluation via Recursive Reweighting and Moment Matching
- QMuon: Smoothing Muon for Spectral Optimization and Quantum Speedups
- Q-Probe: Scaling Image Quality Assessment to High Resolution via Context-Aware Agentic Probing
- Q-Residual Physics: Hamiltonian-Structured Quantum Residual Learning for Embodied Dynamics
- Q-Restore: Image Restoration via Prior Alignment between Image Generation and Quality Assessment
- Qrita: High-performance Top-k and Top-p using Pivot-based Truncation and Selection
- Q-Save: Towards Scoring and Attribution for Generated Video Evaluation
- QuadContra: Contrapositive Quadruplet Learning for Federated Semi-Supervised Remote Sensing Image Scene Classification
- Quadratic integrate-and-fire neurons exhibit less fragmented loss landscapes and outperform leaky integrate-and-fire neurons in spike-based gradient descent
- QuadricViT: Feed-Forward Superquadric-based Model for Efficient 3D Occupancy Prediction
- Quality-Aware Belief Sharing and Soft Back-Fusion for Multi-Agent Source Term Estimation
- Quality-Diversity Optimization as Multi-Objective Optimization
- QuantDemoire: Quantization with Outlier Aware for Image Demoiréing
- Quantifying Centrality for Complex Data
- Quantifying Concentration Phenomena of Mean-Field Transformers in the Low-Temperature Regime
- Quantifying Faithful Confidence Expression in Large Reasoning Models
- Quantifying Potential Observation Missingness in Inverse Reinforcement Learning
- Quantifying Prior Dominance in RAG Systems
- Quantifying the Failure Rate of Single-Run Training Configuration Comparisons Across ML Domains
- Quantifying the Generalization Advantage of Causal World Models
- Quantifying the Utility of User Simulators for Building Collaborative LLM Assistants
- Quantitative Certification of Agentic Tool Selection
- QuantitativeFinance-Bench: Benchmarking AI Agents on Real-World Quantitative Finance Tasks
- Quantized Evolution Strategies: High-precision Fine-tuning of Quantized LLMs at Low-precision Cost
- Quantized Functional-Basis Networks: Recoverable Discrete Codes for Neural Layers and Adapters
- QuantLRM: Quantization of Large Reasoning Models via Fine-Tuning Signals
- Quant.npu: Enabling Efficient Mobile NPU Inference for on-device LLMs via Fully Static Quantization
- Quantum Best Arm Identification with Limited Round of Adaptivity: Lower Bounds and Algorithms
- Quantum Composite Hypothesis Testing with Small Error
- Quantum Diffusion Graph Neural Networks
- Quantum Error Correction with RL decoders under Non-Markovian Noise
- Quantum Feature eXpansion for Tabular Data: A Large-Scale Empirical Study with Design Guidelines
- Quantum Federated Learning under Shot Noise and Heterogeneity: Convergence Analysis
- Quantum Generator Kernels
- Quantum Maximum Likelihood Prediction via Hilbert Space Embeddings
- Quantum Speedup of Multi-armed Bandits at Scale by Tackling Memory Decoherence
- Quantum Speedups for Log-Concave Sampling from Local Structure
- Quantum Speedups for Stochastic Optimization with Heavy-Tailed Noise
- Quantum-Steered Diffusion: Stereochemically Correct Protein Generation via NISQ-Guided Score Matching
- QuantWeather: Quantile-Aware Probabilistic Forecasting for Subseasonal Precipitation
- Quasimetric Distance with Flow-Based Planning for Offline Hierarchical Goal-Conditioned RL
- QuasiMoTTo: Quasi-Monte Carlo Test-Time Scaling
- Qubrio: High-Performance Quantum Compilation via Multi-Agent LLM Collaboration
- Query as a Resource: Activity-Cost Guided Remote Sensing Domain-Incremental Object Detection
- Query Learning Nearly Pauli Sparse Unitaries in Diamond Distance
- Query Lower Bounds for Diffusion Sampling
- Query-Only Attention Preserves Plasticity in Fully Online Continual Learning
- QuestBench: A Course Curated Benchmark for Expert-Level Cross-Domain Deep Search in Language Models
- Question-Conditioned Adaptive Frame Selection for Streaming Perception via Conditional Information Gain
- Question Knows Where to Look: A Query-Adaptive Framework for Zero-Shot Video Understanding
- QUEST: Q-Learning for Uncertainty-Guided Efficient Search Teams
- Queue-Audit Specification for Fixed-Capacity Next-Admissible Safety Queues in Clinical Trials
- Quicksviewer: An LMM with Early Video Compression using Annealed Gumbel-Softmax
- QuIL: Quantifying Quantisation-induced Information Loss in Large-Scale Models
- QUIVER: Quantum-Informed Views for Enhanced Representations in Large Machine Learning Models
- Qu-LoRA: Discovering and Overcoming the Frozen-Adapter Bottleneck in LoRA-Based Continual Learning
- Quotient flag complexes data structure
- QuoVLA: Quotient Space for Vision-Language-Action Models
- Q-VAR: Quantizing Visual AutoRegressive Models
- R²-Write: Reflection and Revision for Open-Ended Writing with Deep Reasoning
- RABBiT: Rapidly adaptive BOLD foundation model via brain-tuning for accurate zero-shot and few-shot prediction of speech-elicited responses in the brain
- RA-CFGCache: From Branch-Level Criteria to Guided-Risk Control under Classifier-Free Guidance
- Radar2Shape: 3D Shape Reconstruction from High-Frequency Radar using Multiresolution Signed Distance Functions
- RADAR: Relative Angular Divergence Across Representations
- RADAR: Text-Guided Medical Image Segmentation via Residual Aggregation and Dense Alignment Representations
- RadOmni: Advancing Foundation Model for Non-contrast CT with Omni Radiology Knowledge
- RAG-Anything: All-in-One RAG System
- RAGForget-Bench: Low Direct Leakage Does Not Mean Retrieval Forgetting in Hybrid RAG
- RAGtime-PIANO: Efficient Secure Remote RAG
- RAIC: Representation-Aware Selective Inference for Unicode Attack Robustness
- RAM-H1200: A Unified Evaluation and Dataset on Hand Radiographs for Rheumatoid Arthritis
- Random-Effects Centroids for Domain Generalization
- Random Matrix Theory of Early-Stopped Gradient Flow: A Transient BBP Scenario
- Randomness is sometimes necessary for coordination
- Random-Set Graph Neural Networks
- Random-Set Large Language Models
- Random-Subspace Frank–Wolfe over Strongly Convex Sets
- Rank-Accuracy Trade-off for LoRA: A Gradient-Flow Analysis
- RankAlign: Unsupervised Vision-Language Representation Alignment via Rank Transformation
- RankGFN: Rank-Aware GFlowNets for Budget-Constrained Molecular Discovery
- Ranking-Aware Calibration for Reliable Multimodal Reinforcement Learning
- Ranking Collapses but Selection Survives: Checkpoint Readout Under Constrained Adaptation
- Rank Intervals for Leaderboards: A Hierarchical Framework for Model Evaluation
- Rank-Transformed Dissimilarity Profiles for High-Dimensional Classification
- RankVQ: Low-Rank Parameterized Commutative Vector Quantization for KV Cache Compression
- RAO-Nav: Probing Omni-Language Models for Zero-shot Semantic Audio-Visual Navigation
- Rapid Grassmannian Averaging with Chebyshev Polynomials
- RAPNet: Relative Pose and Action Primitives Network for Bi-Equivariant Robotic Manipulation
- RAP: Retrieval-Augmented Paper-to-Poster Generation via Declarative Markup from Human-Crafted Exemplars
- RAQUEL: Robust Validation of Machine Unlearning through Query Executions over Aligned Databases
- Rare-Class Signal Suppression in Long-Tailed Multi-Expert Fine-Tuning
- RASpec: Auditable Risk-Adaptive Speculative Decoding for Regulated Agentic AI
- Rational ANOVA Networks
- Rational Clarification by Assistive Agents via Value-of-Information Reasoning
- Raw2Event: First Paired Raw, RGB and Real-Event Benchmark, with Multi-Dimensional Diagnostics for the Sim-to-Real Gap
- RAWild: Toward Sensor-Agnostic RAW Object Detection via Physics-Guided Curve and Grid Modeling
- RBS-ATTENTION: Radius-Bounded Sparse Prefill for Long-Context Large Language Models
- RD-LTC-Topo: Reaction-Diffusion Augmented Liquid Neural Networks with Topological Priors for Spatiotemporal EEG Decoding
- REACT: A Conditioning Framework for User-Adaptive sEMG Hand Pose Estimation
- Reaction Coordinate Descent: A Thermodynamic Framework for Efficient Multi-Modal Agents
- Read in English, Answer in Hindi? Exploring Data Compositions for Cross-Lingual Knowledge Transfer
- Reading $\neq$ Seeing: Diagnosing Typography Blindness in Vision-Language Models
- Reading the Room: Foundations, Design and Challenges of Normative Competence in LLMs
- Reading the Unreadable: Text-Aware Image Super-Resolution Needs Reasoning
- Read-Only Zero-Shot Classifier Expansion from Pairwise Semantics
- Read-Write Diffusion: Data-Level Pre-Conditioning for Denoising Diffusion Probabilistic Models
- RealDrag: The First Dragging Benchmark with Real Target Image
- REAL-Q: E2E LLM Quantization via Dynamic Gradient Descent
- Real-Time Causal Geometric Flows for Adaptive Therapeutic Drug Monitoring
- Real-Time Conversational Agents: Toward Natural Multimodal Interaction
- Real-Time Multimodal Conversational AI
- Realtime-VLA FLASH: Speculative Inference Framework for Diffusion-based VLAs
- Real vs. Semi-Simulated: Rethinking Evaluation for Treatment Effect Estimation
- Reason Before You Speak with Latent Diffusion
- Reasoning as Compression: Unifying Budget Forcing via the Conditional Information Bottleneck
- Reasoning Denoiser: Denoising Reasoning Traces for Hallucination Detection in Large Reasoning Models
- Reasoning-Grounded Sparse Autoencoders for Interpretable Vision-Language Models (Extended)
- Reasoning Guided Embeddings: Leveraging MLLM Reasoning for Improved Multimodal Retrieval
- Reasoning Is More Than the Model: Harness-Aware Evaluation of Agents on Verifiable Reasoning Tasks
- Reasoning or Copying? How Target-Conditioned Rationales Teach Distilled Students to Read the Hidden Label
- Reasoning or Familiarity? Evaluating LLMs on Classic, Misère and Novel Board Games
- Reasoning-Structured Videos: A Stratified Diagnostic Suite for Compositional Consistency in World Models
- Reasoning to Rank: Exploiting Large Language Models for Recommendation
- Reasoning-Trace Collapse: Evaluating the Loss of Explicit Reasoning During Fine-Tuning
- Reasoning via Test-Time Instance-Level Policy Gradient in Latent Space
- Reasoning with Neologisms: Can Soft Tokens Learn Composable Reasoning Skills Without Forgetting?
- Reasoning with Sampling: Cutting at Decision Points
- Reason in the Words You Speak: Idiolectal Paraphrasing Off-Policy Traces for Reasoning Distillation in VideoLLMs
- Rebellious Student: Reversing Teacher Signals for Reasoning Exploration with Self-Distilled RLVR
- Recalibrate, Don't Replace: Class-Pair Calibration for Robust Graph Neural Networks
- REChart: Reasoning-Efficient Chart Editing with Large Reasoning Models
- RECIPE: Learning to Rank Complete Precursor Sets for Inorganic Retrosynthesis
- RecoAtlas: From Semantic Plausibility to Set-Level Utility in LLM Recommendation Agents
- Recognizing and Restructuring Latent Experts for Large Language Model Compression
- Recommender-Creator Game
- Reconcile Before You Generate: Multimodal Evidence Reconciliation for Medical Report Generation
- Reconciling Contradictory Views on the Effectiveness of SFT in LLMs: An Interaction Perspective
- Reconciling Safety and Performance via Dual-Expert Offline Imitation Learning
- Reconstructing and Analyzing High-Resolution Hyperparameter Loss Landscapes via Surrogate Modeling
- Reconstructing the Early Universe from Sparse Tracers with Point-Cloud Flow Matching
- Reconstruction Converges Before Decoder Reproducibility in Sparse Autoencoders
- Recoverability Maps and Intervention Deadlines for Action-Chunked Robot Policies
- Recovering Articulated Objects from Monocular Interaction Videos
- Recovering No-Trade Regions: Pontryagin-Guided Policy Projection for Transaction-Cost Control
- Recovering Physical Dynamics from Discrete Observations via Intrinsic Differential Consistency
- Recurrent Point Transformer via Test-Time Training
- Recurrent Residual Quantization: A Progressive Multi-Precision Representation for LLMs
- Recursive Flow Matching
- Recursively Trained Diffusion Models: Limiting Collapse Distribution and Spectral Characterization
- Recursive Multi-Agent Systems
- Recursive Semantic Divergence for LLM Agent Consistency
- ReDeck: Step-Level Render-Grounded Refinement for Document-to-Slide Generation
- Reducing Heterogeneous Constraint Violations in Large Neighborhood Search
- RedVLA: Physical Red Teaming for Vision-Language-Action Models
- ReelBench: Towards Systematic Evaluation of Multi-Shot Narrative Video Generation under Image Reference
- Re-evaluating Position and Velocity Decoding for Hand Pose Estimation with Surface Electromyography
- Re-Examine, Don't Re-Sample: Critic-Localized Inference for Interpretable Medical Reasoning
- Refactoring Code Through Library Design
- Reference-Driven Multi-Speaker Audio Scene Generation from In-the-Wild Priors
- Reference-Guided Training: Adaptive Gradient Scaling via Per-Sample Loss Comparisons
- Reference-Specific Unlearning Metrics Can Hide the Truth: A Reality Check
- Referring-Aware Visuomotor Policy Learning for Closed-Loop Manipulation
- RefineAny3D: Depth Refinement as Semantic Alignment for Monocular 3D Detection
- Refine Connections, Close the Gap: A Reliable Enhancement Framework for Driving Scene Topology
- Reflective AI Must Empower Human Agency for a Digital Renaissance
- Reflective Causal Agents
- Reflective VLA: In-Context Action Consequences Makes VLAs Generalize
- ReflectMT: Internalizing Reflection for Efficient and High-Quality Machine Translation
- REFORM-3D: A Representation-Centric Evaluation Framework for 3D Medical Vision Foundation Models
- Reformulating Neural Operators in $d+1$ Dimensions for Embedding Evolution
- ReFPO: Reflow Regularization for Flow Matching Policy Gradients
- Reframing General Intelligence as a Civilizational Property
- Refusal direction is the safety training trajectory: raising refusal-subspace stable rank can reduce ablation attack effectiveness
- REGATE: Confidence-Calibrated Integration of Temporally-Aligned Exogenous Texts for Dynamic Graphs
- ReGDec: Decoupling Region Cues from Global Semantics for VLM-Grounded Open-Vocabulary Dense Perception
- REGINA: Regularized Encoder with Latent Cycle-GAN for In-vitro Neural Cell Perturbation Approximation
- ReGKD-VLM: Distilling Vision-Language Model Embeddings via Gradient Topology Transfer
- ReGround: Grounding All Moments in Long Videos via Retrieval, Reranking and Refinement
- ReGuidance: Diffusion Steering with Strong Latent Initializations Solves Hard Inverse Problems
- Regularized Centered Emphatic Temporal Difference Learning
- Reimagining Meaningful Model Multiplicity
- Rein3D: Reinforced 3D Indoor Scene Generation with Panoramic Video Diffusion Models
- REIN: Fine-tuning Steerable Language Model without Intermediate Guidance
- Reinforcement Learning Agents Are Swimmers
- Reinforcement Learning for Adaptive Expert Routing in Mixture-of-Experts Diffusion Transformers
- Reinforcement Learning for Experimental Sciences: Bridging the Simulation-to-Reality Gap
- Reinforcement Learning for View-Adaptive Distillation in 3D Gaussian Compression
- Reinforcement Learning Itself is a Good Continual Learner in MLLMs
- Reinforcement Learning via Reward-Releasing Actions
- Reinforcement Learning with Abstention: Interaction-Aware Regret Bounds
- Reinforcement Learning with Decomposed Subtasks
- Reinforcement World Model Learning for LLM-based Agents
- ReinforceTree: PPO-Guided Adaptive LoRA Routing for Continual Learning
- Reinforcing Chain-of-Thought Reasoning with Self-Evolving Rubrics
- Reinforcing Multimodal Reasoning Against Visual Degradation
- ReLaT-E: End-to-End Relational Latent Tuning for Multimodal Video Generation
- Relational Invariant Approach to Attack-Agnostic Data Poisoning Detection in Tabular Data
- Relation-Aware Adaptive-Depth Heterogeneous Graph Neural Network for Future Bid Winner Prediction
- Relative Q-Learning
- ReLAttack
- Reliability-Calibrated Semantic Knowledge Guidance for Low-Light Image Enhancement
- Reliability-Disentangled Distillation: Purifying Bits and Semantics for Multimodal Hashing
- Reliability-Guided CLIP Prior Alignment for Exposure Correction
- Reliable Autonomous Agent System Requires Human-Agent Collaboration
- Reliable Federated Multi-View Learning via Conflict-Aware Evidence Calibration
- Reliable Neuroimaging Model Interpretation via Closed-Form ROI Influence Posteriors
- Reliable Tensor Hypergraph Learning with Structural Alignment for Multimodal Survival Analysis
- RELIANCE: Reasoning Evaluation with Logical Integrity and Accuracy for Confidence Enhancement
- REMAP: Evaluating Dual Perspective Spatial Reasoning
- REMAP: Regularized Matching and Partial Alignment of Video Embeddings
- Remember with Confidence: Uncertainty Quantification for Spatio-temporal Memory with Probabilistic Guarantees
- ReMem: Compressed Memory as a Native Modality for Efficient Long-Context LLMs
- ReMiT: RL-Guided Mid-Training for Iterative LLM Evolution
- Remix, Don’t Expand: Chunked Template Routing for Compute-Efficient Transformers
- Render and Compress: Tokenizer-Free Context Representation for Language Models
- Render, Don't Decode: Weight-Space World Models with Latent Structural Disentanglement
- Rennala MVR: Improved Time Complexity for Parallel Stochastic Optimization via Momentum-Based Variance Reduction
- RePaGLU: Efficient LLMs via Structural Reparameterization on GLU Layers
- Repeated Revision Collisions: A Matched Shared-Path Audit of Typed Revision Containment
- RepFusion: Leveraging Multimodal Priors for Denoising in Representation Space: A Closer Look
- Replay-buffer engineering for noise-robust quantum circuit optimization
- Replay Refinement for Neural Schrödinger Bridge Training
- Replicable Bandits with UCB based Exploration
- Represent as Complements, Retrieve by Association
- Representational Geometry Determines Learning Dynamics in BMI Tasks
- Representational Geometry Reveals How Context Structures Concept Spaces in Language Models
- Representation Alignment Rests on Linear Structure
- Representational Stability Predicts Semantic Transparency: Cross-Architecture Morphological Probing in LLMs
- Representation-informed Gradient Compression
- Representation Learning Enables Scalable Multitask Deep Reinforcement Learning
- Representation Learning in Factored-Latent Contextual Linear Bandits
- Representation-Level Constrained Preference Optimization for Language Model Detoxification
- Representations for the Physical Sciences
- Representative Attention For Vision Transformers
- RepUp: Robust Representation-Centric Feature Upsampling against Visual Perturbations
- Repurposing Pre-trained LLMs as High Fidelity Continuous Text Autoencoders
- REQwen: LLM-Augmented Reverse Engineering of Deep Neural Network Binaries at Scale
- Reranking with Intra-modal Visual Association for Text-to-Image Person Re-Identification
- ReRank-TTA: Learning to Rerank at Test Time for Robust Handwritten Text Recognition
- RESCAST-100K: A Comprehensive Dataset for Cross-Domain Residential Load and Indoor Temperature Forecasting
- ResID: Reinforced Style Injection in Diffusion for Arbitrary Image Style Transfer
- Residual-Autoregressive Context for 3D Gaussian Splatting Compression
- Residual Canonical Alignment for Stable Multiview Representation Learning
- Residual Count Frontiers for Hybrid Average-Reward Reinforcement Learning via Bridge Certificates
- Residual Evidence Under Nuisance Geometry: A Framework for Shared-Source Verification
- Residualized Temporal Sparse Autoencoders for Interpreting Diffusion Models
- Resilient Latent Readouts for Long-Context Question Answering
- Resilient Online Gaussian Processes with Anomaly-Aware Gates
- Resilient Semi-Supervised Inference with Heterogeneous Unlabeled Data
- Resolution as a First-Class Decision: Task-Conditioned Routing for Efficient Multimodal Large Language Models
- Resolving Time-Frequency Ridge Crossings via Frequency-Rate Lifting
- Resonate‑and‑Fire Spiking Neural Network for End‑to‑End EEG Emotion Recognition in Real-Time
- Resource Block Tree: A Capacity-Driven Hierarchical Memory for Long-Horizon Agents
- Respecting Physical Priors: Reconstructive Contrastive Learning for Signal-Domain Sim-to-Real Transfer
- ResPlan: A Large-Scale Vector-Graph Dataset of 17,000 Residential Floor Plans
- Response-powered Localized Conformal Prediction
- Response Resilience in Large Language Models
- Responsible Communication of Machine Learning Research in Biomedicine
- ReSPO: R\'enyi Sequence-Level Policy Optimization for Gradient Starvation in Off-Policy Learning
- Restore or Reveal? Measuring Training Data Leakage in Image Restoration with Metric-Based Membership Inference
- Restore Text Without Breaking Vision: Vision-Preserving On-Policy Distillation for VLMs
- Restoring Semantic Coherence in Long-Sequence dMLLM Decoding
- Restoring the Lost Magnitude: Rethinking CLIP for Zero-Shot Anomaly Detection
- Retain-Neutral Surrogates for Min-Max Unlearning
- Rethinking Action Chunking: Activity Sparsity Governs Effectiveness
- Rethinking adaState: Self-Evolving Anchors for Streaming Video Generation
- Rethinking aeroDesignGym: A Hierarchical Benchmark for Agentic Aircraft Conceptual Design
- Rethinking All Evidence: Enhancing Trustworthy Retrieval-Augmented Generation via Conflict-Driven Summarization
- Rethinking a Scalable Measure of Loss Landscape Curvature for Analyzing the Training Dynamics of LLMs
- Rethinking Audio-Visual Synchronization: Preventing Modality Collapse in Deepfake Detection
- Rethinking Bayesian Optimization for Co-Optimizing LLM Training Configurations
- Rethinking beyond Decoupled PEFT: Geometry-Aware Low-Rank Adaptation via Riemannian Reparameterization
- Rethinking beyond Frame Selection: Rethinking Long-Video Understanding with MLLMs
- Rethinking beyond the Trial-and-Error Loop: Hybrid Projection and Automated Tuning for Distributed Training
- Rethinking Brain Decoding with CLIP: The Role of Adversarial Robustness
- Rethinking Causal Action Tokenization with Conditional Annealing in Flow Matching
- Rethinking Cell Embedding Evaluation for Foundation Models in the Era of AIVC
- Rethinking Chemical Reaction Prediction: Vision-Centric Reaction Mechanism Prediction with Chemist-like Reasoning
- Rethinking Circuit Discovery with Learnable Corruptions
- Rethinking cLOO: Conflict-Aware Leave-One-Out for Multi-Hop RAG Attribution
- Rethinking closing the Gap: Cross-Modal Geometry Transfer for Unimodal Survival Prediction
- Rethinking conditional Memory Enhanced Item Representation for Generative Recommendation
- Rethinking Contrastive Loss in CLIP Post-training: A Complementary Framework with Frozen Text Encoder
- Rethinking co-PiLOT: Constrained Physics-Informed Latent Optimization for Target-Driven Inverse Design
- Rethinking Cross-View Geo-Localization: A Probabilistic Approach
- Rethinking Data Generation for Long-Horizon Tool Calling
- Rethinking dECA: Decentralizing Block-Wise Adam for Efficient LLM Full-Parameter Fine-Tuning on Non-IID Data
- Rethinking Defense for Computer-Use Agents: Context Deception Attacks are Simple to Defend
- Rethinking Design Choices in Offline MARL: The Role and Stabilization of Value Decomposition
- Rethinking Diffusion for VI--ReID: From Pixels to Priors
- Rethinking Distillation in Pedestrian Re-Identification: A Spatial Aggregation Perspective
- Rethinking Drone-Satellite Cross-View Localization under Long-Term Temporal Drift
- Rethinking el Agente Disco: Agentic Transition-State Discovery via Reasoning-Guided Search
- Rethinking Entropy Allocation in LLM-based ASR: Understanding the Dynamics between Speech Encoders and Large Language Models
- Rethinking evil Spectra: How Optimisers can Amplify or Suppress Emergent Misalignment
- Rethinking Experience Utilization in Self-Evolving Language Model Agents
- Rethinking fast Rates for Offline Contextual Bandits with Forward-KL Regularization under Single-Policy Concentrability
- Rethinking frameVGGT: Coherence-Preserving Memory for Bounded Streaming Geometry
- Rethinking Future Blood Pressure Prediction from PPG as Morphology-Aware Representation Learning
- Rethinking Geometric Depth in Monocular 3D Object Detection: A Projection-Consistent Reformulation
- Rethinking Handwritten Character Recognition
- Rethinking hessian approximations beyond optimization
- Rethinking How to Remember: Beyond Atomic Facts in Lifelong LLM Agent Memory
- Rethinking hybrid Zeroth- and First-Order Method for Stochastic Nonsmooth Minimax Optimization
- Rethinking Identification and Semantic Meanings: Difference-based Knowledge Graph Embedding
- Rethinking impact of the Correlation between Aleatoric and Epistemic Uncertainty on Out-of-Distribution Detection
- Rethinking Incompleteness: Formalizing Protocol Divergence and Train-Once Learning for Robust IMVC
- Rethinking Intelligence: Brain-like Neuron Network
- Rethinking internal Probe Signals Are Insufficient for Training-Free LLM Routing: Three Failure Mechanisms in Cross-Model Difficulty Estimation
- Rethinking in-Weight Memorization or In-Context Deduction? Disentangling Sequence Modeling Strategies via Cellular Automata
- Rethinking Knowledge Distillation for Diffusion Language Models
- Rethinking Language Consistency in Multilingual Reasoning for LLMs
- Rethinking Language Model Scaling under Transferable Hypersphere Optimization
- Rethinking learning Chromatic Scene Field for Consistent Novel View Rendering
- Rethinking learning Global Fairness and Utility Across Acceptance Rates
- Rethinking learning to Deaggregate: Large-scale Trajectory Generation with Spatial Priors
- Rethinking learning Where to Look: Observation Policy Optimization for Thinking with Images
- Rethinking LLM Fine-Tuning via Weight Space Reparameterization: Preserving Safety during Downstream Adaptation
- Rethinking Local Learning: A Cheaper and Faster Recipe for LLM Post-Training
- Rethinking LoRA Initialization for Robust Asymmetric Learning Rates
- Rethinking Machine Unlearning for Biometric PII
- Rethinking masking Causality and Conditional Dependence
- Rethinking maskSAM 3: Text-Prompted SAM Adaptation for Volumetric Medical Image Segmentation
- Rethinking mIND: Monge Inception Distance for Generative Models Evaluation
- Rethinking Multinomial Logistic Mixture of Experts with Sigmoid Gating Function
- Rethinking multi-scale Predictive Representations for Goal-conditioned Reinforcement Learning
- Rethinking nightTrap: Benchmarking Needs-Review Routing in Night Camera-Trap Workflows
- Rethinking Nonparametric Teaching via Geometry
- Rethinking omitted-Variable Sensitivity Analysis for Generalizing Randomized Trials
- Rethinking only the Change: Difference-Aware Visual Token Compression for Multi-Frame GUI Agents
- Rethinking Parallel Multi-Agent Systems: A Cost-Aware Framework for Efficient Coordination
- Rethinking passive Prediction Isn't Enough: Temporal Foundation Models Should Become Actionable World Models
- Rethinking Polynomial Bases in Kolmogorov-Arnold Networks for Time Series Forecasting
- Rethinking Positional Encoding for Neural Vehicle Routing
- Rethinking position: Next-Generation Game Engines Should Be Built on Interactive Generative Video
- Rethinking proDG: Prototypes for Data-Free Generative Post-Hoc Explainability
- Rethinking Psychometric Evaluation of LLMs: When and Why Self-Reports Predict Behavior
- Rethinking Ratio-Based Trust Regions for Policy Optimization in Multi-Agent Reinforcement Learning
- Rethinking Reasoning Post-Training with General Chat Boosting and Dual-Reward Refinement
- Rethinking Reconstruction Capacity for Unified Wafer Anomaly Detection
- Rethinking rethinking Attention in Depth for Operator Learning
- Rethinking Rubric Generation for Improving LLM Judge and Reward Modeling for Open-ended Tasks
- Rethinking Scaling Laws for Vision-Language Adapters: Learning Rate, Adapter Capacity, and Data Exposure
- Rethinking sliced Inner Product Gromov–Wasserstein Distances
- Rethinking source-Matched Flow Matching via Radial-Angular Transport
- Rethinking Spectral GNNs: From Three Theoretical Glitches to Shift-Domain Message Passing
- Rethinking strengthening Unsupervised Graph Out-of-Distribution Detection with Structural Resonance
- Rethinking structSAM: Structure- and Spectrum-Preserving Token Merging for Segment Anything Models
- Rethinking structure-Aware KV Cache Compression for Video Language Models
- Rethinking surprise Minimisation as a Substrate for Cultural Emergence in Multi-Agent LLMs
- Rethinking taming Preconditioner Drift: Unlocking the Potential of Second-Order Optimizers for Federated Learning on Non-IID Data
- Rethinking the Divergence Regularization in LLM RL
- Rethinking the Readout: Unlocking Video Backbones for AI-Generated Video Detection
- Rethinking Training Targets, Architectures and Data Quality for Universal Speech Enhancement
- Rethinking trust Regions Sell, But Who's Buying? Generalized Power-Law Surrogates for Clipped and KL-Constrained Reinforcement Learning
- Rethinking Unbiased Sampling in GNNs: A Bias–Variance Optimal Approach
- Rethinking understanding Private Evolution as Learning-Augmented Clustering
- Rethinking Visual Attribution for Chest X-ray Reasoning in Large Vision Language Models
- Rethinking Visual-Language-Action Model Scaling: Alignment, Mixture, and Regularization
- Rethinking Visual Reasoning in Text-to-Image Reward Modeling
- Rethinking voiceNet: Fine-Grained Voice Understanding Beyond Emotion at Scale
- Rethinking whom to Trust? Adaptive Collaboration in Personalized Federated Learning
- RetinAI: A Low-Cost AI System with Wearable Headset and Retinal Imaging for Eye Tumor Home Screening
- Retrieval-Augmented Interpretable Learning: Towards Task-Specific Zero-Shot Models in Healthcare
- Retrieval-Augmented Memory Is Not Enough: The Monotonic Retrieval Paradox in LLMs
- Retrieval is Enough: Training-Free Interpretability with a Tool-Using Agent
- Retrieving Earlier: Query-Dependent Retrieval Depth in Hybrid Search
- RetroAnalyze: Auditing Single-Step Retrosynthesis Across Datasets, Benchmarks, and Models
- RetroDKR: Enhancing Retrosynthesis Prediction with Dual Knowledge Retrieval
- Revisable by Design: A Theory of Streaming LLM Agent Execution
- Revisiting AdaGrad in Stochastic Convex Optimization: Last Iterates, High Probability, and Lower Bounds
- Revisiting Adaptive Frequency-Domain Soft-Thresholding for Noise-Robust Image Classification
- Revisiting Associative Recall in Modern Recurrent Models
- Revisiting Constant Stepsize Stochastic Approximation with Decision-Dependent Markovian Noise
- Revisiting Deterministic Diffusion through Reverse Transition Kernels
- Revisiting Diffusion Model Predictions Through Dimensionality
- Revisiting Efficient Labeling for Lidar Semantic Segmentation
- Revisiting GAN with Bayes-Optimal Discrimination
- Revisiting Linear Dynamics for Time-Series Anomaly Detection: The FLUDD Framework
- Revisiting Numerical Forecasting Models for Language-Based Human Trajectory Prediction
- Revisiting Online Time Series Forecasting: From Leakage Correction to Architecture Design
- Revisiting Policy Gradients for Restricted Policy Classes: Escaping Myopic Local Optima with $k$-step Policy Gradients
- Revisiting Preference Optimization from Reward Margin Perspective
- Revisiting Single Image Reflection Removal with Prior-Informed Dual-Stream Diffusion
- Revisiting Subgradient Dominance in Robust MDPs: Counterexamples, Hardness, and Sufficient Conditions
- Revisiting the Heatmap-Guided MCTS Paradigm for Traveling Salesman Problem
- Revisiting Transformer Layer Parameterization Through Causal Energy Minimization
- Reviving Filter Banks: Accurate and Interpretable Local Image Filtering via Structured Latent Spaces
- ReVSeg: Incentivizing the Reasoning Chain for Video Segmentation with Reinforcement Learning
- Reward as An Agent for Embodied World Models
- Reward Contrast, Not Algorithm Labels: A Diagnostic Audit of Critic-Free Group-Relative RL for LLMs
- Reward-Estimated Hypergradient for Bilevel Reinforcement Learning with Black-Box Follower
- Reward-Free Flow Q Learning with f -Divergence Guided Critics
- Reward-Guided Joint Generation of Survival Data
- Reward-Guided Knowledge Distillation
- Reward Hacking in Multimodal Reinforcement Learning via Linguistic Priors
- Rewarding Structural Conformance of Reasoning using Process Mining
- Reward Is Not a Universal Interface for Generative Reinforcement Learning
- reward-lens: A Mechanistic Interpretability Library for Reward Models
- Reward Modeling for Multi-Agent Orchestration
- Reward-Preserving Attacks For Robust Reinforcement Learning
- Reward Stealing Attack on Large Language Models
- Reward Transport: Property Control in Flow Matching via Noise-Space Alignment
- Reward-Type Ablation Reveals Mechanism-Dependent Algorithm Rankings in Mixed-Motive Multi-Agent Evaluation
- Reweighting is not Harmless: Understanding the Role of Dataset Characteristics
- Rewire then Express: Cascaded Flow Matching for Single-Cell Perturbation Response Prediction
- RF-Agent: Neuro-Symbolic Multi-Agent Reasoning for Real-Time UAV RF Detection and Protocol Verification
- RF-Logic: Scalable Recency-Frequency Logic Rules Learning for Interpretable Temporal Knowledge Graph Forecasting
- RF Prior: Preserving Global-Context Priors for Efficient Instance Segmentation Transfer
- RFSchemBench: A Multimodal Benchmark for RF Circuit Schematic Understanding
- RGBD20K: A Large-Scale Benchmark for RGB-D Semantic Segmentation
- RGF: Recursive Generative Framework for the Edge-Cut Separable Problems
- RGTrack: Reliability-Guided Historical Prompt Purification for Robust Visual Tracking
- RG-VideoRAG: Reliability-Gated Information Gain Retrieval-Augmented Generation for Long-Form Video Understanding
- RHDSNet: Residual Hemodynamic Distillation For Breast Tumor Segmentation from Incomplete DCE-MRI with Only Single Pre-Contrast MRI
- RI-Bench: An Evaluation Framework for Role Integrity in Hierarchical LLM-based Multi-Agent Systems
- RICE-PO: Turning Retrieval Interactions into Credit Signals for Reasoning Agents
- RIDE: An Open Dataset and Benchmark for Train Delay Prediction
- Riemannian Admissibility Flow for Offline-to-Online Safe Reinforcement Learning
- Riemannian Flow Models with Reinforcement Learning for Molecular Crystal Structure Prediction
- Riemannian Lyapunov Framework: Optimization as Closed-Loop Control on Riemannian Manifolds
- Right Value, Wrong Rule: A Self-Verifying Sudoku Benchmark for Step-Level Rule Attribution
- RigidFormer: Learning Rigid Dynamics using Transformers
- RigRecon: Efficient Rig-Aware Street Reconstruction via Dual-Path Spatio-Temporal Interaction
- RILA: A Radar-Native Structured Interface from Sparse mmWave Point Clouds to Large Language Models
- RIME: Rival-Margin Evidence for Zero-Shot Fine-Grained Recognition
- RipBench: A Unified Benchmark for Multi-Level Rip Current Detection, Classification and Segmentation
- Ripple Cascade Prediction: Learning Causal Cooperative Propagation in Privacy-Preserving Social Systems
- Ripple in Still Water: Zero-Shot Clustering in Heterogeneous Federated Learning with Wavelet Scattering Transform
- RISE: Reliable Improvement in Self-Evolving Vision-Language Models
- RiSE: Residual Subspace Expert for Generalizable Text-Centric Image Forgery Localization
- Rising Tide: Overcoming Learnt Exploration Avoidance
- Risk-Averse Decision-Focused Learning: Minimizing CVaR of Regret
- Risk-Aware Hierarchical Fallback for Degraded Image Classification
- Risk-Calibrated Candidate-Set Decoding for Duplicate-Aware Object Detection
- Risk-Controlled Detection of Hallucinated References in Large Language Models
- Risk-Controlled Post-Processing of Decision Policies
- Risk Controlled Safe Stopping for Sequential Data Acquisition
- Risk-Sensitive Non-Parametric and Kernel Bandits
- RiskShield-VLA: Path-Dependent Safety via Directed Hazard Occupancy
- Risk Under Pressure: Compute-Aware Evaluation of Adversarial Robustness in Language Models
- RiT: Vanilla Diffusion Transformers Are Enough in Representation Space
- RLAR: An Agentic Reward System for Multi-task Reinforcement Learning on Large Language Models
- RLDF: A Region-Level Deepfake Dataset and Benchmark for Partial-Face Manipulation Detection
- RLearner-LLM: Balancing Logical Grounding and Fluency in Large Language Models via Hybrid Direct Preference Optimization
- RL-Guided Contraction of Symbolic Tensor Networks for Quantum Circuit Equivalence
- RL-Guided Temporal Localization for Dual-Channel Retrieval in Long-Horizon Agent Memory
- RL Optimization Amplification: A Public-Safe Boundary Model and Full Controlled Benchmark Results
- RNNs with equivalent attractor landscapes can implement distinct neural algorithms
- RoboEvolve: Co-Evolving Planner-Simulator for Robotic Manipulation with Limited Data
- RoboJailBench: Benchmarking Adversarial Attacks and Defenses in Embodied AI Systems
- RoboMemArena: A Comprehensive and Challenging Robotic Memory Benchmark
- RoboPAD: Post-Training Adaptation of Robot Foundation Models
- Robot Harmonic Product Networks: Constraint-Aware Learning of Robot Kinematics and Dynamics
- Robot Learning with World Models: Capabilities, Frontiers, and Challenges
- Robots as Deep Data Machines
- Robust Actor-Critic Learning under Distributional Uncertainty
- Robust Amortized Simulation-Based Inference via Learned Error Models
- Robust and Efficient Continual Model Merging via Global Singular Subspace Separation and Restoration
- Robust and Fast Training via Per-Sample Clipping
- Robust and Scalable Collaborative Learning via Pull-Based Epidemic Communication
- Robust and Scalable Multi-bit LLM Watermarking via Coded Payload Spreading
- Robust Approximate Nearest Neighbor Search for Any Dataset
- Robust Behavior Cloning Via Global Lipschitz Regularization
- Robust Concept Unlearning in Diffusion Models via Directional Stability Regularization
- Robust Continual Learning under Synthetic Contamination with Pre-Trained Models
- Robust Convex Decomposition-based Mesh Reconstruction from the Point Cloud
- Robust Flow Matching under Target Corruption and Label Noise
- RobustGenBench: A Benchmark for Robust Generalization to Adversarial and Common Perturbations, with Applications to Vision and Vision-Enabled Large Language Models
- Robust Graph Diffusion Model
- Robust High-Dimensional Sufficient Dimension Reduction via Stein’s Identity and Score-Based Inverse Regression
- Robust Individual and Group Fair Classification
- Robust Inference-Time Steering of Protein Diffusion Models via Embedding Optimization
- Robust Multimodal Information Bottleneck for Learning Minimal Sufficient Invariant Representations
- Robustness in RLHF via Trimmed-Mean Aggregation under Adversarial Contamination
- Robust Path Attribution in Piecewise Linear Neural Networks via Dummy-Constrained Optimization
- Robust Policy Evaluation in Image-Based Contextual Bandits via Vision Transformers
- Robust Reasoning Benchmark
- Robust Reinforcement Learning under Heavy-Tailed Dynamics
- RobustSCI: Beyond Reconstruction to Restoration for Video Snapshot Compressive Imaging under Real-World Degradations
- Robust Shielding for Safe Reinforcement Learning
- Robust stochastic first order methods in heavy-tailed noise via medoid mini-batch gradient sampling
- Robust Stream Classification using Time Neutralising Decision Trees
- Robust Surrogate Modeling for Explainable Graph Neural Networks
- Robust Task-Aware State Estimation from Pre-trained Perception
- RoCoMatrix: Coordinating Multi-Agent Manipulation with Compositional Collaborative Constraints
- Rollout-Calibrated Graph Caching for Audio Flow Matching
- Rollout Cards: A Reproducibility Standard for Agent Research
- Rollout fit is not enough for surrogate selection over frozen world-model context
- Rollouts at the Boundary: Value-of-Information Allocation for Test-Time Reinforcement Learning
- Root Cause Analysis of Measurement and Mechanistic Anomalies
- ROSE: Readout-anchored Operator Steering and Explanation for 3D Vision Foundation Models
- ROSE: Risk-Aware Orthogonal Subspace Navigation for Lifelong Knowledge Editing in Multimodal Large Language Models
- ROSE: Role-Aware Routed Update Scheduling for Efficient MoE Diffusion LLMs
- RoSeViT: Role-Separated Vision Transformers for ARC Visual Reasoning
- RotaryQuant: Fitting 120B MoE Models on Consumer Hardware via Fused Compressed-Space Attention
- Rotation-Compatible Tensor Features for Hierarchical Multi-Index Models
- Rotations on Latent Hyperspheres: a Geometry-Aware Guiding Framework for Diffusion Models
- Rotting Multi-Agent Bandits over Random Networks
- Rough Sets Meet Conformal Prediction: A Model-Free Nonconformity Score from Indiscernibility
- Roundtable Policy: Confidence-Weighted-Consensus Aggregation Improves Multi-Agent-System Reasoning
- Routeba: Refine Motion Intent and Dynamics States with Spatial and Temporal Routing
- RouteCast: Instance-Dependent Context Routing for Time Series Forecasting
- RouteOR: Disjunctive Composition of Pretrained Diffusion Experts via State-Dependent Routing
- RouterGen: Towards Substantial Router Generalization via Pseudo-Labeling and Feature Disentanglement
- Routers Learn the Geometry of Their Experts: Geometric Coupling in Sparse Mixture-of-Experts
- Routing as a Singular Reparameterization: A Closed-Form Pushforward Prior and Exact Bayesian Complexity in a Minimal Proxy
- Routing-Invariant Offline Evaluation of Flow-Matching VLAs via Oracle-Anchor Inversion
- Routing Safety Signals to Right Neurons: Multilingual Safety Alignment via Targeted Neuron Masking
- Row-Private Symmetric Cone Programming: Scale-Efficient Algorithm and Lower Bound
- RRD: Routing-and-Residual Distillation for Efficient MoE Recovery in Large Language Models
- RSA: Recursive Sparse Attention with Hierarchical Deep–Shallow Memory and Sparse Activation
- RSDA: Rethinking Unsupervised Domain-Adaptive Object Detection via Riemannian Statistical Alignment
- RSFaith-Bench: When Correct Answers Come with Unfaithful Evidence in Remote Sensing MLLMs
- RSPReg: Reliability-aware Structural Prototype Learning for Point Cloud Registration
- Rubric-Guided Self-Distillation: Post-Training Without Rubric Verifiers
- RUBRIC-MME: Real-User Behavior-grounded Rubric for Multimodal Interaction Capability Evaluation
- Rubric Optimization for LLM-Based Automated Grading with Data Synthesis
- RuleSmith: Multi-Agent LLMs for Automated Game Balancing
- Runtime Analysis of Cartesian Genetic Programming on MAX: A Proven Exponential Speedup
- Runtime Monitoring of Perception-Based Autonomous Systems via Embedding Temporal Logic
- Run-Wise Certificates for Agent Decision Accountability: Auditable Routing on Tool-Use DAGs
- RupeeBias: Auditing Demographic Bias in Indian Economic Guidance from Large Language Models
- RustMizan: A Compilable, Contamination-Aware Benchmarking Framework for Rust Vulnerabilities
- Rust-ortho: A Multi-Agent Framework for Project-Level C-to-Rust Translation via Functional Abstraction
- RVPO: Risk-Sensitive Alignment via Variance Regularization
- RVVCoder: ISA-Grounded LLMs for High-Performance RISC-V Vector Kernel Generation
- S$^{2}$-PINN: Stochastic Separable Physics-Informed Neural Networks
- S2T-RLHF: Hierarchical Credit Assignment for Stable Preference-Based RLHF
- S3I: Score-based Structured Scene Inference for Analysis-by-Synthesis Visual Perception
- SABER: State-Aware Budget Estimation and Routing for Sparse Attention
- SaDiT: Efficient Protein Backbone Design via Latent Structural Tokenization and Diffusion Transformers
- SAE Interventions are Unreliable: Post-Intervention Recovery of Suppressed Behavior
- Safe Actions Can Form Unsafe Traces: Benchmarking and Shielding Compositional Emergent Risk in AI Agents
- Safe and Generalisable Hierarchical Multi-Agent Reinforcement Learning via Constraint Manifold Control
- SafeCascade: Efficient Modality-Progressive Video Content Moderation
- SafeClawBench: An Operating-System Perspective on Evaluating the Security of Claw-like Agent Systems
- SAFE-DRIFT: Data Selection for Supervised Fine-tuning with Controllable Off-Target Drifts
- Safeguarding Multi-Agent LLM Debates with Temporal Intent Detection
- SafeGuide: Adaptive Inference-Time Safety Control for Diffusion Models
- Safe, or Simply Incapable? Rethinking Safety Evaluation for Phone-Use Agents
- SAFE-PACE: A Stability-Aware Frozen Euler Autoencoder with a Physics-Attention Coefficient Estimator
- Safe, Scalable, and Accurate Bayes Posterior Sampling for Large-Data Generalized Linear Mixed Models
- Safe Score Matching: Diffusion Policies with Hamilton-Jacobi Reachability for Online Safe Reinforcement Learning
- Safe Semi-Supervised Margin Distribution Network
- Safety Midtraining: Internalizing LLM Safety as a Foundational Capability
- Safety Modulation: Enhancing Reinforcement Learning Safety via Cost-Modulated Rewards
- Safety Steer: Internal Safety Interventions and Legal analysis of existing frameworks for diffusion models
- SAGAS: Semantic-Aware Graph-Assisted Stitching for Offline Temporal Logic Planning
- SAGE: A Self-Evolving Agentic Graph-Memory Engine for Structure-Aware Associative Memory
- SAGE: Evidence-First Biomarker Discovery through Multi-Agent Reasoning
- SAGE-FM: Gate-Certified Forecast Adaptation with Overlap-Conformal Reliability Repair
- SAGE: Retain-Aware Post-Hoc Sanitization of Final Unlearning Vector
- SAGE: Shapelet-guided Adaptive Experts for ECG-Language Model
- SahaBose-KFAC: Making KFAC Stable via Spectral Annealing and Curvature Condensation
- SAIL-DAVA: Data Asset Value Appraisal based on Sparse Attention and Imbalanced Learning
- SAIR: Cost-Efficient Multi-Stage ML Pipeline Autoscaling via In-Context Reinforcement Learning
- SALART-VQA: Diagnosing Whether VLMs Understand Salient Artifacts in Generated Images
- SaliMory: Orchestrating Cognitive Memory for Conversational Agents
- SALT: When More Rollouts Don’t Help in Group-Based Policy Optimization and How to Make Them Matter
- SAM3-CORE: Competitive Memory Readout for Robust Video Object Segmentation
- Same Architecture, Different Capacity: Optimizer-Induced Spectral Scaling Laws
- Same Words, Different Judgments: How Preferences Vary Across Modalities
- SA-MoGrasp: Scene-Aware Vision-Language Reasoning for Last-Meter Mobile Grasping
- Sample Complexity of Bias Detection with Subsampled Point-to-Subspace Distances
- Sample complexity of stochastic optimization with integer variables
- Sample Efficient Generative Molecular Optimization with Joint Self-Improvement
- Sample-Efficient Optimization over Generative Priors via Coarse Learnability
- Sample-Specific Model Selection for Crowd Counting via Dynamic Attention and Feature Fusion
- Sampling-Free Privacy Accounting for Matrix Mechanisms under Random Allocation
- Sampling from Flow Language Models via Marginal-Conditioned Bridges
- Sampling Is Not Curiosity: Why LLM Agents Should Investigate
- SAM-RISE: Black-Box Saliency for Vision-Language Models via Semantic Region Ablation, with Positive and Negative Evidence Attribution
- SANEval: Open-Vocabulary Compositional Benchmarks with Failure-mode Diagnosis
- Sanitize Before You Summarize: Memory Laundering as a Safety Blind Spot in LLM Agents
- Sanity Checks for Long-Form Hallucination Detection
- SARA: Semantically Adaptive Relational Alignment for Video Diffusion Models
- SarcBench: A Bilingual Benchmark for Contextual Sarcasm Understanding, Response, and Generation
- SASA: Subspace-Aware Sparse Autoencoders for Effective Mechanistic Interpretability
- SASContrast: Self-Supervised ECG Analysis for Stress Detection
- Satisficing Mixed-Strategy Equilibrium and Quantile Regret in Zero-Sum Games
- SATW: Sentence-Adaptive Text Watermarking for Efficient Multi-Bit Attribution
- SAVeR$^2$: Reasoning-based Safety Alignment for Large Reasoning Models via Verifiable Rewards
- Savitar: Curve-Aware Interaction-Structured Kernels for Low-Budget Bayesian Optimization in Rare-Winner Combinatorial Spaces
- Say When You Don’t Know: Training LLMs to Expose Self-Assessment
- SC$^3$: A Multi-Solvent Solubility Challenge and Benchmark
- Scaffold-Mediated Post-Training: Co-Evolving Model Parameters and Procedural Scaffold Graphs
- Scalable and Decentralized Training of Mixture of Experts with Collaborative Gating
- Scalable Attention via Lightweight Up-Projection
- Scalable Delphi: Large Language Models for Structured Risk Estimation
- Scalable Distributed Stochastic Optimization via Bidirectional Compression: Beyond Pessimistic Limits
- Scalable Fair Learning via Cramér-von Mises Regularization
- Scalable Forward-Forward Learning: Block-wise Contrastive Forward-Forward with Masked Autoencoder
- Scalable Identification of Higher-Order Interactions in Networked Dynamical Systems
- Scalable Knowledge Editing for Mixture-of-Experts LLMs via Tensor-Structured Updates
- Scalable Maximum Entropy Reinforcement Learning for Diffusion Policies via Adjoint Matching
- Scalable Multi-Agent Contrastive Reinforcement Learning
- Scalable, Noise-Aware Kernel Methods for Interferometric Image Reconstruction
- Scalable Oversight for Superhuman AI via Recursive Self-Critiquing
- Scalable Paraphrase-Robust Language Model Fingerprints via Secret Semantic Targets
- Scalable PCA Under Infinite-Variance Noise via Subspace Aggregation
- Scalable Supervised Optimal Transport of Gaussian Mixture Models
- ScaleAC: Scale Actor-Critic by Replay Ratio
- Scale-and-Shift Knowledge Injection for Online Continual Post-Training
- Scale-Aware Conformal Inference: Calibrating Error Feedback for Time Series Forecasting (Extended)
- Scale-Invariant Empirical-Bayes Laplace Approximation for ReLU Networks
- Scale-Invariant Graph Neural Networks for Cross-Scale Branching in Branch-and-Bound
- Scale SAE: Specialized Multi-Expert Sparse Autoencoders with Feature Scaling
- Scales++: Compute Efficient Evaluation Subset Selection with Cognitive Scales Embeddings
- Scale-Sensitive Shattering: Learnability and Evaluability at Optimal Scale
- Scale Where It Matters
- Scaling Agentic Capabilities, Not Context: Efficient Reinforcement Finetuning for Large Toolspaces
- Scaling Laws for Classical Machine Learning on Tabular Business Data: A Classroom-Scale Empirical Study
- Scaling Laws for HyperNetwork-Based Knowledge Injection in Large Language Models
- Scaling Laws for Strategic Interactions
- Scaling Limits of Long-Context Transformers
- Scaling Linear Mode Connectivity and Merging to Billion Parameter Pretrained Transformers
- Scaling Optimization-Oriented Hypernetworks for Implicit Neural Representations
- Scaling Point-in-Time Language Models: Economic Evaluation of Embeddings
- Scaling Reward Modeling without Human Supervision
- Scaling Wrist sEMG for Hand Pose Estimation
- SCAMP: Black-Box Knowledge Poisoning Attack on Multimodal Retrieval-Augmented Generation
- SCAR: Summary-Driven Consolidation Attacks on Rule Formation in Hierarchical Memory Agents
- SCASRec: A Self-Correcting and Auto-Stopping Model for Generative Route List Recommendation
- SCDBench: A Benchmark for LLM-Based Smart Contract Decompilers
- Scenario-Conditioned Policy Composition for Large-Scale Multi-Agent Spatial Coordination
- SceneGraphVLM: Dynamic Scene Graph Generation from Video with Vision-Language Models
- Scenes as Objects, Not Primitives : Instance-Structured 3D Tokenization from Unposed Views
- SceneScaffold: Active Scene-State Construction for Unified 3D Scene Understanding
- SceneScript: Text-Only Pre-Training Boosts LLM-based 3D Scene Understanding
- SceneShifter: Training-free Multi-Scene Temporal Control for Audio-driven Human Animation
- ScFlow: Spectrum-Aware Correction for Stabilizing Training-Free Spatial Control in Rectified Models
- Schedule Distillation: Learning Penalty Schedules for Resource-Constrained Shortest Paths
- SchemaVerse: A Benchmark for Learning Compositional Global Schema Construction
- ScholarGym: Benchmarking LLM Capabilities in the Information-Gathering Stage of Deep Research
- Scholarly Communication Should Employ World Models to Make Implicit Knowledge Explicit
- SCHTs: A Semi-Structured Dynamic Sparse Training Framework for Hardware-Efficient Deep Learning
- SCI-Defense: Defending LLM-Based Ranking Against Manipulation Attacks
- SciKernelBench: Evaluating LLM Kernels for Higher-Order Autodiff and Scientific Residuals
- SciReC: Diagnostic Evaluation of Multimodal Multi-Turn Relational Reasoning with Adaptive Interaction
- scMAS: A Multi-Agent System for Single-Cell Annotation with Foundation Model
- SCOPE: From Analytical Scale Oracles to Candidate Scale Search for NVFP4 Quantized Training
- ScopeLoRA: Coupling Spectral Structure and Uncertainty-Aware Low-Rank Adaptation
- ScopeSAE: Model-Scope Feature Discovery with Interpretable Layer Selection
- Score-based ab initio reconstruction from heterogeneous cryo-EM datasets
- Score-Based One-step MeanFlow Policy Optimization
- SCoRe: Candidate Semantic Denoising and Relational Reasoning for Training-Free Zero-Shot Composed Image Retrieval
- Score is Not Circuit: Evaluating Mechanistic Localization as a Score-to-Circuit Stack
- SCoReQ：Scalable Complex 2-bit Residual Quantization for Approximate Nearest Neighbor Search
- SCoRe: Smoothed Certified Release for Prompt Repair
- Score-Stability Theory for Robust Outlier Detection
- ScoringBench: A Benchmark for Evaluating Tabular Foundation Models with Proper Scoring Rules
- SCOUT: Multi-Modal Acquisition for Data-Efficient Closed-Loop Metal Additive Manufacturing
- ScrapeGraphAI-100k: Dataset for Schema-Constrained LLM Generation
- Screening Lipid Nanoparticles through Structure-Ratio Alignment
- ScreenTutor: Learning Computer-Use Agents from Unlabeled Tutorial Videos
- ScrewSet: A Large-Scale Benchmark for Fine-Grained Industrial Screw Classification and Corruption Robustness
- Script-based 3D Indoor Scene Generation
- Scroll Search: Model-Free Learning for Decentralised Control
- SCRYER: A scalable framework for forecasting neural population activity
- SCTI: Self-Calibrated Trident Identification of Black-Box LLM Watermarks
- SDFlow: Similarity-Driven Flow Matching for Time Series Generation
- SDFNet: Decomposition-Guided Support Distribution Forecasting for Long-Term Time Series
- SDHilb: Schrödinger Dynamics-Guided Neural Network with Adaptive Multi-Scale Hilbert Transform for Time Series Forecasting
- SD-LLaVA: A Dual-Stream Framework for Semantic-Augmented BEV Understanding
- SDP-VAE: Semantic-Guided Dynamic Feature Parsing for Image Generation
- SDR: Set-Distance Rewards for Radiology Report Generation
- SEAICE-BENCH: A Benchmark for Sea Ice Prediction and Analysis Capability of LLMs
- SEAL: Robust Semantic Watermarking for LLM-Generated Code
- SeaPilot: Mobile Agent with Self-refining Environment Alignment
- Search-Augmented Masked Diffusion Models for Constrained Generation
- Searching Videos as Trees: Self-Correcting Agents for Grounded Long Video QA
- Search More, Think Less: Rethinking Long-Horizon Agentic Search for Efficiency and Generalization
- Search-R3: Unifying Reasoning and Embedding in Large Language Models
- SEAR: Sample Efficient Action Chunking Reinforcement Learning
- SEAR: Segment-level Early Anomaly Recognition via Cross-Task Anticipation Gating and Temporal Contrastive MIL
- Season-Aware Hybrid Convolutional-Transformer for Antarctic Sea Ice Concentration Forecasting
- Second-Order Complexity Theory for Neural Networks
- Second-Order Drifting Models
- Second Workshop on MLxOR: Mathematical Foundations and Operational Integration of Machine Learning for Uncertainty-Aware Decision-Making
- SecPO: Principled Adversarial Training for Prompt Injection Security
- SecureForge: Finding and Preventing Vulnerabilities in LLM-Generated Code via Prompt Optimization
- Securing Large Language Model Agents via Structured Graph Abstraction
- SEED: Evaluating Adaptive Evaluator Reuse and Deployment Evidence for Bounded Prompt and Workflow Self-Evolution in Coding Agents
- Seeing Across Skies and Streets: Feedforward 3D Reconstruction from Satellite, Drone, and Ground Images
- Seeing Beyond the Next Step: World-Model-Guided Human-Like Navigation in Multi-Agent Scenes
- Seeing Further on the Shoulders of Giants: Knowledge Inheritance for Vision Foundation Models
- Seeing in 4D: From Monocular Video to an Immersive 4D Gaussian World
- Seeing Together: Multi-Robot Cooperative Egocentric Spatial Reasoning with Multimodal Large Language Models
- Seeing Without Exposing: Adaptive Privacy Control for Open-World, Context-Hungry MLLMs
- Seeking the Unfamiliar but Memorable: Conceptual Creativity as Meta-Learning
- Seen but Not Reported: Representation–Behavior Decoupling in Medical VLMs
- SEER: Stable Equivariant Encoders via Reconstruction
- SEFIR: Safety Detection in Frozen Multimodal Large Language Models via Internal Evidence Fusion
- Segment and Select: Vision-Language Segmentation in 3D Scenarios
- Segment-Based Multi-Hypothesis Modeling for Protein–Ligand MD Trajectory Generation
- SegRWKV: Linear-Complexity RWKV for Efficient 3D Medical Image Segmentation
- Seirênes: Adversarial Self-Play with Evolving Distractions for LLM Reasoning
- SE-KGC: Structure-Enhanced Knowledge Graph Completion with Graph Semantics and Higher-Order Topology
- Selection, Not Fusion: Radar-Modulated State Space Models for Radar-Camera Depth Estimation
- Selective Exposure of Hidden State for Low-Budget Manipulation Transfer from Passive Video
- Selective Latent Thinking: Adaptive Compression of LLM Reasoning Chains (Extended)
- Selectively Scarce Data with Domain Knowledge is an Underexplored Problem in RL
- Selective Machine Unlearning for Vision Transformers via Attention-Guided Contrastive Learning
- Selective Out-of-Distribution State Correction via Behavior Density in Offline Reinforcement Learning
- Selective Prediction Reduces the Negative Effects of Automation Bias Overall but Increases False Negatives
- Selective Unlearning Depends on How Knowledge Is Stored
- Selective Unlearning in Multi-Agent Deep Deterministic Policy Gradient
- Selectivity and Shape in the Design of Forward-Forward Goodness Functions
- SelecT: Selection-based Spatio-Temporal Video Grounding with MLLMs
- Self-Agent System for Reflective Reasoning
- Self-attention Re-weighting (SAR): Rethinking Attention for Efficient Vision Transformer Tuning
- Self-Blinding and Counterfactual Self-Simulation Mitigate Biases and Sycophancy in Large Language Models
- Self-Consistent Gradient Structure Regularization for Improved Generalization in Neural Networks
- Self-Consistent Latent Reasoning: Long Latent Sequence Reasoning for Vision-Language Models
- SelfCritic-VLA: Language as Intrinsic Critic for Vision Language Action Models in Autonomous Driving
- Self-Distillation Narrows the Learning-Efficiency Gap
- Self-Diversifying Regularization for Robust Out-of-Distribution Detection
- Self Driving Datasets: From 20 Million Papers to Nuanced Biomedical Knowledge at Scale
- Self-Evolving Audio Reasoners: When Reasoning Models Become Their Own Critics
- Self-Evolving Diversity-Driven Search for Robust AI Systems
- Self-Evolving Modular World Model for Embodied Long-Horizon Tasks in Dynamic Environments
- Self-Evolving Multi-Agent Systems via Decentralized Memory
- Self-Evolving Spatial Belief for Video Spatial Reasoning
- SelfHalt: Learning to Stop — Autonomous Halting in LLM Reasoning via Completion Attractor Steering
- Self-Hinting Language Models Enhance Reinforcement Learning
- Self-Improvement of Language Models by Post-Training on Multi-Agent Debate
- Self-Learning Digital Twins for Sustainable Peat Agriculture Using Physics-Informed Graph Neural Networks
- Self-Organized Conformal Prediction: Reducing Regional Coverage Gaps with Unsupervised Group Discovery
- Self-Pruned Key-Value Attention: Learning When to Write by Predicting Future Utility
- Self-Refined Distillation
- Self-ReSET: Learning to Self-Recover from Unsafe Reasoning Trajectories
- Self-Rewarded Multimodal Coherent Reasoning Across Diverse Visual Domains
- Self-Supervised Combinatorial Optimization with Constraints via Frank–Wolfe
- Self-Supervised Denoising for Single Volume MR Angiography via Slice-Aware Diffusion
- Self-Supervised Discovery of Latent 3D Rotations
- Self-Supervised Learning with a Multi-Task Latent Space Objective
- Self-supervised pretraining for an iterative image size agnostic vision transformer
- Self-supervised Residual Neural Kalman Filtering with Stability Guarantees
- Self-Taught Policy Improvement for Vision-Language Game Agents
- Self-Triggers: Backdoor Attacks via Internal Representation Poisoning
- Self-Tuned Q-Learning and SARSA: Policy Control Without Worrying About Step-Size Calibration
- Self-Verifying Agents: Counterfactual Tool-Call Auditing for Robust Multi-Step Reasoning
- Semantically Complementary Spectral Views Learning for Graph-Level Anomaly Detection
- Semantic Calibration of Media Streams
- Semantic Coherence Decomposition for Reliable Large Language Model Reasoning
- Semantic Compass: Self-Check Test-Time Scaling for Prompt-Faithful Image Generation
- SemanticDialect: Semantic-Aware Mixed-Format Quantization for Video Diffusion Transformers
- Semantic Field Guidance for Transferable Targeted Attacks on Closed-Source MLLMs
- Semantic-Guided Exposure Correction for Context-Aware Brightness Adjustment
- Semantic-Guided Expression-Invariant Kinship Verification
- Semantic Human Motion Prediction TBD
- Semantic-Level Invariant Representation Learning for Cross-Hospital Clinical EEG Modeling
- Semantic Motion Anchors: Bridging Motion and Meaning in Co-Speech Gestures
- Semantic Online Facility Location for Agentic Systems
- Semantic Priors Meet Statistical Evidence: Robust Forests for Few-Shot Tabular Learning
- Semantic Robustness Auditing for VLMs
- Semantic Shadowing: A Chaos-Inspired Framework for Output Variability in LLMs
- SEMA: Semantic Factors for Skill Discovery
- SemaVoice: Semantic-Aware Continuous Autoregressive Speech Synthesis
- SemiFLOYD: Semi-Supervised Federated Learning with YOLO-based Detection for Unlabeled Client Data
- Semiparametric Efficient Bilevel Gradient Estimation
- Semi-Supervised Conditional Independence Testing in High-Dimensional Partial Linear Models
- Semi-Supervised Multi-Domain Learning for Vision-Language Models
- SemMamba: Semantic Control by Dynamic Modulation for Time-Series Forecasting
- SENSE: Semantic Neural Speech Synthesis from Brain Dynamics via Spatial Graph Encoding
- SensorCI : A Contextual Integrity Benchmark for Privacy in Industrial Time-Series
- SENTINEL: A Multi-Level Formal Framework for Safety Evaluation of Foundation Model-based Embodied Agents
- Separating Persistence from Accessibility in Long-Term Agent Memory
- Separator Profiles and Regularity Barriers for Direct Gaussian Transport
- Seq-LoRA: Sequential Bayesian Low-Rank Adaptation for Large Language Models
- Sequential Certification of LLM Evaluators with Logit-Free Drift Triggers
- Sequential Grouping Enables Efficient Test-Time Alignment for Closed-Loop LLM-based Motion Planning
- Sequential Local Operator Alignment for Training-Free Model Merging
- Sequentially-Controlled Interactive Multi-Particle Flow-Maps for Online Feedback-Driven Search
- Sequential Policy Learning for Optimising 3D Deformable Registration under Region Constraints
- Sequential Probability Assignment against Smoothed Adversaries with Unknown Base Measure
- Sequential Subspace Noise Injection Prevents Accuracy Collapse in Certified Unlearning
- Sequential Testing for Expert Birth in Self-Evolving Mixture-of-Experts
- Serialization Tax in Shared-Latent Exchangeable Decisions
- Serving a Free Lunch for Fine Grained 3D Geometry via Entropy Guided Attention
- Settling the Sample Complexity of Deterministic Agnostic PAC Learning
- Set-Valued Policy Learning
- SFA Transformer: Rethinking Visual Representation Learning with Spiking Fourier Attention
- SF-DST: Adapting Vision-Language Models for Anomaly Detection via Asymmetric Modulation and A-LoRA
- SFT-then-RL Outperforms Mixed-Policy Methods for LLM Reasoning
- SGD at the Edge of Stability: The Stochastic Sharpness Gap
- SGD in Multiclass Logistic Regression: Sequential Learning and Scaling Laws
- SGMamba: Semantic-Guided Mamba for 3D Human Pose Estimation
- SH$^2$: A Mathematician-Curated benchmark for Assessing Research-level Math Capabilities of LLMs
- ShadFormer: Quantum Ground State Prediction over Hamiltonians and Scales from Classical Shadows
- ShadowFM: A Foundation Model for Quantum Property Estimation from Classical Shadows
- ShadowTransfer: A Geographic Transfer Benchmark for Overhead Shadow Detection
- SHAKY PREPEND: A Multi-Group Learner with Improved Sample Complexity
- ShapVal: Cooperative Workload-Conditioned Mixed-Precision Quantization for MoE via Shapley Residual Interpolation
- Shared Canonical Component Removal Beyond Expert Views
- Shared-Coordinate Prompt Selection for Domain Adaptation of Vision-Language-Action Models
- Shared Neuro–Visual Latent for Brain Decoding and Discovery
- Shared-Response Agreement Fabrication: A Model-Side Attack on Multi-Explainer Visual Explanations
- Sharp Finite-Sample Rates for Identifiability in Causal Representation Learning
- Sharpness-Aware Muon: Matrix-Aware Optimization with Momentum-Guided Perturbations
- Sharp Optimal Algorithm for Derivative-Free Stochastic Convex Optimization in One Dimension
- Shattering the Autoregressive Curse: Dynamic Epistemic Entropy Orchestrated Erasable Reinforcement Learning for LLMs
- Sheaf Neural Networks with Learnable Grothendieck Topology
- SHERPA: Seam-aware Harmonized ERP Adaptation for Open-Domain 360$^\circ$ Panorama Generation
- ShifaMind: A Multiplicative Concept Bottleneck for Interpretable ICD-10 Coding
- Shifting AI Efficiency from Model-Centric to Data-Centric Compression
- ShipBench: A Drawing-Grounded VLM Benchmark for Ship Structural Reasoning
- SHiPPO: Recurrent Memory with Transported Polynomial Projections
- Shortcut or Signal? Diagnosing Transferable PPG Morphology in Cuffless Blood Pressure
- Should We Pay This Much for Robustness? Efficient Proxy Certificates with Marginal Guarantees
- Show Your Work in Swahili: Reinforcement Learning for Native-Language Mathematical Reasoning
- Shutdownable Agents through Length-Neutral Policy Optimization
- SIFT: Sequence-Informed Fine-Tuning via Token-Level Outlier Detection
- SIGA: Scientific Simulation Coding Agent Adapter- A Geophysics Case Study
- SIGM: Size-aware Integer-grid Maps for Adaptive Hexahedral Mesh Generation
- Signal-Adaptive Trust Regions for Gradient-Free Optimization of Recurrent Spiking Neural Networks
- SignalBench: Comparing Dense Feedback Methods for Long-Horizon Agents
- Signal-Guided Discrete Diffusion for Categorical Inverse Problems
- Signal or Noise? Understanding Generative Models for Real-World Sensor Time Series
- Signature Approach for Contextual Bandits with Nonlinear and Path-dependent Rewards
- Sign-Aware Gated Sparse Autoencoders: Modeling Anticorrelated Features with Bi-Jump-ReLU Activations
- SignCtrl: Controlling Diffusion Dynamics for Sign Language Production
- SignRot: LLM Quantization with Massive Outlier-Aware Sign-Adjusted Rotation
- Signs Beat Floats: Low-Rank Double-Binary Adaptation for On-Device Fine-Tuning
- Sign Structure in Ternary Neural Network Weights
- SignVerse-2M: A Two-Million-Clip Pose-Native Universe of 55+ Sign Languages
- SILAGE: Memory-Efficient, Full-Gradient-Free Nonconvex Optimization for Nested Finite Sums
- Silence is Strategic: Empowering Agents to Act with Minimal User Reliance
- Sim2Science: ML with Imperfect Scientific Models
- Similar Code, Shared Neuron: Rethinking Sparse Representation in Refactoring-Aware Code Retrieval
- Simple and Better Algorithms for Individually Fair Diversity Maximization
- Simple and Effective Query-Adaptive Sparse Attention
- Simple Baselines are Competitive with Code Evolution
- SimpleHRL: Minimal Hierarchical Structure Improves Out-of-Distribution Generalization in RL Post-Training of LLMs
- SimplexFlow-GRPO: Flow-Matching Speech Synthesis with Hybrid ODE-SDE Sampling and Adaptive Weighted Multi-Objective Reinforcement Learning
- SimPlex-GT: Node-to-Cluster Attention for Graph Learning under Mixed Homophily and Heterophily
- Simply Stabilizing the Loop via Fully Looped Transformer
- SimReadyNet: PDE-Aware Learning for Simulation-Ready Vascular Segmentation
- Simulating Students or Sycophantic Problem Solving? On Misconception Faithfulness of LLM Simulators
- Simultaneous Individual, Group and Multigroup Fairness in Set Covering Problems
- SimVerse: Benchmarking Inverse Mental Simulation in the Physical World via Interactive Games
- SinAE: A Single-Architecture Flow-Matching Autoencoder for Cross-Domain Atomic Systems
- Single-cell foundation models lose regulatory signal under perturbation fine-tuning: a cross-layer audit and a rank-collapse account
- Single LLM Debate, MoLaCE: Mixture of Latent Concept Experts Against Confirmation Bias
- Single Metrics Are a Trap
- Single-Minima Neural Network for Modeling Lyapunov Functions
- Single-Run Computable Random-Subset KL Bounds for SGLD
- Sink Before You Speak: Manipulating LLM Generation via Prompt-specific Representation Subspaces
- Sinkhorn Information Theory
- SIREM: Speech-Informed MRI Reconstruction with Learned Sampling
- SiTe: A Statistical Significance-Aware Framework for Robust XAI
- SiT-v2: Pretrained Surface Transformer Improves the Brain Decoding of Images, Videos, Sounds and Language
- SizingAgent: Incentivizing Physics-Grounded Reasoning in LLM-Based Analog Circuit Sizing
- Sketching the Readout of Large Language Models for Scalable Data Attribution and Valuation
- Skill1: Unified Evolution of Skill-Augmented Agents via Reinforcement Learning
- Skill-Aligned Annotation for Reliable Evaluation in Text-to-Image Generation
- Skill-Coupled Policy Optimization with Calibrated Group-Wise Advantage Estimation
- SkillFlow: Flow-Driven Recursive Skill Evolution for Agentic Orchestration
- SkillForge: Co-Evolving Skills and Agents via Dynamic Skill Lifecycles
- SkillGenBench: Benchmarking Skill Generation Pipelines for LLM Agents
- SKILLGRAPH: Skill-Augmented Reinforcement Learning for Agents via Evolving Skill Graphs
- SkillGrid: A Unified Framework for Evaluating Skill Composition in LLM Agents
- Skill-Inject: Measuring Agent Vulnerability to Skill File Attacks
- Skill-in-Skill: A Skill-Native Approach to Auditable Coding Agents
- SkillLens: Diagnosing the Efficacy of Agent Skills in Language-Agent Benchmarks
- SkillMaster: Toward Autonomous Skill Mastery in LLM Agents
- SkillPatchBench: Repairing Hidden Vulnerabilities in Agent Skills While Preserving Utility
- Skill-SD: Skill-Conditioned Self-Distillation for Multi-turn LLM Agents
- SkyLink-SC: Unified Multi-Task Semantic Communication for Low-Altitude Applications
- SlackBench: Benchmarking Agents on Collaborative Projects Grounded in Real Code Repositories
- SlakoNet DB: a paired bandgap benchmark for tight-binding and graph neural networks across crystal regimes
- SL Algebra: A Control IR for Structured, Grounded, and Auditable LLM Inference
- SLayerGen: a Crystal Generative Model for all Space and Layer Groups
- SLAY: Geometry-Aware Spherical Linearized Attention with Yat-Kernel
- SLAyiNG: A Diverse and Community-validated Dataset of Queer Slang
- SLBRD: A Single-Loop and Byzantine-Resilient Decentralized Bilevel Algorithm
- SLDR: Defending Against Malicious Fine-tuning via Selective Layers Recovery and Dynamic Routing
- SLEIGHT-Bench: A Benchmark of Evasion Attacks Against Agent Monitors
- Sliced Wasserstein Meets Quantum Optics: Provable Wavefunctions Tomography with Scarce Noisy Measurements
- SliceGraph: Mapping Process Isomers in Multi-Run Chain-of-Thought Reasoning
- SLIC: Reinforcement Fine-Tuning Small LMs for Multi-Turn Analog Circuit Optimization
- SLIDERS: Systematic Reviews via Automated Evidence Synthesis and Reconciliation
- SLiDE: Structured Linear Dynamics for Forecasting with Exogenous Inputs
- SLM-Agents: 1st Workshop on SLMs for Agentic Systems
- SlopCodeBench: Benchmarking How Coding Agents Degrade Over Long-Horizon Iterative Tasks
- SLS-Bench: A Benchmark for Incident Log Summarization with Synthetic Observability Data
- Small Experts, Big Students: Distilling Long-Horizon RL Policies into LLM Agents via Imitation Learning
- Small Models, Strong Priors: Architectural Inductive Bias for Parameter-Efficient Neural PDE Solvers
- SmartEval: A Benchmark for Evaluating LLM-Generated Smart Contracts from Natural Language Specifications
- Smart Picks in the Dark: Towards Efficient RLVR for Reasoning via Tracing Metacognitive Pivots
- SMC-GU: A Scalable and Model-agnostic Certified Unlearning Framework for Graph Neural Networks
- SMEARGLE: Sketch the Draft, Skip the Attention (Extended)
- SMEBench: Benchmarking Neural Music Editing with Synthesized MIDI Edits
- SMILE: Bridging Continuous Optimization and Discrete Symbolic Recovery
- SMMBench: A Benchmark for Source-Distributed Multimodal Agent Memory
- sMMC-22M: A Context-Aware Dataset and Benchmark for Single-Cell Spatial Transcriptomics
- SMOG: Scalable Meta-Learning for Multi-Objective Bayesian Optimization
- SMolLM: Small Language Models Learn Small Molecular Grammar
- Smoothed Score Queries and the Complexity of Sampling
- Smoothing Dark Areas in Molecular Latent Diffusion
- Smooth Partial Lotteries for Stable Randomized Selection
- Smooth Piecewise Cutting for Neural Operator to Handle Discontinuities and Sharp Transitions
- SnapAudit: Active Auditing of Differentially Private In-Context Learning via Snapshot-Based Simulation
- SNAP: Segment Nuclei via Automatic Prompting
- SnareNet: Flexible Repair Layers for Neural Networks with Hard Constraints
- SOAP-Bubbles: Effective and Scalable Variational Learning with Structured Covariances
- SOAR: Scale Optimization for Accurate Reconstruction in NVFP4 Quantization
- SOAR: Semantic Organ-Aware Pretraining for 3D CT Image Understanding
- SocialAgent: Second Workshop on Large Language Models for Social Reasoning and Simulation
- Social Laws for Multi-agent Coordination in Stochastic Environments
- Social Matching Bandits: Learning with Network-Mediated Preferences
- SocialPilot: A Simulated Dataset and Benchmark for Visual Navigation in Dynamic Human Environments
- Social Structure Matters in 3D Human-Human Interaction Generation
- Social Theory Should Be a Structural Prior for Agentic AI: A Formal Framework for Multi-Agent Social Systems
- SocialWeave: Generating Spatially Coherent Multi-Human Interaction
- Socio-Conformal Calibration in Complex Survey Data: Marginal Validity Is Not Enough for Subgroup Reliability
- SocioTalk: Decoupling Physical Articulation and Social Dynamics for 3D Dyadic Conversational Head Generation
- Soft Forward-Backward Representations for Zero-shot Reinforcement Learning with General Utilities
- SofT-GRPO: Surpassing Discrete-Token LLM Reinforcement Learning via Gumbel-Reparameterized Soft-Thinking Policy Optimization
- Softmax Regression for Continuous-Time Competing Risks: A Neural Network Approach with Time Augmentation
- SoftPQ: Robust Instance Segmentation Evaluation via Soft Matching and Tunable Thresholds
- SoftStep: data-adaptive sparse similarity weighting improves deep neighbor-based regression
- Soft Token Alignment for Cross-Lingual Reasoning
- Soft Vector Quantization via Entropy Regularized Neural Activation Patterns
- SOLA: Second-Order Link Adaptation from ACK/NACK Feedback
- Solver-as-Teacher: Solver-Guided On-Policy Post-Training Framework for PDE Foundation Models
- Solver-Free Verification of Neural Barrier Certificates via LP-Free ReLU Region Enumeration
- Solving the Pre-image Problem in Kernel PCA: The Reproducing Property is All You Need
- Sophon: A Procedural Diagnostic for Spatial Reasoning in Vision-Language Models
- SOrbitBench: A Large-Scale Multi-Regime Benchmark and Evaluation Pipeline for Satellite Orbit Prediction
- SOTA or Luck? The Winner’s Curse in LLM Leaderboards
- Sound Probabilistic Safety Bounds for Large Language Models
- Source-Free Domain Adaptation for Semantic Segmentation in Open-Set World
- SP$^2$ec: Adaptive Self-Speculative Decoding for Vision-Language Models
- SpaceMind++: Toward Allocentric Cognitive Maps for Spatially Grounded Video MLLMs
- Space-Optimal Streaming Algorithms via Efficient Encodings
- Spade-SLAM: Sparse Correspondence Anchors for Dense Self-Supervised Visual SLAM
- SPADE: Stochastic Payoff-Based Algorithm for Decentralized Equilibria in Decision-Dependent Games with Coupled Constraints
- SpaHetG: A Heterogeneous Graph Neural Network for Spatial Multi-Omics Integration of Transcriptomics and Metabolomics
- SpanFormer: Multi-Level Adaptive Sparsity for Object Detection in High-Resolution Wide Shots
- SPAN - Shapley pruned autoregressive network for chaotic attractor reconstruction
- SPANUQ: Span-Level Uncertainty Quantification for Large Language Model Generation
- SPARCE: Sparse Counterfactual Attribution via Gradient-Guided Diffusion for Time-Series Root Cause Analysis
- SPARC-RTL: Stochastic Prompt-Assisted RTL Code Synthesis
- SpaRFL: Sparse Low-Rank Federated Learning for Memory-Constrained Clients
- Spark: Path-Aware Experiential Self-Evolution for VLMs Spatiotemporal Reasoning
- SPARK: Spectral Plasticity via Adaptive Rank-preserving Kernels
- SPARK: Structural Priors Activated from closed-set Recognition Knowledge for On-the-Fly Category Discovery
- Sparse Autoencoders for Interpretable Out-of-Distribution Detection
- Sparse Hemodynamic Field Recovery in Intracranial Aneurysms via Physics-Informed Neural Networks
- SPAR: Self-Referential Prompt-Conditioned Attention Reweighting for VLM Decoding
- Sparsely-gated tiny linear experts
- Sparsely-Supervised Data Assimilation via Physics-Informed Schrödinger Bridge
- Sparse Relational Bottlenecks for Ambiguity-Heavy Dense Registration
- Sparse Shift Autoencoders for Identifying Concepts from Large Language Model Activations
- Sparse-View Interpretable 3D Animal Behavior Representations for Neural Encoding and Decoding
- SparseWake: A Temporal Hydrodynamic Sensing Benchmark for Close-Neighbor State Estimation
- Sparsity Annealing Improves the Performance of Cannistraci-Hebb Dynamic Sparse Training
- Sparsity for Free: A Budget-Induced Equilibrium in Joint Topology–Parameter Search
- SpatialAct: Probing Spatial Reasoning-to-Action Capabilities of VLM Agents in 3D Scenes
- Spatial Gene Expression Prediction as Field Reconstruction from Histology
- Spatial-IQ: Deconstructing Spatial Intelligence via Hierarchical Capability Tests
- Spatial Localized Approach for LLM Watermarks Detection under Human Edits
- SpatialOracle: Geometry Supervision for Spatial Orientation Reasoning in VLMs
- Spatial Representation Distillation and Knowledge Routing for Vision-Language-Action Models
- SpatialScaffold: Active Construction of 3D Structure in Text Space for Multimodal Spatial Reasoning
- SpatialWorld: Benchmarking Interactive Spatial Reasoning of Multimodal Agents in Real-World Tasks
- SP-CACW: Convergence-Aware Client Weighting for Selfish Personalized Learning
- SPDAlign: An Interpretable Alignment Framework for EEG Forward Modeling Shifts
- SPD: Skeleton-Preserving Distillation from Quadratic to Linear-Time Vision Models
- SpecAttent - Spectral Attention Representations in Dense Heterophile Graphs
- SpecGuard-Chem: Oracle-Compiled Evaluation Contracts for Scientific Language Agents
- Specialists Hold, Generalists Discount: Asymmetric Equilibrium in LLM Routing Auctions
- Specialize Roles, Mix Deployments: Pushing the Cost-Accuracy Frontier of LLM Agent Teams
- SpecKV: Frequency-Domain KV-Cache Compression for Autoregressive Time-Series Foundation Models
- SPECS: Faster Test-Time Scaling through Speculative Drafts and Dynamic Switching
- Spectral Collapse in Diffusion Inversion
- Spectral Conflict: the Difficulty of Joint Continual Learning and Unlearning via Low-Rank Adaptation
- Spectral Flattening Is All Muon Needs: How Orthogonalization Controls Learning Rate and Convergence
- Spectral Flow Certificates for Depth-Aware Long-Range Propagation in Graph Neural Networks
- Spectral Generator Neural Operator for Stable Long-Horizon PDE Rollouts
- Spectral Geometry of Attention: From Information Routing to Uncertainty
- Spectral Gradient Surgery for Domain Generalizable Dataset Distillation
- Spectral Homeostasis in Recommender Systems
- SpectralHop: Efficient Unsupervised Graph Anomaly Detection With Chebyshev Structure Encoders
- SpectralKV: Redundancy-Aware KV Cache Compression via Spectral Coreset Selection
- Spectral Mamba: Decoupled State Space Models for Multispectral Continual Learning
- Spectral Measures of Mamba Conductances Predict and Shape Effective Receptive Fields
- Spectral Power Law: Understanding Concentration Dynamics of MuonSAM with Spectral Deviation
- Spectral Progressive Diffusion for Efficient Image and Video Generation
- SpectralQuant: Low-Effective-Rank KV Keys Enable Calibrated Cache Compression
- Spectral Structure in Trained Neural Operators: Power-Law Heads and Marchenko--Pastur Bulks
- Spectral Superposition: An Operator-Theoretic Lens on Feature Geometry
- Spectral Transformer Neural Processes
- Specula: Speculative Recovery for Disaggregated LLM Inference via Compressed KV Checkpoints
- Speculate with Memory: Lossless Acceleration for LLM Agents
- Speculative Decoding Has Disparate Impacts
- Speculative Decoding Is Per-Token, Not Path
- Speculative Encoding for Efficient Gigapixel Whole Slide Image Analysis
- SpecX: A Large-Scale Benchmark for Multi-Modal Spectroscopy and Cross-Paradigm Evaluation
- Speech Tokenizers are Vulnerable: Transferable Semantic Attack and Robust Tokenizer
- Spherical Harmonic Optimal Transport: Application to Climate Models Comparisons
- Spherical Test-Time Robust Adaptation for Vision-Language Model
- SpikeDet-Tiny: Sub-Milliwatt Spiking Neural Network Object Detection for Event Cameras on Microcontrollers
- SpikeProphecy: A Large-Scale Benchmark for Autoregressive Neural Population Forecasting
- Spike-SFT: Selective Parameter Enhancement and Fusion for Efficient Spiking Neural Networks
- SpinDoctor: Finding Innocuous Prompts for Biased Image Generation
- SpiralFovea: Sparse Biologically-Inspired Tokenization for Vision Transformers and State Space Models
- Splatting the Invisible: Geometry and Appearance Scene Completion from Sparse Views
- Split-and-scale Latent 3D Representations
- Split Conformal Prediction for Uncertainty Quantification in Brain-Computer Interfaces
- Spontaneous High-Order Generalization in Neural Theory-of-Mind Networks
- SPOOF: Simple Pixel Operations for Out-of-Distribution Fooling
- Sporadic Gradient Tracking over Directed Graphs: A Theoretical Perspective on Decentralized Federated Learning
- Spotlights for Tumours: Effective MRI Breast Cancer Subtyping Using a Long Context Video Classification Foundation with an Attention Hack
- SpreadsheetBench 2: Evaluating Agents on End-to-End Business Spreadsheet Workflows
- Spreadsheet-RL: Advancing Large Language Model Agents on Realistic Spreadsheet Tasks via Reinforcement Learning
- SPRING: Solver-guided Process Rewards for Novel Logical Reasoning Steps Generation
- Spurious Bumps? A statistical analysis of the METR time-horizon plot
- SQUARE: Structured Quantum Representation Adapters for Nonlinear Interaction Modeling in Frozen Language Models
- SRC-Bench: A Project-Centric Dataset for Security-Relevant Commit Identification in OSS Dependencies
- SREGym: A Live Benchmark for AI SRE Agents with High-Fidelity Failure Scenarios
- SR-GRPO: Reinforcement Learning with Self-Refinement from Verifiable Rewards
- SRNN: Spatiotemporal Relational Neural Network for Intuitive Physics Understanding
- SRS-VLA: Self-Reflective Simulation for Long-Horizon Vision-Language-Action Planning
- SR-tSNE: Differentiable Rank Preservation for Neighbor Embeddings
- SSDGExplainer: Structure-Semantic Dual-Guided Explainer for Graph Neural Networks
- SS-ESOAP: Self-Scaled Adaptive Preconditioning for Physics-Informed Learning
- SSLLM: Auditable Depression-Aware Reasoning from Social and Wearable Signals
- Stability in Multi-Step Reasoning via Jacobian-based Error Accumulation Analysis
- Stabilized Physics-Aware Operator Learning for Low-Dose Nonlinear Scattering Inversion
- Stabilized Proximal Point Method via Trust Region Control
- Stabilizing Asymmetric Cross-Attention in Multi-Modal Fusion
- Stabilizing Knowledge, Promoting Reasoning: Dual-Token Constraints for RLVR
- Stabilizing Low-Rank Adaptation in Decentralized Learning
- Stabilizing RL Fine-Tuning of Diffusion VLAs with Sparse MoE Action Heads: Advantage- and Noise- Aware Load Balancing and Geometric-Mean Ratio
- Stabilizing RL+Search for Imperfect-Information Extensive-Form Games
- Stable FP4 Training via Transposition-Invariant Block Quantization
- Stable-Layers: Fine-Tuning Image Layer Decomposition Models with VLM-Scored Reinforcement Learning
- Stable Matching with Predictions: Robustness and Efficiency under Pruned Preferences
- StableTrack: Stabilizing Multi-Object Tracking on Low-Frequency Detections
- Stable Vectorization of Persistent Laplacians via Spectral Descriptors
- STAC-R: Subspace-Tracked Activation Compression with Residuals
- Stage- and Risk-Adaptive Policies for Retrieval Through Yielding Occlusions in Robotic Manipulation: A Closer Look
- StagedWorkspace: Managing Artifact State for Knowledge-Work Agents
- Stage Light is Sequence$^2$: Multi-Light Control via Imitation Learning
- StageSearch: Multi-Stage Retrieval Learning for Efficient Reasoning
- StainNFT: Curriculum-Gated Multi-Reward Post-Training for Pathology-Faithful Virtual Staining
- Staircase Activations Improve Low-Bit Post-Training Quantization
- StakeBench: Evaluating Language Understanding Grounded in Market Commitment
- StaminaBench: Stress-Testing Coding Agents over 100 Interaction Turns
- STAPLE: Static Sparse Upcycling of Pretrained Checkpoints via Embedding Modules
- STARE: A Statistical Evaluation Framework for Financial Trading Agents
- STARE: Surprisal-Guided Token-Level Advantage Reweighting for Policy Entropy Stability
- STARE-VLA: Progressive Stage-Aware Reinforcement for Fine-Tuning Vision-Language-Action Models
- STAR-Math: Multi-Agent Mathematical Reasoning under Persistent Meta-Strategic Supervision
- STaRR: Spatial-Temporal Token-Dynamics-Aware Responsive Remasking for Diffusion Language Models
- STARRY: Spatial-Temporal Action-Centric World Modeling for Robotic Manipulation
- STARS: Spike Tail-Aware Relational Synthesis for ANN-to-SNN Data-Free Knowledge Distillation
- STAR: STacked AutoRegressive Scheme for Unified Multimodal Learning
- StateBridge: State-Transition Shortcuts for Context Reuse in SSM Serving
- State Criticality via Optimal Interventions under Decaying Agency
- State-Dependent Curriculum Optimization over Transfer Graphs
- State Flow Credit Assignment: Improving Credit Assignment with Generative Modeling
- Stateful Token Reduction for Long-Video Hybrid VLMs
- State Management with Taxonomies for LLM Agents
- State of Thought Enables Endogenous Reasoning
- StatePoisonBench: A Benchmark Methodology for Persistent State Contamination in Long-Horizon Agents
- Static Recovery Is Not Dynamic Stability: Dynamics-Aware Benchmarking of Protein Motif Scaffolding
- Static Softmax: Leveraging Offline Attention Statistics Accelerates Online Softmax
- Static-to-Dynamic: Animating Still Mattes via Generative Motion for Video Matting
- Statistical Distinguishability of Rank 0 and Rank 2 for Elliptic Curves via Finite Mestre Statistics under Finite Conductors
- Statistical Inference for Structurally Constrained Nonlinear Regression
- Statistical Inference in Causal Partial Identification under Smooth Densities
- Statistical Matching via Schr\"odinger Bridge beyond Conditional Independence
- STATISTICAL UNLEARNING OF DISTRIBUTIONS: A HYPOTHESIS TESTING APPROACH
- ST-CIR: A Benchmark for Multi-Granularity Composed Image Retrieval with Heterogeneous Scene Text
- STDec: Spatio-Temporal Stability Guided Decoding for dLLMs
- Steady as She Goes: Disturbance-Aware Enhancement for Steady-State Regulation in Cooperative Multi-Agent Systems
- Stealth Apart, Harm Together: Skill Cascading Attacks on Skill-Based Agent Systems
- SteerablePoster: Adaptive Denoising Trajectory Steering for Precision Typography in End-to-End Poster Generation
- Steered LLM Activations are Non-Surjective
- STEER: Enabling Steerable Recommendation Systems with User Memory and Think-Then-Recommend
- Steering Vectors as a Training Signal in LLM Post-Training
- Steer-to-Detect: Probing Hidden Representations for Detection of LLM-Generated Texts
- StegoBench: Evaluating steganography potential in language models through supervised learning
- StemBind: When MLLMs Get Lost Between Rules and Instances in Abstract Visual Reasoning
- STEMFly: Enhancing UAV Vision-Language Navigation via Sensor Grounding, Temporal Diversity and Episodic Memory
- StepAdaptive: Uncertainty-Calibrated Compute Allocation for LLM Reasoning
- StepCAD: Mesh-to-CAD Code Generation via LLM Policy and Geometry-Guided Search
- Stepping VLMs onto the Court: Benchmarking Spatial Intelligence in Net Sports
- StepRouter: From Effort Priors to Utility Posteriors
- StereoAdapter-2: Global Iterative Refinement for Underwater Stereo Depth Estimation
- Stiefel Manifold Optimization Accelerates Reinforcement Learning
- StitchEdit: Stitching Depth Priors into Image Editors via Per-Layer Gradient Probing
- ST-LoRA: Spectral Transfer Adaptation for Fourier Neural Operator Fine-tuning
- Stochastic Interpolant Bridge Estimator: Variance-Reduced Free Energy Estimation via Learned Nonequilibrium Bridges
- Stochastic Matching via Local Sparsification
- Stochastic Penalty-Barrier Methods for Constrained Machine Learning
- Stochastic-Preference Robust Advantage Optimization for Multi-Reward Text-to-Image Alignment
- Stochastic Reconfiguration as Statistical Filtering for Overparameterized Neural Quantum States
- Stochastic Reset Pathfinding: Path-Level Regret for Cascading Bandits over Graph Paths
- StomataBench: Measuring Taxonomic Generalization in Stomatal Detection
- Stop Generating, Start Integrating: Implicit Semantic Augmentation for LLM Post-Training
- Stop or Restart? Principled Inference Control for Large Reasoning Models via the Pandora's Box
- Stop Patching AI Safety
- Stop Quantum Machine Learning; do AI-for-Quantum instead
- Stop Sensing, Start Inferring: Uncertainty-Driven Sparse Multimodal Perception for Wearable Health AI
- Stop Trusting Prices in AI Markets and Rebuild Market Infrastructure with Verifiable Reasoning
- Stored but Not Queryable: Diagnosing Cross-Modal Transfer Failure in Speech and Text LLMs
- STRABLE: Benchmarking Tabular Machine Learning with Strings
- StraTA: Incentivizing Agentic Reinforcement Learning with Strategic Trajectory Abstraction
- Strategic Decision Focused Learning
- Strategic Exploitation in LLM Agent Markets: A Simulation Framework for E-Commerce Trust
- Strategic Feature Selection and Regularization
- Strategic PAC Learnability via Geometric Definability
- Stratified Joint Causal Inference for Sparse-Tick Multi-Platform Time Series
- StreamAttention: Energy-Efficient and High-Utilization Attention on Systolic Hardware
- Stream-CQSA: Avoiding Out-of-Memory in Attention Computation via Flexible Workload Scheduling
- Streaming Speech-to-Text Translation with a SpeechLLM
- Streaming Video Editing with Easy Adaptation
- StreamPhy: Streaming Inference of High-Dimensional Physical Dynamics via State Space Models
- Stress-Testing Bundled Frozen-Backbone Memory Gains under Partial Observability
- Stress-Testing Neural Network Verifiers with Provably Robust Instances
- Stress-Testing Vision Models in the Goldilocks Zone
- Stress-Testing White-Box Hallucination Routing with Benchmark-Native Catalog Swaps
- Striatal Beat Frequency Automata: Bio-Inspired Continuous Learning in Edge Devices
- STRIDE: A Robust Graph Framework for Event-driven Vision under Distortions
- STRIDE: Automated Evaluation of Text-to-Trajectory Alignment across Diverse Contexts
- STRIVE: Structured Spatiotemporal Exploration for Reinforcement Learning in Video Question Answering
- STRMs: Spatial Temporal Reasoning Models for Vision-Based Localization Rivaling GPS Precision
- Stronger, Tractable Notions of Regret for Online Submodular Maximization
- STR-Pruner: Spatio-temporal Token Refinement for Robust Spiking Transformer Pruning
- StructBridge: Structure-Grounded 3D Indoor Object Generation via 3D Latent Diffusion Bridge and Normal Refinement
- StructCoT: Structured Chain-of-Thought Reasoning for Multimodal Large Language Models
- STRUCT: Diagnosing 3D Scene Understanding in Vision–Language Models
- StructLens: A Structural Lens for Language Models via Maximum Spanning Trees
- Structural Collapse in ECG Foundation Model: Repairing Cross-Site Transfer with the Anatomical Decision Bottleneck
- Structural Control for Long-Horizon Reasoning: A Safe Model and Episode-Level Benchmark
- Structural-Coupled Morphological Convolution for Remote Sensing Pansharpening
- Structural Extrapolated Data Generation
- Structurally Separated DAG Learning with Multi-Scale Normalized Closure
- Structurally Separated Uncertainty in Supervised Latent Variable Models
- Structural Reward Learning for Long-Form Continuation Selection: Model and Full Benchmark Test
- Structural Stability Control for Long-Horizon Reasoning in Large Language Models
- Structural Uncertainty in Tensor Neural Networks: A Distribution-Free Approach
- Structure-agnostic Causal Representation Learning
- Structure-Aware Representation Entanglement: Improving Training Efficiency of Diffusion Transformers with Register Tokens
- Structured Memory for Edge Language Models: Persistent Context and Corpus Retrieval via O(1) SSM State Injection
- Structured Neural SDEs for Functional Calibration
- Structured Prosody Modeling for Long-Form Text-to-Speech via Semantic-Conditioned Cross-Scale Regularization
- Structured Recurrent Mixers for Massively Parallelized Sequence Generation
- Structured Reference Distributions for Retrieval-Augmented Radiology Report Generation
- Structured Residual Connectivity Matters for Diffusion Transformers
- Structured State-Space Regularization for Generation-Friendly Image Tokenization
- Structured Voxel Diffusion for 3D Generation
- Structure-Preserving Unpaired Image Translation to Photometrically Calibrate JunoCam with Hubble Data
- Structure-Semantic Co-optimized Latent Diffusion Model for Fast Visual Anagram Synthesis
- Structure-Semantic Guided Closed-Loop Medical Anomaly Detection via Multi-Agent Collaboration
- Structure, Subspace and System: Push the Real Limit of Extremely Low-Bit Quantization for MoE-LLMs
- StruSR: Structure-Aware Symbolic Regression with Physics-Informed Taylor Guidance
- StyleStream 2.0: Fast and Controllable Streaming Voice Style Conversion
- Subcritical Signal Propagation at Initialization in Normalization-Free Transformers
- Subdata Selection: A Unified Framework for Optimal Selection and Statistical Efficiency Assessment
- SUBGOALCALC: Certified Decomposition for Lean Theorem-Proving Agents
- Subjects Are Domains: A Position on Multi-Sensor Clinical AI Under Distribution Shift
- Subliminal Learning Is Steering Vector Distillation
- Subliminal Prosody Learning: Auxiliary Emotion Supervision Redistributes Affective Representations Across ALM Layers
- Subliminal Transfer of Unsafe Behaviors in AI Agent Distillation
- Sublinear regret is a property of Nature
- Sublinear Time Quantum Sensitivity Sampling
- Submodular Clustering beyond $1-1/e$
- Submodular Cover under Partition and Fairness Constraints
- Subprocess-Constrained Markov Decision Processes
- Substrata: Know What You Don't Know
- Sudo Thermal Camera: Heat-Adjustable Generative Video from a Single RGB Image
- SUDP: Secret-Use Delegation Protocol for Agentic Systems
- Sufficient or Necessary? Interventional KEEP/HIDE Attribution for Auditing EHR Transformers
- Suffix-Constrained Greedy Search Algorithms for Causal Language Models
- SUMI: Scalable Unified Model for 3D Point Cloud Inference
- SunTzu: A Long-Horizon Decision Making Benchmark for Large Language Models
- Super Apriel: One Checkpoint, Many Speeds
- Superior Ordering of In-Context Examples Is Learnable by Neural and Interpretable Rule-Based Models
- Super-Level-Set Regression: Conditional Quantiles via Volume Minimization
- Superpositioned Thinking: Overcoming Premature Commitment in Diffusion LLMs
- Superposition Has Memory: The Geometry of Representational Collapse is Provably Irreversible
- Supervised Distributional Reduction via Optimal Transport and Dependence Maximization
- Supervision Recovery for Time Series Anomaly Detection via Counterfactual Pairing
- Support Before Frequency in Discrete Diffusion
- Support-Safe Variational Hybrid Filtering for Contact-Mode and Sparse-Law Recovery
- SurF: A Generative Model for Multivariate Irregular Time Series Forecasting
- SURF: Steering the Scalarization Weight to Uniformly Traverse the Pareto Front
- Surgical duration prediction enabled patient admission control
- Surgical Post-Training: Proximal On-Policy Distillation for Reasoning with Knowledge Retention
- Surprises in Proper Positive-Only Learning
- Surrogate-Assisted Training of Parametrized Quantum Circuits for Hybrid Actor-Critic Reinforcement Learning
- Surrogate-Conditional Pareto Trade-offs in Active Learning on Foundation-Model Embeddings
- Surrogate-Constrained Latent Posterior Flow for Atmospheric Data Assimilation
- SurroGate: Inference Control via Generative Profiling and Trust-Regulated Feedback
- SurvCancel: A Longitudinal Dataset and Benchmark for Dynamic Order Cancellation Prediction in On-Demand Ride-Sharing Systems
- Survival VAE: Local Feature Explanations\\ via Double-Pass Risk Stability
- SVG-3D: Mining Decision Boundaries with Generative Splatting Priors for Zero-Shot 3D Classification
- SVIB: Adaptive Geometric Routing for Compositional Vision-Language Models
- SVIPE: Sequentially Verified Prior Elicitation for Guaranteed Knowledge-Driven Bayesian Inference
- SVoT: State-aware Visualization-of-Thought for Spatial Reasoning via Reinforcement Learning
- SWARM-FR: Benchmarking Virtual Cell Metrics
- Swarm Search: Efficient, High Recall Graph-based Retrieval Robust to Index Quality
- SWE Atlas: Benchmarking Coding Agents Beyond Issue Resolution
- SWE-Crafter: Scaling Executable Multilingual Software Engineering Data with Meta-Skill Agents
- SWE-Cycle: Benchmarking Code Agents across the Complete Issue Resolution Cycle
- SWEEP: Achieving Global Convergence and Enabling Pareto Front Exploration for Deep Neural MORL
- SWE-EVO: Benchmarking Coding Agents in Long-Horizon Software Evolution Scenarios
- SWE-Git-Bench: A Focused Worktree-Level Benchmark for Real Merge Conflict Resolution
- Swift Sampling: Selecting Temporal Surprises via Taylor Series
- Swimba: Switch Mamba Model Scales State Space Models
- SWIN: Seed Weight Integration for LLM Fine-Tuning
- SwitchLingua V2: Agent-Driven Code-Switching via Digital Clones
- Sycophancy in RLHF Can Destabilize Closed-Loop Agentic AI Systems
- SymBionic-VLA: Ego-Motion-Aware Vision-Language-Action Models for Bionic Prosthetic Control
- SymbolLKG: Towards Verifiable Logical Reasoning via Logical Knowledge Graph and Symbolic Solvers
- Symb-xMIL: Symbolic Explanations for Multiple Instance Learning in Digital Pathology
- Symmetric Interventions for Eliciting Model Intent
- Symmetric Linear Dynamical Systems are Learnable from Few Observations
- Symmetries in Weight Space Learning: To Retain or Remove?
- Symmetry and Geometry in Neural Representations NeurIPS Workshops 2026
- Symmetry- and Semantic-Aware Symbolic Discovery of Ordinary Differential Equations
- Symmetry Guarantees Statistic Recovery in Variational Inference
- SymOnet: Expression-driven Foundation Model for Multi-operator Learning with Data-free Fine-tuning
- SymPlex: A Structure-Aware Transformer for Symbolic PDE Solving
- SymSelect: Symmetry Alignment under Sample-Dependent Validity
- SynAE: A Framework for Measuring the Quality of Synthetic Data for Tool-Calling Agent Evaluations
- SynCo: Learning Cross-Modal Synergy by Contrasting Interaction Residuals
- SynLV: A Synthetic Benchmark for Decision-Time Incompleteness in Longitudinal Survival Prediction
- Syntax-Guided Synthesis via a Counterexample-Guided LLM
- Synthesizing Bridges: A Diffusion-Contrastive Graph Network for Multimodal Remote Sensing Classification
- Synthetic Data is (Probably) Not All You Need
- SynthForensics: Benchmarking and Evaluating People-Centric Synthetic Video Deepfakes
- SynthHair: Leveraging MetaHumans for a High-Quality 4K Hair Matting Dataset
- SynVA: A Modular Toolkit for Vessel Generation and Aneurysm Editing
- SynZIF-8: A Synthetic SEM Dataset and Benchmark for 3D Sub-Micron Crystal Perception
- T2S-Bench: Benchmarking and Prompting Comprehensive Text-to-Structure Reasoning
- T2V-AttnDisrupt: Inducing Hallucinations in LVLMs via Misrouting Visual Evidence Retrieval
- TabClustPFN: A Prior-Fitted Network for Tabular Data Clustering
- TabFM-AD:Unlocking Tabular Foundation Model for Semi-Supervised Anomaly Detection
- TabMamba: Stage-Wise Structure-Aware Modeling Mamba for 3D Brain Tumor Segmentation
- TabMI-Bench: Evaluating Mechanistic Interpretability Methods Across Tabular Foundation Model Architectures
- TabularSetSSM: A Permutation-Invariant State Space MoE with Two-Way Mixing for Tabular Data
- Tackling Decentralized Information in Dynamic Games by Aligning Latent State Representations
- Tackling Dense Contact: A Multi-View Benchmark for 3D Human Pose in Rugby
- TAC: Target-Anchored Coverage for Source–Target Curation in Cross-Center Medical Image Segmentation
- TACTIC: Task-Agnostic Exemplar-Free Continual Learning via Hyperspherical Geometry and Feature Disentanglement
- TACT: Mitigating Overthinking and Overacting in Coding Agents via Activation Steering
- TADA: A Generative Framework for Speech Modeling via Text-Acoustic Dual Alignment
- TADA! Tuning Audio Diffusion Models through Activation Steering
- TAD: Temporal-Aware Trajectory Self-Distillation for Fast and Accurate Diffusion LLM
- Tag-Along Attacks: For LLMs, by LLMs, with LLMs
- TailAdapt: Heavy-Tailed Sparse Variational Adaptation for Long-Tailed Class Incremental Learning
- Tail-Aware Flow Matching with Mixture-of-Experts for Heavy-Tailed Conditional Density Estimation
- Tail-Conditional Regret Bounds for Best-of-N Alignment under Imperfect Reward
- Tailoring Teaching to Aptitude: Direction-Adaptive Self-Distillation for LLM Reasoning
- Tail-Sensitive Objectives Reshape Neural Operator Failure Profiles Under Shift
- Taking the Road Less Scheduled with Adaptive Polyak Steps
- Talking to AI: Robustness in Large Language Models
- Talk Less, Work More: Communication-Efficient Decentralized Stochastic Approximation
- TALON: Confidence-Aware Speculative Decoding with Adaptive Token Trees
- TAMD: Training-free Token-level Adaptive Max-min Decoding for Multiple Objectives
- TamedFlow: Trajectory-Aware Distillation for Self-Speculative Diffusion Transformers
- Taming Gradient Perturbations: Probabilistic Reparameterization for Robust Physics-Informed MRI Protocol Optimization
- Taming Outlier Tokens in Diffusion Transformers
- Taming the Entropy Cliff: Variable Codebook Size Quantization for Autoregressive Visual Generation
- Taming the Reachability Gap: Anchored Preference Optimization on the Denoising Manifold
- Taming the Tails of Stochastic Gradient Descent in the Interpolation Regime
- Tandem Reinforcement Learning with Verifiable Rewards
- Tango3D: Towards Alignment for Global and Local 2D-3D Correspondence
- TANGO-D: Diffusion Cloze-Trace Verification for Stepwise Reinforcement Learning of Mathematical Reasoning
- TaPE: Certified Robustness Against Table Permutations with Tabular Positional Encoding
- Tapered Language Models
- TAPS: Task Aware Proposal Distributions for Speculative Sampling
- Target-Consistent Dual Illumination Estimation for Robust Single-Image Exposure Correction
- Targeted Review for AI-Assisted Biodiversity Surveys: Active Continuous-Score Occupancy Modeling
- Targeted Tests for LLM Reasoning: An Audit-Constrained Protocol
- Target Policy Optimization
- Target Value as Potential: Reward Shaping for Transfer in Deep Reinforcement Learning
- Task-Conditioned Reliability Learning in Heterogeneous Multi-Agent Systems
- Task-Driven Dynamic Topology Routing for Multi-Agent Collaboration
- Task–Environment Interactions in Multi-Task Learning: Decomposition and Intervention
- Task-Level Natural Language Priors as Learning Signals for Low-Resource LLM Training
- Task Monodromy: Path Independence on Semantic Quotients as a Foundation for Neural Algorithmic Generalization
- Task Vector Geometry Underlies Dual Modes of Task Inference in Transformers
- TASM: Tail-Aware Support Modeling for Long-Tailed OOD Detection
- TASTE: From LLM Reasoning to an Auditable Rubric for Founder Evaluation
- TAVCD: Per-Step Adaptive Contrastive Coefficients for Mitigating Hallucination in Vision-Language Models
- TAVIS: A Benchmark for Egocentric Active-Vision Imitation Learning
- Tcell: Mitigating Harmful Fine-tuning for Large Language Models via Gradient Alignment
- T-Con: Diagnosing Divergence Mismatch in Time Series Contrastive Learning
- TDTRec: Semantic Prefixes With Behavioral Refinement For Generative Recommendation
- TeachClaw: Capability-Aware Adaptive Teaching for Test-Time Reasoning in Large Language Models
- Teaching VLMs What to Say, Not How to Reason: Rethinking Counterfactual Reasoning in Autonomous Driving
- Teach the Smaller to Learn the Larger: \\ Small to Large meets Federated Fine-tuning of \\ Large Language Models
- Teach-to-Reason: Competition-Guided Reasoning with a Self-Improving Teacher
- TeamBench: Evaluating Agent Coordination under Enforced Role Separation
- Team of Experts: Dependence-Aware Aggregation of Probabilistic Ensembles
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- TEA-Time: Transporting Effects Across Time
- TEE-X: TEE-aware Acceleration Framework for Large Vision Models at the Edge
- TEFormer: A Topology-Enhanced Transformer for Architecture Performance Prediction
- Telescope: Spatial Information Allocation via Learnable Hyperbolic Foveation for Ultra-Long-Range Object Detection
- Telescoping Flow: Adapting Rectified Flow for Manifold Data and Time-Agnostic Inference
- TempEdit: Temporally Guided Video Editing via Edit Energy Redistribution
- TempGlitch: Evaluating Vision-Language Models for Temporal Glitch Detection in Gameplay Videos
- Temporal Abstention: Training LLMs to Respect Knowledge Cutoff Instructions
- Temporal Concentration from Rollout Errors: Implicit Preference Optimization For Text-to-Video Diffusion
- Temporal Context Invariance for Robust Synthetic-to-Real HAR
- Temporal Modeling Platform for Unified Spatial Omics
- Temporal Predictive Coding for Training-Free ANN-to-SNN Conversion
- Temporal Preference Concepts and their Functions in a Large Language Model
- Temporal Reasoning Is Not the Bottleneck: A Probabilistic Inconsistency Framework for Neuro-Symbolic QA
- Temporal Reference-Aligned Directional Aggregation for Byzantine-Robust Federated Learning
- Temporal Robustness in AI Systems: Hidden Load, Boundary Buffer, and Path Memory
- Temporal-Scale Sensitivity in Time-Series Tokenization and Scale-Robust Token Estimation by Gated Sum
- Temporal Sheaf Diffusion
- Temporal Sheaf Neural Networks with Evolving Local Frames
- Temporal Slice Learning for AI-Generated Video Detection with 400× Fewer FLOPs
- Temporal Twins
- Tempus: A Temporally Scalable Resource-Invariant GEMM Streaming Framework for Versal AI Edge
- TempusBench: An Evaluation Framework for Time-Series Forecasting
- Tensor-based Second-order Causal Discovery
- Tensor Brain: Structured Probabilistic Modeling for Neural Population Activity with Tensor Networks
- Tensorizing Engram: Sharing Latents Across N-Gram Embeddings is Beneficial in LLMs
- TePD: Temporal Privileged Distillation for Amodal Counting Under Structured Occlusion
- TeRA: Test-driven Research Agents with Composite Specifications and Self-Repair
- TermiGen: High-Fidelity Environment and Robust Trajectory Synthesis for Terminal Agents
- TerminalWorld: Benchmarking Agents on Real-World Terminal Tasks
- Termination Censorship in World-Model Training: A Coverage-Gated Pre-Training Diagnostic
- Termination Misalignment in Large Reasoning Models
- TerraMesh-Masks: Open‑Vocabulary Segmentation for Earth Observation
- TerraVis: Towards Evaluation of World-Grounded Visual Consistency in Text-to-Image Generation via MLLM Workflows
- Territory Paint Wars: Diagnosing and Mitigating Failure Modes in Competitive Multi-Agent PPO
- TESLA: Native 4D Gaussian Splatting Generation with Temporally Structured Latents
- Testable and Actionable Calibration for Full Swap Regret
- Testable Learning of General Halfspaces under Massart Noise
- Test-Time Conditioning with Representation-Aligned Visual Features
- Test-Time Context Construction: An Information Bottleneck Perspective
- Test-Time Defense Against Adversarial Attacks via Stochastic Resonance of Latent Ensembles
- Test-Time Dynamic Classifier Construction for Exemplar-Free Class-Incremental Learning
- Test-Time Dynamics Verification of World Action Models for Contact-rich Manipulation
- Test-Time Hinting for Black-Box Vision-Language Models
- Test-time Multi-agent Coordination by Decomposed Value Gradient Flow
- Test-time reward-guided alignment of language models by importance sampling on pre-logit space
- Test-time RL alignment exposes task familiarity artifacts in LLM benchmarks
- Test-time Scaling for Diffusion Language Models with Frequency-Aware Remasking
- Test-time Scaling for LLM Hallucination Detection
- Test-time Scaling of Diffusions with Flow Maps
- Test-Time Scaling with Diffusion Language Models via Reward-Guided Stitching
- Test-Time Top-K Reranking of Electromagnetic Designs from Partial Frequency Responses
- Test-Time Training Undermines Safety Guardrails
- Tethered Predictive-Inertial Proposals with Objective Verification for Diffusion-Prior Inverse Problems
- Text-Attention Guided In-Context Forecasting for Multimodal Time Series Prediction
- Text-Based AI Tools for Research Integrity Must Be Audited on Linguistic Fairness Before Deployment
- TextEtch: Mask-Aware, Background-Preserving Text Removal
- \textit{FedNoise}: Rethinking Adaptive Diffusion Noise Scheduling for Differentially Private Heterogeneous Federated Learning
- TextSeal: A Localized LLM Watermark for Provenance & Distillation Protection
- Textual Planning with Explicit Latent Transitions
- T-FIX: Text-Based Explanations with Features Interpretable to Experts
- TFMix: Lightweight Dual-Domain Pre-Training for Time Series Forecasting with Adaptive Frequency Warping and Post-Pretraining Forgetting
- TGGM: Temporal Graph Generative Model for Multi-Graph Datasets
- The Access–Similarity Lens: An Operational Copyright Framework for Generative Models
- The Adversarial Gait: Detecting Visual Adversarial Attacks against Vision-Language Models via Self-Targeted Gradient Characterization
- The Agent Behind the Cue: Abductive Biases in Conditional Reward Hacking
- The Agentic Oversight Tax: Human Supervision of AI Agents Has a Cost that Must be Accounted For
- The Agentic Web: Networked, Continually-Adapting Agent Ecosystems
- The AI Workflow Store: Agents Must Move Beyond On-the-Fly Improvisation
- The Algorithm Is Not the Behavior: Learned Priors Override Look-Ahead in a Chess-Playing Neural Network
- The Alien Space of Science: Sampling Coherent but Cognitively Unavailable Research Directions
- The Alignment Tax Concentrates in Output Projections
- The Alternation Depth Principle for Neural Operator Design
- The Approximation Ratio for the Risk of Myopic Bayesian Active Learning for Linear Regression
- The Authority Expectancy Effects in Multi‑User Conflict
- The BabyVLM Workshop: Toward Developmentally Plausible Multimodal Systems
- The balance between feature learning and collapse in generative dynamical systems
- The Benchmarking Transparency Crisis in Large Language Models: What Reviewers Ask For That State-of-the-Art Models Never Report
- The Best Probing Layer Is Not The Last: Effective Rank as a Zero-Shot Selector
- The Bias of Nonlinear Two-Time-scale Stochastic Approximation under Constant Step-Sizes
- The Bicameral Model: Bidirectional Hidden-State Coupling Between Parallel Language Models
- The Billboard Model: DNA Foundation Models Encode Motif Identity but Fail to Learn Regulatory Grammar
- The Block Catastrophe of Marginal Calibration in Certified Requirements Traceability
- The Capability Frontier: Benchmarks Miss 82% of Model Performance
- The Case for Model Science: Verify, Explore, Steer, Refine
- The Causal Description Gap: Information-Theoretic Separations Across Pearl's Hierarchy
- The Causal-Information Floor: When Safe RL Certifiers Cannot Escape Latent Confounding
- The Compliance Trap: How Structural Constraints Degrade Frontier AI Metacognition Under Adversarial Pressure
- The Compliance Trap: Why Capable Language Models Cannot Be Both Robust and Reliable at Authority-Critical Contexts
- The Conditioning Design Space of Time-Series Diffusion: Null, Operator, and Guidance
- The Context Gathering Decision Process: A POMDP Framework for Agentic Search
- The Convention Gap: Measuring Implicit Communication in Cooperative AI Evaluation
- The DAWN of World-Action Interactive Models
- The Deceptive Evaluation Gap: Why Regulated LLM Agents Need a Verifier, Not a Judge
- The Degeneracy Distillery
- The Denoising Score Matching Loss is a Scalable Local Intrinsic Dimension Indicator
- The Dual Averaging Power-Prox Method with Application to Heavy-Tail Incremental Gradient
- The Dual Mechanisms of Spatial Variable Binding in Vision–Language Models
- The Dynamics Beneath Language: A Statistical Physics-Inspired Framework for Analyzing Large Language Models
- The Effective Dimension of Private Learning: Tight Bounds for DP-SGD Under Gradient Subspace Structure
- The Epistemic Integrity Panel: Measuring What Preference Optimization Misses in Multi-turn Agentic Reliability
- The Era of Agentic Organization: Learning to Organize with Language Models
- The Evaluation Axis: Detecting and Erasing Eval-Aware Behaviour in Reasoning Models
- The Evasion Bound: Information-Theoretic Limits of Scalar Invariance Penalties
- The Expressivity Boundary of Probabilistic Circuits: A Comparison with Large Language Models
- The FACTS Leaderboard: A Comprehensive Benchmark for Large Language Model Factuality
- The Final Mile of AI Should Be Optimized for Action, Not Information
- The FinLM Suite: pushing the limits of open source financial language modeling
- The Flow-Limit of Reflect-Reflect-Relax: Existence, Stability, and Discrete-Time Behavior
- The Fourth Quadrant: When Good Generalization Requires Bad Empirical Fit
- The Frozen-Embedding Bottleneck in Multi-Hop Retrieval
- The Geometric Inductive Bias of Grokking: Bypassing Phase Transitions via Architectural Topology
- The Geometric Structure of Models Learning Sparse Data
- The Geometry of Differentiable Discrete Computation
- The Geometry of Harmfulness in Multi-Turn Attacks
- The Geometry of Phase Transitions in Generative Dynamics via Projection Caustics
- The Geometry of Token Prediction: Voronoi Tessellations and Expressibility Gaps in Language Model Representations
- The Granular Gambit: Checkmating Behavior-based CAPTCHAs
- The Hidden Code Behind Agent Extensions Is an Overlooked Supply-Chain Risk
- The Holonomy of Thought: Taming Compounding Error via Curvature Regularization
- The Human Brain Encodes Directed Graphs of Reasoning during Naturalistic Discourse Comprehension
- The Illusion of Forgetting: Rank Leakage in Knowledge Editing and Its Mitigation
- The Illusion of Global Calibration: Auditing Group-Aware Risk in Class-Incremental Learning
- The Illusion of Privacy in Differentially Private Variants of SMOTE: Reproduction and Auditing
- The Implicit Bias of Hyperbolic Representation Learning for Multiclass Data: A Busemann Risk Perspective
- The Implicit Bias of Logit Regularization
- The Information Lives in the Disagreement: Multimodal Saliency for Efficient VLMs
- The Interaction Tax: How Communication Erases Diversity in Multi-Agent Teams
- The Interplay of Data Structure and Imbalance in the Learning Dynamics of Diffusion Models
- The Intervention Value: When Governing LLM Agents Helps and When It Harms
- The Inverted Bloom: Knowledge Architecture and Conviction Prompting as Complementary Defenses Against AI Sycophancy
- The Joint Gromov Wasserstein Objective for Multiple Object Matching
- The Kernel Reality Check: Benchmarking and Distilling Efficient Attention at Scale
- The Labeling Problem in Hallucination Detection Benchmarks: An Empirical Evaluation
- The Laplacian Keyboard: Beyond the Linear Span
- The Last Visible Pixel: Probing Fine-Scale Perception in Vision-Language Models
- The Latent Space of Matter: LLM-Driven 3D Knowledge Exploration with Discipline-Adaptive Semantic Axes
- The Linear Representation Hypothesis Is a Consequence of Approximate Lie-Group Equivariance, Not a Coincidence of Training
- The Lipschitz Trap: Why Unlearning in Language Models Returns After Ten Samples
- The Mamba Induction Circuit: Localization, Two-Stage Grokking, and Geometric Brittleness
- The Maximal-Cone Solution to Nonnegative Matrix Factorization
- The Mode of Null-A: Compositional Computation of a Generalized Inverse
- The Multi-Query Paradox in Zeroth-Order Optimization
- The Neural Race Model
- The Normalization Tax: Why Near-Optimal KV Cache Quantization Still Needs Per-Group Scales
- The Optimizer Behind the Recurrence: Stability and Convergence in Test-Time Training
- The Ordinal-Cardinal Gap in LLM Evaluators
- Theoretical guarantees for Banded Approximations of Gaussian Processes
- Theoretical Guarantees for Sub-sampled Kernelized Q-Learning with Large-scale Dependent Trajectories
- Theoretical Limits of Language Model Alignment
- Theoretically Principled Balanced Deepfake Detection
- Theoretical Understanding of Dynamic Mixture of Experts in Continual Learning
- The Origin of Edge of Stability
- Theory for group-robust continual learning
- Theory Guided and Interpretable Neural Operator Design for Partial Differential Equation Learning
- Theory of Mind and Persuasion Beyond Conversation: Assessing the Capacity of LLMs to Induce Belief States via Planning and Action
- The Override Gap: A Magnitude Account of Knowledge Conflict Failure in Hypernetwork-Based Instant LLM Adaptation
- The Panel Complexity of Sortition: Is 12 Angry Men Enough?
- The PAST-QAFTROM X Architecture: Phase-Structured Quantum Fourier Features for Trainable, Barren Plateau-Free Kernel Learning
- The Personalized-Agent Paradigm Requires Coordinated Intervention to Prevent Metacognitive Lock-in
- The Point of No Return: Counterfactual Localization of Deceptive Commitment in Language-Model Reasoning
- The Post-GCN Decade Revisited: Curvature-Stratified Evaluation of Relational Learning
- The Power of a Random Sample in Online Algorithms
- The Price of Locality: Why Forward-Forward Underperforms Backpropagation?
- The Procrustean Bed of Time Series: Optimization Bias in Point-wise Loss Functions
- The Promise and Limits of Return-Conditioned Supervised Learning in Offline POMDPs
- The Pro-Worker AI Benchmark: Measuring Whether Large Language Models Augment or Replace Human Intelligence
- The Proxy Is the Metric: How Verifiable Reward Choice Changes Conclusions About GRPO
- The Quantization Benefits of Residual-Free Transformers
- The Quantization Cliff: Disproportionate Reasoning Degradation at the W4-to-W3 Boundary
- The Quantization Trap: Breaking Linear Scaling Laws in Multi-Hop Reasoning
- The Reflexivity Threshold: A Phase Transition for Multi-Agent Performative Prediction
- The Regret-Inference Tradeoff: Balancing Hypothesis Testing and Reward Seeking in Contextual Bandits
- The Representational Limit of Scalar Interactions: An Interventional Decomposition
- The Rescue Effect: Spatio-Semantic Early Exit Bypasses Quantization Collapse in CLIP
- The Ringelmann Effect in Multi-Agent LLM Systems: A Scaling Law for Effective Team Size
- Thermo-VL: Extending Vision–Language Models to Thermal Infrared Perception
- The Role of Ego-Neighbor Separation and Graph Topology in Memorization for Graph Neural Networks
- The Role of Preference Data and Unembeddings in the Convergence Rate of DPO
- The Ruler and the Judge: Benchmark-Conditional Evaluation of LLM-as-a-Judge
- The Score Kalman Filter
- The Shakespeare Effect: Latent-Context Underspecification in Fine-Tuned Language Models
- The Shape of a Program: Path Signatures for Trace-to-Program Induction
- The Silence that Speaks: Implicit Estimation via Communication Gaps
- The Silent Brush: Evaluating Artistic Style Leakage in AI Art Generation
- The Silent Hyperparameter: Quantifying the Impact of Inference Backends on LLM Reproducibility
- The Smart Buildings Control Suite: A Diverse Open Source Benchmark to Evaluate and Scale HVAC Control Policies for Sustainability
- The SMeL Test: A sanity check for media literacy in language models
- The Spectral Amplification Hypothesis: A Law of Scale, Mismatch, and Representation
- The State-Prediction Separation Hypothesis
- The Sum Leaks More Than Its Parts: Compositional Privacy Risks and Mitigations in Multi-Agent Collaboration
- The Surprising Effectiveness of Reasoning with Unsurprising Sampling
- The Surprising Effectiveness of Video Diffusion Models for Hand Motion Reconstruction
- The Tabular Polygraph: Neurosymbolic Hallucination Detection in Synthetic Data
- The Third Workshop on Agents in the Wild: Safety, Security, and Beyond
- The Third Workshop on GenAI for Health: Agentic Systems, Clinical Trust, and Future Potential
- The Third Workshop on Long-Context Foundation Models
- The Tilted Sampling Problem: Optimism, Posterior Sampling, and Logarithmic Regret
- The Timing Attack: Sequential Detection Against a Detector-Aware Adversary
- The Tone Trap: How Presentation Style Bypasses LLM Safety at Scale
- The tractability landscape of diffusion alignment: regularization, rewards, and computational primitives
- The Tripartite Transformer: Artificial Glial Networks for Homeostatic Regulation and Robust Out-of-Distribution Detection
- The two clocks and the innovation window: When and how generative models learn rules
- The Unembedding Bottleneck: A Mechanistic Analysis of Single Digit Counting in LLMs
- The Vision Bottleneck: Why Vision-Language Models Fail at Causal Reasoning
- The Vocabulary of Refusal: Alignment as a Sparse Token Combination in Large Language Models
- The Web Doesn't Sit Still: Adversarial Self-Evolving Attacks on Search Agents
- The Weights–Context Frontier: When to Train, When to Scaffold
- The World is Not Mono: Enabling Spatial Understanding in Large Audio-Language Models
- Think about how AIs think about themselves
- ThinkDrag: Semantic Drag-Based Image Editing with Visual Reasoning
- Thinking in Boxes: 3D Editing in Real Images Made Easy
- Thinking in Pictures: A Systematic Benchmark for Reasoning-driven Image Generation
- Thinking Past the Answer: Evaluating Harmful Overthinking in Large Reasoning Models
- Thinking Seeds: Leveraging Historical Diversity for Position-Aware RL in LLMs
- Thinking with Patterns: Breaking the Perceptual Bottleneck in Visual Planning via Pattern Induction
- Thinking with Spatial Code for Physical-World Video Reasoning
- Think, then Score: Decoupled Reasoning and Scoring for Video Reward Modeling
- Think When Needed: Adaptive Reasoning-Driven Multimodal Embeddings with a Dual-LoRA Architecture
- Thompson Sampling using Prior-fitted Diffusion Transformers
- Thought without systematicity: Evaluating reasoning models on rule induction tasks
- Three Laws of Spectral Distributions:Temperature, Truncation, and Merging under Rényi Effective Rank
- ThriftAttention: Selective Mixed Precision for Long-Context FP4 Attention
- ThunderGen: A Comprehensive Dataset and Multi-Topology Synthetic Estimator for Accelerated Analog Circuit Simulation
- TIDE-Bench: Task-Aware and Diagnostic Evaluation of Tool-Integrated Reasoning
- TIDE: Efficient and Lossless MoE Diffusion LLM Inference with I/O-aware Expert Offload
- TIDE: Every Layer Knows the Token Beneath the Context
- TIDES: Implicit Time-Awareness in Selective State Space Models
- TIDES: Test-time Inference Drift Exploitation via Scaling
- TIDE: Tuning-Integrated Dynamic Evolution for LLM-Based Automated Heuristic Design
- TIER: Two-Level Inter- and Intra-Layer Exact-Memory Reallocation for Linearized LLMs
- TIGER: A Topological Sampling Intervention Framework for Diffusion-based Graph Generation
- TIGER: Text-Driven Interaction Generation and Peer-Aware Residual Control for Dual Humanoids
- Tight Gap-Dependent Regret Bounds and Problem-Independent Bounds for Cost-aware Cascading Bandits
- Tight PAC-Bayes Generalisation Guarantees for Large Language Model Safety Monitoring
- Tikhonov-Stabilized Bezier Representation Forecasting for Training-free Diffusion Acceleration
- TimeClaw: A Time-Series AI Agent with Exploratory Execution Learning
- TimeCoder: LLM-Driven Synthesis and Refinement for Intervention-Aware Time Series Modeling
- Time Dependent Loss Reweighting for Flow Matching and Diffusion Models is Theoretically Justified
- TimePhaser: Non-Stationary Memory and Phase-Aligned Sparse Evidence Retrieval for Long-Time Forecasting
- TimeQuest: Scaling Inference-Time Compute in Video-LLMs via Decoupled Temporal Navigation
- Time Resolves What Observation Channels Destroy: Modeling Dynamical Systems from Lossy Observations
- TimeSage-MT: A Multi-Turn Benchmark for Evaluating Agentic Time Series Reasoning
- Time Series as Language: A Universal Tokenizer for General-Purpose Time Series Foundation Models
- Time Series Causal Discovery via Context-Conditioned and Causality-Augmented Pretraining
- Time-series Decoding under Heterogeneous Channel Montages without Structural Modifications
- TiME: Temporal Stabilization of Mamba-based Models for Episodic Reinforcement Learning
- TimeTok: Granularity-Controllable Time-Series Generation via Hierarchical Tokenization
- Tiny but Trusted: Efficient Vision-Language Reasoning for Time-Series Anomaly Detection
- TinyCube-MC: A Multiple-Choice Benchmark for LLM Reasoning on the 2x2x2 Rubik's Cube
- TIPO: Tracking Instances via Proposals and Overlaps for Transformation Tracking and Understanding
- TIR-Bench: A Benchmark for Understanding Tool-Reasoning Interactions in Mathematical Reasoning
- TIT-Score: Evaluating Long-Prompt Based Text-to-Image Alignment via Text-to-Image-to-Text Consistency
- To Advertise or Not to Advertise: Learning Abstention for Conversational Ad Insertion in LLM Assistant
- To Align or Not To Align: Check Your COMPASS Before You Train
- To Ask or Not to Ask: Learning to Require Human Feedback
- TOC-Bench: A Temporal Object Consistency Benchmark for Video Large Language Models
- TodyComm: Task-Oriented Dynamic Communication for Multi-Round LLM-based Multi-Agent System
- ToE: A Hierarchical and Explainable Claim Verification Framework with Dynamic Multi-source Evidence Retrieval and Aggregation
- Tokenization Multiplicity Leads to Arbitrary Price Variation in LLM-as-a-service
- Tokenizer Choice Shapes Generalization in State-Centric Learning for Planning
- Token-Level Credit Assignment Reduces Sample Complexity by a Factor of Horizon in the Token MDP
- Token-Level Credit Assignment via In-Context Value Learning
- TokenPruneAlign: Dynamic Token Importance Masking with Alignment-Aware Sparse Attention for Efficient Constitutional AI Fine-Tuning
- TokenSculpt: Pruning with Min-Max Spatio-Temporal Duplication for Video Grounding
- ToLD: Efficient Time Series Forecasting via Tokenized Truncated Latent Diffusion
- Too Aligned to be Real: Detecting AI-Generated Images via Cross-modal Alignment Shift
- Tool-Augmented VLM Agents for Zero-Shot 3D Visual Grounding
- Tool-Choice Scaling Robustness: Dissecting When More Tools Make Large Language Model Agents Worse
- ToolMATH: A Diagnostic Benchmark for Long-Horizon Tool Use under Systematic Tool-Catalog Constraints
- Too Long; Didn't Reason (TL;DR): Gradients Can Drive Efficient LLM Reasoning
- Tool Use enables Undetectable Steganography in Multi-Agent LLM Systems
- Tool Use Reduces Depth-Induced Collapse in OOD Reasoning
- Tool Verification for Test-Time Reinforcement Learning
- Topic Is Not Agenda: A Citation-Community Audit of Text Embeddings
- Topo-AeroVLN: Cognitive Topological Mapping for Brain-Inspired Aerial Vision-Language Navigation
- TopoGen-RD: Reaction-Diffusion Guided Diffusion Models for Topological Structure Control
- TopoGraphRAG-Bench: Evaluating Multimodal GraphRAG on Layout-Grounded Evidence Reasoning
- Topological Expressivity Theory of Neural Networks
- Topological Invariance and Breakdown in Learning Dynamics
- Topological Steering
- Topology Collapse in Single-Cell RNA-seq Denoising: Persistent Homology Bounds and Contrastive PCA
- Topology Distortion Diagnoses Representation Fragility But Cannot Optimize It Away: A Paired View Study
- Topo-R1: Detecting Topological Anomalies via Vision-Language Models
- TopoRadar: Topology-Aware Multi-View Radar Semantic Segmentation
- TopoSSM: Learning Persistent Homology for Memory in State Space Models
- TopoU-Net: A U-Net Architecture for Topological Domains
- TOPPO: Rethinking PPO for Multi-Task Reinforcement Learning with Critic Balancing
- TorchUMM: A Unified Multimodal Model Codebase for Evaluation, Analysis, and Post-training
- To Retrieve or Not to Retrieve? Most of the Benefit is Structural, Not Semantic.
- TOUR: A Trajectory-Level Unlearning Benchmark for Offline Reinforcement Learning
- To Use or not to Use Muon: How Simplicity Bias in Optimizers Matters
- Toward a Unified Statistical Theory of Unsupervised Pretraining and Supervised Neural Knowledge Graph Learning
- Toward Bandit Convex Optimization under Generalized Smoothness
- Toward Calibrated, Fair, and accurate Deepfake Detection
- Toward Discrete-Internals Transformers
- Toward Factual Space: Codebook-Guided Contrastive Modeling for Detecting Realistic Fake News
- Toward IIT-Inspired Consciousness in LLMs: A Reward-Based Learning Framework
- Toward Open Weight Models Without Risks: Separating Public and Private Capabilities in LLMs
- Toward Optimal Regret in Robust Pricing: Decoupling Corruption and Time
- Toward Optimizing Thalamic Deep Brain Stimulation for Cortical Modulation: A Surrogate Brain Approach
- Toward Privileged Foundation Models: LUPI for Accelerated and Improved Learning
- Toward Robust Multimodal Retrieval for MLLMs with Transport Alignment
- Towards Accurate Sparse Training via Hessian-Aware Reparameterization
- Towards Acoustic World Models: Benchmarking Physics-Grounded Reasoning in Audio-Language Models
- Towards All-in-One Image Restoration: A Sparse Mixture-of-Experts Framework with Task-Adaptive Routing
- Towards Anytime-Valid Statistical Watermarking
- Towards a Transferable Defense Against Multi-Turn Attacks in LLMs
- Towards Automatically Detecting and Correcting the Ambiguous Problem Descriptions in LLM Code Generation
- Towards Characterizing Scientific Image Utility and Upgradability
- Towards Compact and Robust DNNs via Compression-aware Structural Stability Optimization
- Towards Effective and Transferable Physical Camouflage against Multi-View BEV-based 3D Perception in Autonomous Driving
- Towards Effective Multi-Agent KV Cache Selection using Gradient-Based Approximation
- Towards Efficient Compression of Large Models via Low-Rank Representation Aware Fine-tuning
- Towards Efficient Continual Learning: Task Discrimination via Incremental LDA for VLMs
- Towards Fair Graph Generation Without Sensitive Attribute
- Towards Generalization of Block Attention via Automatic Segmentation and Block Distillation
- Towards High Semantic Fidelity: Hyperdimensional Symbolic Messages in Multi-Agent Communication
- Towards Joint Quantization and Token Pruning of Vision-Language Models
- Towards Literature-Grounded Agentic Co-crystal Synthesis Design
- Towards Multi-Human-Value Alignment via Value Localization in LLMs
- Towards Multimodal Lifelong Understanding: A Multi-scale Proxy Dataset
- Towards Poisoning Robustness Certification for Natural Language Generation
- Towards Principled Dataset Distillation: A Spectral Distribution Perspective
- Towards Principled Fine-Grained MoE Expert Pruning via Pseudo-Boolean Approximation
- Towards Problem-Parameter-Agnostic Federated Reinforcement Learning with Policy Gradient Updates
- Towards Provably Unbiased LLM Judges via Bias-Bounded Evaluation
- Towards Real-Time Full-Waveform LiDAR Transformers via Intensity-Guided Token Reduction and Physics-Aware Augmentation
- Towards Real-world Human Behavior Simulation: Benchmarking Large Language Models on Long-horizon, Cross-scenario, Heterogeneous Behavior Traces
- Towards Reliable Single-View Graph Contrastive Learning: Multi-Level Refinement of Neighborhood Signals
- Towards Robust and Stable Single Decision Trees under Distributional Shift
- Towards Scalable Data Diversification for Language Model Pretraining via Leverage Score Sampling
- Towards Self-Supervised, Generalizable and Decomposable 4D Driving Scene Reconstruction
- Towards Spatial Reasoning and Understanding via Modeling Modality Conflict, Bias and Alignment
- Towards Spatio-Temporal World Scene Graph Generation from Monocular Videos
- Towards Structure-Agnostic Tensor Network Layers via Isometric Structured Operators
- Towards Temporal Interest Modeling in Recommendation via Reinforcement Learning
- Towards Uncertainty-Aware Federated Granger Causal Learning
- Towards Understanding and Measuring Cognitive Atrophy in LLM Behaviour
- Towards Understanding Why Momentum Improves SGD for Non-Smooth Non-Convex Optimization
- Towards Universal Black-box Attacks on Graph Neural Networks
- Towards Universal Semantics With Large Language Models
- Towards Visual Query Segmentation in the Wild
- Toward Visually Realistic Simulation: A Benchmark for Evaluating Robot Manipulation in Simulation
- TPBench: A Turning-Point Benchmark for Dialogue Compression
- TPO: Tri-level Distributionally Robust Learning for OOD Direct Preference Optimization
- TPRL: Adaptive Visual Token Pruning in LVLMs via Language-Guided Reinforcement Learning
- TPSD: Tree-based Parallel Speculative Decoding with Subtree Grafting
- TRACE: Domain-Adaptive Video Object Detection via Trajectory Self-Supervision and Count Constraints
- Trace the Source: Behavioral Drift Discrimination via Profile Replay for Knowledge Tracing
- TrACE: Trajectory-Adaptive Error Mitigation for Variational Quantum Eigensolvers
- TraceVul: Tracing Vulnerabilities Beyond Function Boundaries
- Tracing Geometry and Entanglement in Small Quantum Recurrent Memory Models
- Tracing the Arrow of Time: Diagnosing Temporal Information Flow in Video-LLMs
- TrackCraft3R: Repurposing Video Diffusion Transformers for Dense 3D Tracking
- TrackEverything: Dense 3D Point Tracking in Long Videos via 3D Scene Representations
- TrackFish3D: Self-Supervised 3D Tracking of Schooling Fish from Multi-view Videos
- Tracking Memorization without Validation Data
- Tracking vs. Deciding: A Data-Composition Bottleneck in Searchless Sequence Chess Transformers
- Track, Lift, Complete: Rethinking 3D Point Tracking as Occlusion Completion
- Trade-off in Estimating the Number of Byzantine Clients in Federated Learning
- TRaffic-Aligned Continuous Evaluation (TRACE): Closed-Loop Evaluation Dataset Evolution via Topological Coverage Analysis of Production Traffic
- Train at the Moving Edge: Rollout-Efficient RL for Large Reasoning Models
- Training Against Misalignment: Offline Preference Training with Simple Synthetic Safety Data
- Training a Predictive Coding Network on ImageNet using Equilibrium Propagation
- Training-Based Backdoors Are Not Cryptographic
- Training Deliberative Monitors for Black-Box Scheming Detection
- Training Flow-based Generative Models with Adaptive Curriculum Sampler
- Training-Free Entangler Selection for Quantum Neural Networks via Hilbert–Schmidt Geometry
- Training-free Episode-adaptive Dense Feature Debiasing via Fourier Phase Randomization
- Training-Free Looped Transformers
- Training-Free Pose Refinement under Coarse Geospatial Priors
- Training-Induced Escape from Token Clustering in a Mean-Field Formulation of Transformers
- Training Language Agents to Learn from Experience
- Training Language Models to Explain Their Own Computations
- Training Optimal Large Diffusion Language Models
- Training Quality Determines Efficiency Boundaries in Test-Time Reasoning
- Training Reasoning Models on Saturated Problems via Failure-Prefix Conditioning
- Training spiking neurons with a hippocampus
- Training Transformers for KV-Cache Compressibility
- Training with (Swap) Regret Loss in a Single-Layer Self-Attention Model: A Case Study on the Probability Simplex
- Train Short, Inference Long: Training-free Horizon Extension for Autoregressive Video Generation
- Trajectory-Aligned Gradient Suppression for Test-Time Generalization in Recurrent Transformers
- Trajectory as the Teacher: Few-Step Discrete Flow Matching via Energy-Navigated Distillation
- Trajectory-Aware Timestep Sampling for Conditional Visibility in In-Context Diffusion Editing
- Trajectory-Consistent Diffusion Policies for Offline Reinforcement Learning
- Trajectory-Consistent Dropout for Uncertainty Decomposition in Hamiltonian Neural Networks
- Trajectory Constraints for Imaging Inverse Problems
- Trajectory Diagnoser: Rubric-Grounded Staged Diagnosis for Long-Horizon LLM Agents
- Trajectory Influence Functions for Diffusion-Based Control: Data Attribution for Safe Planning
- Trajectory-Level Bayesian Coresets for Time-Series Clustering
- TrajLift: Encoding Verbal Memory Dynamics via Heat Diffusion on Semantic Hierarchies
- TrajMamba: Trajectory-Driven State Space Forecasting for Incomplete Multivariate Time Series
- Transferable, Time-Parallel Graph Dynamics via Space-Time Factorization
- Transformer Approximations from ReLUs
- Transformer as an Euler Discretization of Score-based Variational Flow
- Transformer Inference Is Communication
- Transformers are Energy Regularizers: Unifying Attention and Dynamics as Energy Minimization
- Transformers as In-Context Samplers: From Closed-Form Diffusion to Estimation-Free Sampling
- Transformers Provably Implement In-Context Reinforcement Learning with Policy Improvement
- Transformers Provably Learn to Internalize Chain-of-Thought
- Transition-Aware Planning for E2E Autonomous Driving with Fine-Grained Temporal Alignment
- Transitioning from Pre-training to Post-training
- TransmissiveGS: Residual-Guided Disentangled Gaussian Splatting for Transmissive Scene Reconstruction and Rendering
- Transport-Constrained Thermodynamic Neural Fields
- Transporting the Past: An Optimal Transport View of Backtracking Counterfactuals
- TravelBundle: A Benchmark for Personalized Travel Bundle Recommendation with Rich Structured Data
- TravelFraudBench: A Configurable Evaluation Framework for GNN Fraud Ring Detection in Travel Networks
- TRD-LoRA: Text-Routed Dynamic LoRA for Cross-Domain Anomaly Segmentation with Vision Foundation Models
- Treat Domain-Specific Languages as Design Variables in LLM Agents
- Tree-based Adaptive Block Elimination with Exemplar Sampling for Best Arm Identification
- TreemapMix: Dirichlet-Controlled Multi-Image Augmentation for Probability and Ordinal Supervision
- Tree of Options: Temporally Extended World Modeling, Planning, and Execution with Large Language Models
- Tree-Structured Synergy of Large Language Models and Bayesian Optimization for Efficient CASH
- TR-Graph: Task Command Repair for Safe LLM-Controlled Robot Systems
- TRIAGE: Taxonomy-Routed Intervention via pre-Answer Geometry Estimation
- TrialAgentBench: Evaluating AI Agents for Clinical-Trial Analysis and Long-Horizon Drug-Development Decisions
- TrialOpsBench: A Trust-Boundary Audit of Leaderboard Admissibility for Three Trial-Operations Decisions
- Trident: Piercing Deep Reinforcement Learning Cyber Defenses via Agentic Language Models
- Tri-Diff: Beyond Retinex via Intrinsic-Guided Latent Diffusion for Low-Light Image Enhancement
- TriFlow: Structured Triplane Latent Flow Matching for Efficient Video Generation
- Triggering Generalist Reasoning via Predictive Uncertainty for Dual-System VLA
- Tri-JEPA: Self-Supervised Multimodal Pretraining via Shared, Unique, and Synergistic Objectives
- TriP: A Triangle Puzzle Approach to Robust Translation Averaging
- Triplet-Granular Semantic Navigation over Knowledge Graph for Retrieval-Augmented Generation
- Tri-Prompting: Controllable Video Generation with Scene, Subject, and Motion Prompts
- TriTD: Tri-Partite Trajectory–Distribution Distillation for Real-Time Autoregressive Video Generation
- TriViS: A Large-Scale Multi-View Benchmark for Vietnamese Sign Language Recognition
- TRON: Tracing Rays to Orchestrate a Neural Renderer for 3D Gaussian Reconstructions
- Tropical Gaussian Anticoncentration: Settling Optimal Instance-Dependent Bounds for Online Learning in Extensive-Form Games
- TR-SSQP: A Trust-Region Method for Constrained Stochastic Optimization under Heavy-Tailed Noise
- TrueMuse: A Benchmark for Data Attribution in Text-to-Music Models
- Truncate Bad, Upweight Good: BoN-Style Distillation via Rank-Based Classification
- Truncated Riemannian (1+1)-ES for Black-Box Optimization with Intrinsic Dimension Guarantees
- Trust, but Don’t Verify: Epistemic Blind Spots in LLM Source Evaluation
- TRUST-ECHO: Reliable Structural Evidence Transfer for Medical Image Segmentation
- TrustFlow: Adaptive Trust Calibration for Language Model Guided Reinforcement Learning
- TRUST-GFS: Tie-safe Reliability and Uplift Screening for Target-guided Genetic Feature Selection
- TrustMix: Framework for Calibrated Predictive Uncertainty under Distribution Shifts
- TrustMod-SM: A Multi-Axis Benchmark for Evaluating Trustworthiness of LLMs in Social Media Content Moderation
- TRustRL: Reinforcement Learning to Promote Safe C-to-Rust Migration
- TrustRoboReward: Preference-Ordered Isotonic Score Editing for Multi-Paradigm Robot Reward Models
- Trust the Batch: A Data-Adaptive Framework for Large Model Policy Optimization
- Trustworthy AI for Good (AI4GOOD) Workshop
- Trustworthy AI Must Account for Interactions
- Trustworthy Reinforcement Learning for Adaptive Pulmonary Management: Uncertainty-Aware Digital Twins with Simulated EIT-Guided Decision Support
- Truth-Shapley: Truthful Data Valuation Against the Data Overvaluation Attack in Federated Learning
- TSA: Temporal Slot Activation for Persistent Object-Centric Video Representation
- TSFMAudit: Data Contamination Auditing in Forecasting Time Series Foundation Models
- TS-Haystack: A Multi-Task Retrieval Benchmark for Long-Context Time-Series Reasoning
- TS–Neyman: Posterior Sampling for Adaptive Stratified Estimation
- TTF: Temporal Token Fusion for Efficient Video-Language Model
- TTGBench: Benchmarking Topological Evolution and Semantic Drift in Text-attributed Temporal Graphs
- TTS-RiskArena: Unsafe at Scale? Evaluating and Mitigating Risk in Test-Time Scaling
- TULIP-Bench: Multi-perspective videos and 3D poses capturing multi-task Parkinson’s disease assessments and deep brain stimulation treatment effects
- TurbCity: A High-Resolution Urban Flow Benchmark for Real-World Turbulence Forecasting
- TurboEvolve: Towards Fast and Robust LLM-Driven Program Evolution
- TurboVGGT: Fast Visual Geometry Reconstruction with Adaptive Alternating Attention
- Turning Poison into Cure: Beneficial Backdoor Defense via Feature Decoupling Training
- TVA-DiT: Unified Tri-Modal Diffusion for Audio–Video Generation with Auxiliary Text Modeling
- TwinlandAI: An Uncertainty-Aware Agentic Digital Twin for Sustainable Land Management
- TwinShield: Secure AI Inference on Untrusted Accelerators via Cryptographic Co-Design
- TwIST: Rigging the Lottery in Transformers
- Two Calls, Two Moments, and the Vote-Accuracy Curve of Repeated LLM Inference
- Two-Clustering Regime of Token Dynamics in Causal Attention
- Two Curses Across Pearl's Causal Hierarchy
- Two Layers of Attention Stability: A Koopman-Operator Analysis of Linear and Softmax Attention
- Two-Level Softmax Sampling Done Right: Correcting Bias from Size Imbalance and Dispersion
- Two-Sided Learning in Matching Markets with Interviews
- Two Stages of Folding: Convergent Mechanisms in AI Protein Folding Trunks
- Tyche: One Step Flow for Efficient Probabilistic Weather Forecasting
- UAET: A Universal Theory of Geometric Inductive Bias in Vision and Language Models Architectures
- UAM: A Dual-Stream Perspective on Forgetting in VLA Training
- UAV-MapVQA: Evaluating Foundation Models on Instance-to-Entity Grounding Question Answering across UAV Imagery and Vector Maps
- U-Bench: A Comprehensive Understanding of U-Net through 100-Variant Benchmarking
- UBP2: Uncertainty-Balanced Preference Planning for Efficient Preference-based Reinforcement Learning
- UFO: A Unifying-domain Free Operator Framework for Generalized Operator Learning
- UI-UG 2: Advancing Mobile UI Understanding and Generation with Fine-Grained Supervision and Reinforcement Learning
- UltraDiff:Transferring High-Fidelity Priors to Compressed Latent Spaces for High-resolution Image Generation
- Ultra Fast PDE Solving via Physics Guided Few-step Diffusion
- UltraFlash: Accelerating Megapixel Visual Synthesis
- Ultrametric traversal: competitive traversal with a random walker
- UltraVR: A Diagnostic Ultra-Resolution Image–VQA Benchmark for Evidence-Grounded Reasoning
- UMARM: Unified Multi-Agent Reward Matching
- UMEDA: Unified Multi-modal Efficient Data Fusion for Privacy-Preserving Graph Federated Learning via Spectral-Gated Attention and Diffusion-Based Operator Alignment
- U-MVP: Encode Locally, Decode Globally for Feed-Forward 3D Gaussian Splatting
- Unbiased Approximate Vector-Jacobian Products for Efficient Backpropagation
- Unbounded Online Class-Incremental Learning via Stable Evolving Neural Collapse
- Unbounded Streaming Text-To-Speech with Prefixed Sliding Window Attention
- UNBOX: UNveiling black-BOX visual models with natural-language
- Uncertain Rewards bring More Possibilities: Reward Noise-driven Exploration for Reinforcement Learning
- Uncertainty-Aware Evidential Fusion for Robust Multimodal Emotion Recognition
- Uncertainty-Aware Fine-Tuning with Variational Last Layers
- Uncertainty-Aware Low-Light Enhancement with Bayesian Deep Networks
- Uncertainty Estimation for Pretrained Medical Image Registration Models via Transformation Equivariance
- Uncertainty Gates the Gradient: Adaptive Updates for Generalized Few-Shot 3D Segmentation
- Uncertainty-Guided Test-Time Adaptation for Robust Cross-Domain Brain Tumor Classification
- Uncertainty Quantification for Large Language Diffusion Models
- Uncertainty Quantification for Multimodal Large Language Models with Incoherence-adjusted Semantic Volume
- Uncertainty Quantification for Open-Ended LLM Physics Reasoning via Physics Semantics
- Uncertainty Quantification in Non-Stationary Time Series via Adaptive Conformal Regression and Reinforcement Learning
- Uncovering and Shaping the Latent Representation of 3D Scene Topology in Vision-Language Models
- Uncovering Heterogeneous Treatment Effects via Latent Diffusion Probabilistic Models
- Uncovering Semantic Hierarchies in Text-Attributed Graphs via Variational EM-based LLM–GNN Synergy
- Understanding Agent Scaling in LLM-Based Multi-Agent Systems via Diversity
- Understanding and Defending VLM Jailbreaks via Jailbreak-Related Representation Shift
- Understanding and Exploiting Weight Update Sparsity for Communication-Efficient Distributed RL
- Understanding and Harnessing Attention Sinks in Diffusion MLLMs
- Understanding and Inducing Explicit Refusal in Prompt-Injection Defense
- Understanding Coordination in Embodied Multi-Agent Systems: The Role of State Observability and Action Prediction
- Understanding DNN Overfitting through Interactions
- Understanding Driver Intention: Intention-Driven Hierarchical Planning for Robust End-to-End Autonomous Driving
- Understanding Graph Self-Supervised Pre-training under Distribution Shifts: A Scaling Law Perspective
- Understanding In-Context Learning for Nonlinear Regression with Transformers: Attention as Featurizer
- Understanding Latent Diffusability via Fisher Geometry
- Understanding Momentum-Based Accelerated Learning Through a Filter Design Perspective
- Understanding parallel samplers in masked diffusion via random walks on graphs
- Understanding Polyak's Momentum in Deep Learning May Require Rethinking Non-Convex Optimization
- Understanding SGD with EMA: A Case Study in Linear Regression
- Understanding the Connections between Policy Gradient and Trajectory Optimization
- Understanding the Curse of Unrolling
- Understanding the Effects of Hyper-Connections on Self-Attention Dynamics: A Bifurcation Analysis
- Understanding the Staged Dynamics of Transformers in Learning Latent Structure
- Understanding the Surprising Generalization Properties of Tabular Foundation Models
- UnDER: Uncertainty-Informed Diffusion Priors for Extrapolative 3D Scene Reconstruction
- Undoing the Fake: Verifiable Face Manipulation Reasoning via Model-Agnostic Inverse Editing
- UniCAD: Unified CAD Generation with Continuous Conditional Guidance
- UniClawBench: A Universal Benchmark for Proactive Agents on Real-World Tasks
- UniCoMark: Unbiased and Robust Multi-bit LLM Watermarking via Co-design of Error-resilient Recovery and Sampling
- UniCon3R: Unified Contact-aware 4D Human-Scene Reconstruction from Monocular Video
- UniCR: A Unified Generative Framework for Cloud Removal from Heterogeneous Remote Sensing Observations
- UniER: A Unified Benchmark for Item-level and Path-level Exercise Recommendation
- Unification and Optimization of Robust Supervised Learning
- Unified AI Alignment Is Impossible
- Unified Flow Matching for Long Horizon Marked Event Sequences
- Unified Forensic Preference Learning for Generalizable Synthetic Image Detection
- Unified Generative-Predictive Modeling for 4D Scene Understanding
- Unified Multiple Granularity Federated Graph Clustering Network
- Unified Regime Control: Identifiable Causal Mediators with Minimax-Tight Regret for Non-Stationary Time Series Forecasting
- UniFlowDock: Flexible Docking with Complete Equivariant Velocity Fields
- Uniform Coresets for Scalable Strategic Classification
- Unify-Agent: A Unified Multimodal Agent for World-Grounded Image Synthesis
- Unifying Clusters from Multi-Modal Graphs: A Spectral Algorithm with Theoretical Guarantee (Extended)
- Unifying Goal-Conditioned RL and Unsupervised Skill Learning via Control-Maximization
- UniHeart: A Unified Framework for 4D Cardiac Reconstruction Across Diverse Modalities
- UniMoE-World: A Unified Mixture-of-Experts Architecture for Scalable Multi-Control Video Generation World Modeling
- UniPCB: A Unified Vision-Language Benchmark for Open-Ended PCB Quality Inspection
- UniPrefill: Universal Long-Context Prefill Acceleration via Block-wise Dynamic Sparsification
- UniPRO: Easy-to-Hard Capability Generalization via Unified Policy and Reward Optimization
- UniPro: Unified Multi-Mode Medical Image Segmentation from 2D Images to 3D Volumes via Propagation
- UniQ: Modality-Aware Post-Training Quantization for Unified Multimodal Models
- UniRank: Unified List-wise Reranking via Confidence-Ordered Denoising
- UniReflect: Can Unified Multimodal Models Reflect Their Generation?
- UniSHARP: Universal Sharp Monocular View Synthesis
- UniTab: Enhancing Tabular Data Understanding with Structured Reasoning and Robust Code Generation
- United We Learn, Divided We Adapt: Conductor‑Guided Three‑Block Neural Networks for Federated Learning
- Universal and Efficient Computation with 2D Attention
- Universal Cross-Prompt Adversarial Attacks on Promptable Concept Segmentation
- Universal Time Series Generation with Neural Controlled Differential Equations
- Universal Transformers for Circuit Computations: Perfect Length Generalization in Tiny Transformers
- UniverSat: Resolution- and Modality-Agnostic Transformers for Earth Observation
- UniVLR: Unifying Text and Vision in Visual Latent Reasoning for Multimodal LLMs
- UniVL: Unified Vision-Language Embedding for Spatially Grounded Contextual Image Generation
- Unlearning Diffusion Policies via Relative Fisher Forgetting
- UnlearningSoup: Is Repeated Tuning Necessary for Large Language Model Unlearning?
- Unleashing the Power of Intrinsic-Entropy-Driven Exploration for Off-Policy Generative RL
- Unlocking Dense Metric Depth Estimation in VLMs
- Unlocking Feature Learning in Gated Delta Networks at Scale
- Unlocking Model Potentials Through Adaptive Multi-Agent Scaffolding for Efficient Issue Resolution
- Unlocking Spatial Grounding in Large Audio-Visual Retrieval models
- Unlocking the Duality between Flow and Field Matching
- Unlocking the Working Memory of Large Language Models for Latent Reasoning
- Unlocking Volition: Proactive Intention Decoding via Interpretable Graph Learning of Multi-Region ECoG
- Unordered but Not Structureless: Rank-Induced Learning in Identity-Indexed Measurement Sets
- UnpredictaBench: A Benchmark for Evaluating Distributional Randomness in LLMs
- Unrolled gradients in disguise: bridging interpolation-based and Jacobian regularization for stable neural dynamics
- Unsupervised Concept Discovery with Dirichlet Concept Diffusion Models
- Unsupervised Continual Learning with Growing Self-Organizing Maps and Synthetic Replay
- Unsupervised Decomposition with Recombination-Consistent Diffusion Models
- Unsupervised Domain Shift Detection via Discrepancy Between Pseudo-Labels and Enhanced Pseudo-Labels
- Unsupervised Process Reward Models
- Unveiling Fine-Grained Visual Traces: Evaluating MultiModal Interleaved Reasoning Chains in Multimodal STEM Tasks
- Unveiling Lookahead’s Potential in Sharpness-Aware Minimization: Convergence and Generalization Revisited
- Unveiling the Depth-Performance Dilemma in Split-Federated Fine-tuning of LLMs
- Unveiling the Value of Motion for Cinematic Camera Trajectories
- Up and Down the Abstraction Ladder: Code-Based Skills for Language Agents
- UPCDD: Unsupervised Point Cloud Dataset Distillation
- Updated, Not Just Remembered: Behavioral Supersession in Long-Term LLM Agents
- Urbex: Agentic Spatial Grounding for City-Scale 3D Scenes
- User-based Gradient Aggregation for Label Protection in Vertical Federated Recommendation
- User-Oriented Multi-Turn Dialogue Generation with Tool Use at scale
- User Satisfaction Optimization for Online LLM Services via Expectation-Confirmation Theory
- Using mechanistic insight to interpret CovFit
- Using Spurious Correlation of Hedged Feature for Controllable Regularisation
- Utility-Constrained Policy Optimization
- UVNG
- V1-Inspired Dynamic Vision System: A Bio-Plausible Video Embedding Framework with Decoupled Shape–Color Pathways and Long-Range Spatiotemporal Perception
- VAANI: Capturing the language landscape for an inclusive digital India
- VACE: Learning Geometrically Structured Representations for Time Series Anomaly Detection
- VAGS: Velocity Adaptive Guidance Scale for Image Editing and Generation
- Valid and Expressive Copulas for Irregular Multivariate Time Series
- Validated Hypotheses as a Lens for Human-Likeness Evaluation in AI Agents
- Validating Paired-Rollout Audits for Reward-Interface Dependence with Exact-TV Anchors
- Validity-Calibrated Reasoning Distillation
- Validity Threats for Foundation Model Research
- Value Entanglement: Conflation Between Different Kinds of Good In (Some) Large Language Models
- Value-Guided Variational Latent Dynamics for Offline Skill Stitching
- Value-Rectified Distillation for Flow-based Offline Reinforcement Learning
- Values Are Not Single Labels: Distributional Value Profiling Across Groups and Contexts
- ValuSpec: Plug-and-Play Candidate Valuation before Target Verification for Tree-Based Speculative Decoding
- Variance-Averse n-Step Offline Reinforcement Learning for Sparse Long-Horizon Environments
- Variance-Aware Lexicographic Generalized Linear Bandits: Unknown Variance and Tighter Bounds
- Variance-Guided Logit Adjustment for Recommender Systems
- Variational Information Bottlenecks for Geometry-Grounded Protein Inverse Folding
- VarioDrive: Adaptive and Diverse Reasoning for VLA Planning in End-to-End Autonomous Driving
- VAR-Q: Tuning-free KV Cache Quantization for Visual Autoregressive Image and Video Generation
- VascuHeal-Twin: A Sports Medicine Digital Twin with Hybrid Physics-Data Learning for Post-Intervention Upper-Extremity Tendon Healing in Obese Patients on a Public Retrospective Radiograph Benchmark
- VASR: Variance-Aware Systematic Resampling for Diffusion Models
- VCPSeg: Visibility-Conditioned Prompting for Missing-Modality Medical Image Segmentation
- VDJ-Flows: Simulating B-Cell VDJ Recombination with Biologically-Prioritized Flow Matching
- VecDBLens: A Modular Framework for Diagnosing Vector Databases Retrieval Pipelines
- VectorLogic: Editable Scientific Illustration Generation via Layout-Aware Multimodal Finetuning
- VEDJE: Video-Efficient Discriminative Joint Encoder for Scalable Video-Text Retrieval
- VEFX-Bench: A Holistic Benchmark for Generic Video Editing and Visual Effects
- VEHBench: A Stage-Local Diagnostic Benchmark for LLM-Assisted Vibration Energy Harvester Design
- Velocityformer: Broken-Symmetry-Matched Equivariant Graph Transformers for Cosmological Velocity Reconstruction
- VeriContest: A Competitive-Programming Benchmark for Verifiable Code Generation
- Verifiable Process Rewards for Agentic Reasoning
- Verified Labels Do Not Identify Off-Support Risk
- Verifier-Derived Dense Credit for Sequence Models
- Verifier-Gated Continual Learning Memory for LLM Agents in Formal and Executable Domains
- Verifier-Native Control and Trajectory Evaluation for Operations Research Repair Agents
- Veri-Sure: Multi-Agent RTL Code Generation with Temporal Tracing, Slicing and Formal Verification
- VeriVul: A Verification-Guided Framework for Generating Realistic Vulnerability Benchmarks
- Vermeer: Autoregressive generative modeling of microscopy predicts protein localization
- VERSE: Variable Ratio Stepsize Extrapolation for Fast Sampling of Diffusion Models
- Vertex-Softmax: Tight Transformer Verification via Exact Softmax Optimization
- VGGT-SLAM$^\sharp$: $\underline{S}$patially $\underline{H}$ierarchical $\underline{A}$ttention with $\underline{R}$evisiT $\underline{P}$riors
- V-HFLoRA: Target-Basis Aggregation for Heterogeneous Federated LoRA
- Vibe AIGC: A New Paradigm for Content Generation via Agentic Orchestration
- Vibe Research Drives Scientific Inquiry Shallower Unless Human Judgment Is Scaffolded
- VibeSearchBench: Benchmarking Long-horizon Proactive Search in the Wild
- Vibe Structure as Harness Engineering: The Vibe Harness Control Model and Full Benchmark Test
- ViBE: Visual-to-M/EEG Brain Encoding via Spatio-Temporal VAE and Distribution-Aligned Projection
- ViCoR: Estimating Visual Necessity via Counterfactual Residuals for Multimodal Medical Data Selection
- ViCoR: Vision-Centric Collaborative Molecular Recognition via Iterative Verification and Revision
- VICO: Visual Environments Co-Evolving for Vision-Language Model Reasoning
- Video Anomaly Detection Research is Ignoring Important Problems
- VideoInspector: Agentic Multi-Video Reasoning with Compositional Segment Training
- VideoMD: Generating Protein Dynamics with Continuous Molecular Videos
- VideoMDM: Towards 3D Human Motion Generation From 2D Supervision
- VideoStar: Adaptive Reward Shaping with Temporal Chain Coherence for Video-Grounded Reasoning
- ViDiC: Video Difference Captioning
- VidUEU-Agent: A Data-Curation Agent for Multimodal Understanding, Editing, and Unified Tasks
- VIGIL: A Reflective Runtime for Self Healing Agents
- ViJudge: Benchmarking LLM-as-a-Judge for Vietnamese through Human Preference and Bias Analysis
- ViLo-TM: Visual-Logic Mapping from Deep CNNs to Tsetlin Machines: A Closer Look
- VIMPO: Value-Implicit Policy Optimization for LLMs
- VIPBench: A Human-Aligned Benchmark for Voice Identity Perception in the Age of Voice Cloning
- VIP-COP: Context Optimization for Tabular Foundation Models
- ViroGym: Realistic Large-Scale Benchmarks for Evaluating Viral Proteins
- Virtual Task Prompting for Multi-Task Scene Understanding
- Vision2Code: A Multi-Domain Benchmark for Evaluating Image-to-Code Generation
- VisionCreator-P1: Closing the Loop for Long-Horizon Physical Visual Generation
- Vision-driven Preference Synthesis for Mitigating Hallucinations in VLMs
- Vision-Language-Action Models Need Neuroscience’s Evaluation Toolkit, Not Just Its Architectures
- Vision-Language Binding in In-Context Image Generation
- Vision-Language Grounding as Bidirectional Concept Correspondence
- Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation
- VisionPress: Intent-Aware Hierarchical Vision Compression for Efficient Multimodal LLMs
- VisionQ: VLM-as-a-Judge Taxonomy, Dataset and Benchmark for Qualitative Analysis in Computer Vision
- Vision to Geometry: 3D Spatial Memory for Sequential Embodied MLLM Reasoning and Exploration
- ViSQ: Vision-to-SuperQuadrics via Global-Local Transformers
- VISTA: Validation-Informed Trajectory Adaptation via Self-Distillation
- Visual Access Boundaries in Vision-Language Model Reasoning
- Visual Anchoring for Scenario-Guided Forecasting
- VisualBridge: Image Generation as an Intermediate for Semantic Time Series Generation
- Visual Chain-of-Thought with Verifiable Region Grounding for Multimodal Mathematical Reasoning
- Visual Distraction Undermines Moral Reasoning in Vision-Language Models
- Visual Echoes: Context-aware Perception for Multimodal Agent Memory
- Visual Grounding First, Multimodal In-context Learning Follows
- Visual Memory Injection Attacks for Multi-Turn Conversations
- VisualNeedle: Benchmarking Active Visual Search in Information-Dense Scenes
- Visual-Redundancy-Controlled Parallel Decoding for Diffusion-Based Multimodal Large Language Models
- Visual Signal Integrity for Detecting Object Hallucinations in LVLMs
- VitaBench 2.0: Evaluating Personalized and Proactive Agents in Long-Term User Interactions
- Vitruvius: Attention Is Not All You Need for Retrieval
- VivaAvatar: Drift-Free Infinite Talking Avatars with Dynamic Backgrounds
- Viverra: Text-to-Code with Guarantees
- VLA-CoRe: Empowering Vision Language Action Model with Cognitive Reasoning
- VLA-GSE: Boosting Parameter-Efficient Fine-Tuning in VLA with Generalized and Specialized Experts
- VLAN: Vision-Language Accessible Navigation
- VL-DocIR: A Benchmark for Vision-Based Long Document Retrieval
- VLM-CFG: Steering Vision-Language Inference through Multi-Modal Guidance using CFG
- VLM-Dreamer: Contrastive Semantic Regularization of World Model Latents for Robust Generalization
- VLSplat: Vision-Language Guided Object-Centric 3D Gaussian Splatting via Scene Graph
- Voice "Cloning" Is Actually Style Transfer
- Voices Behind the Veil: A Large-Scale Synthetic Dataset for Indian Maternal Health Dialogue
- VOID: Backdoor Injection through Knowledge Vacuity in Federated Unlearning
- Volterra Flow Matching for Non-Markovian Cross-Modal Alignment
- VPA-Guard: Defending and Benchmarking Image-to-Video Generation Against Visual Prompt Attacks
- VPRP: Visual Prompt Restoration Pipeline for High-Quality Nighttime Flare Removal
- VP-VLA: Visual Prompting as an Interface for Vision-Language-Action Models: A Closer Look
- Vulnerability Analysis for Safe Multi-Agent Reinforcement Learning
- VVTRec: Radio Interferometric Reconstruction through Visual and Textual Modality Enrichment
- V-Warper: Training-Free Appearance Enhanced Video Diffusion Personalization via Value Warping
- W4VU: Understanding $\mathit{Who}$ did $\mathit{What}$, $\mathit{When}$ and $\mathit{Where}$ in Videos (Extended)
- Wait for the Signal: Simple Frequency-Aware Flow-Matching
- Walking the Hypercube: Unbiased Quantum Partition Functions Without the Matrix
- Walking the Score Manifold: Continuous-time Generative Dynamics on Learned Data Manifolds
- Walking Through 3D Spaces: Spatial Routing for Referring 3D Gaussian Splatting Segmentation
- Walk Me Through Your Work: Explainability in LLMs Lies in the Auditability of Outputs
- Wander: Toward Stable World Modeling with Evolving Persistent State
- WARP Logic Neural Networks
- WASD: Wasserstein-based Knowledge Distillation for Large Language Models
- Watching Itself Watch: Self-Auditing Visual Reliance for Video Reasoning
- Watermarking as a Learned Intrinsic Property of Diffusion Models
- Watermarks Attack Watermarks: Re-Watermarking as a Generic Removal Strategy
- WaveFlow-UIE: A Single-Pipeline Wavelet-Domain Flow Model for Efficient Underwater Image Restoration
- WBENCH: A Comprehensive Multi-turn Benchmark for Interactive World Model Evaluation
- Weak-Form Bayesian Discovery of Ordinary Differential Equations with Derivative Regressors and Support-Recovery Guarantees
- Weakly Supervised Concept Learning for Interpreting and Attributing LVLM Predictions
- WeaveLA: Event Driven Cross-Subtask Latent Memory Weaving for Repetitive Robot Manipulation
- WEAVER, Better, Faster, Longer: An Effective World Model for Robotic Manipulation
- WEBACT: Test-Time Learning of Verifiable Action Interfaces for Web Agents
- WebPI-Bench: A Benchmark for Prompt Injection Robustness in Web GUI Agents
- WebPII: Benchmarking Visual PII Detection for Computer-Use Agents
- Weight Anisotropy in Mean-Field Theory: Learning on Isotropic Data
- Weight-Aware Meta Auxiliary Learning
- Weight-Level Defenses Improve LLM Agent Adversarial Robustness
- Weight-Space Graph Signal Processing for Multi-Resolution Model Merging
- Weisfeiler-Leman Is Incomplete on Simple Spectrum Graphs, so Canonicalize Them
- Welfare, Improvability, and Variance: A Principal-Agent Approach to Optimal Benchmark Item Aggregation
- WellBench: An Open-Source Benchmark for Synthetic Well Log Generation
- Well-Calibrated Yet Unfair: How Standard Conformal Prediction Fails Along the Predicted-Value Range
- We need to explain our explanations
- We Need Translational AI Research
- WFDroneBench: A Benchmark for Sensor Placement and Drone Routing for Wildfire Detection
- What Are We Actually Decoding? Source Attribution for Non-Invasive Brain-to-Language Retrieval
- What Can Labels Alone Audit and Edit in Frozen Representations?
- What does a Bayes-filtered transformer believe? A predictive Monte Carlo approach
- What Does an Observability Forecasting Foundation Model Know?
- What Drives Compositional Generalization? The Importance of Continuous Training Objectives in Visual Generative Models
- What Each Cell Means: Semantic Role Graphs for Structure-Aware Spreadsheet RAG
- What Enables Transformers to Generalize OOD? A Theoretical Perspective
- What Frozen VLAs Already Know About Success: A Probing Study of Value-Like Structure in Foundation Robot Policies
- What Hallucinations Have We Measured? An Empirical Study
- What Happens Before Decoding? Prefill Determines GUI Grounding in VLMs
- What Have EEG Foundation Models Actually Learned About the Conscious Brain?
- What Is Preference Optimization Doing, and Why?
- What Is Worth Representing? Representational Empowerment for Continual Model Construction
- What kills $v$-prediction? A Patch-wise PCA Perspective on Pixel-Space Flow Matching
- What Lies Beneath: Intent Inference via Structured Decomposition for Jailbreak Defense
- What Makes Chain-of-Thought Effective: Test-Time Step-Level Compression for In-Context Learning in LLMs
- What Makes Diffusion-based World Models Powerful Navigators? An Internal Analysis and Enhancement via Cognitive Map (Extended)
- What Makes Interaction Trajectories Effective for Training Terminal Agents?
- What Makes Two Language Models Think Alike?
- What Probing Reveals about Autonomous Driving: Better Predictions Lead to Better Planning
- What Remains in Sight? Autoregressive Video Decoding as Representation-Guided Context Rewriting
- What Reveals 3D Articulation? Learning from Motion
- What Should a Streaming Video Model Remember?
- What should post-training optimize? A test-time scaling law perspective
- What Should Time-Series Transformers Tokenize, and Across What Should Attention Be Applied?
- What Sketches Tell Us about LVLMs: Conventions, Grounding, and Localisation
- What, Where, and Boundary: Hierarchical Cognitive Decomposition for Echocardiography Video Segmentation
- What You Think Is What You Plot? A Benchmark for Decoding Visualization Intent from EEG
- When adversarial threats do not fill the ball
- When Agents Lie to Each Other: Adversarial Robustness of Hierarchical Multi-Agent Medical Consensus for Brain MRI Reporting
- When Alignment Hurts: A Controlled Benchmark for Ranking-Preserving 3D Face Metrics
- When and Why Adversarial Training Improves PINNs: A Neural Tangent Kernel Perspective
- When and Why is Optimistic Multiplicative Weights Slow? The Geometry of Energy Dissipation
- When and Why SignSGD Outperforms SGD: A Theoretical Study Based on $\ell_1$-norm Lower Bounds
- When Anecdotes Beat Accuracy: A Bayesian Framework for AI Product Evaluation
- When Are Predictions Enough? An Evaluation Protocol for Frozen Expert Composition
- When Are Quantum Kernels Trainable? A Structural Diagnostic for Circuit Architecture Search
- When Are Simple Models Statistically Indistinguishable From the Best Possible Models?
- When Autoregressive Consistency Hurts Safety Alignment
- When Brain Networks Travel: Learning Beyond Site
- When can we trust untrusted monitoring? An AI control safety case across collusion strategies
- When Concatenation Fails: A Dimensionality--Fusion Tradeoff for Foundation Model-Driven Clinical Prediction
- When Congestion Misleads: Sensitivity Inversion in Congestion-Aware Expert Routing
- When Correct Demonstrations Hurt: Rethinking the Role of Exemplars in In-Context Learning
- When Diffusion Models Do Dictionary Learning
- When Do Agents Start to Talk? Phase Transitions in Sparse Semantic Communication
- When Do Benchmarks Predict LLM Adoption?
- When Do Counterexamples Help LLM Repair?
- When Do Discovered Circuits Generalize? Auditing Prompt-Template Robustness in Mechanistic Interpretability
- When Does Adam Help under Heavy-Tailed Noise? A Coordinatewise Theory and Controlled Experiments
- When Does Agent Memory Become Actionable? A Matched Within-Stack Study of Boundary Placement
- When Does Background Knowledge Help the Learner? A Sharp Dichotomy for Parameterized Concept Fitting
- When Does Complex Solution-Level Test-Time Search Help? A Phase View of Search Utility and Decoupled Topology Search
- When Does Derivative Training Help? A Controlled-Grid Benchmark for Derivative-Enhanced Machine Learning
- When Does Dynamic Tanh Replace LayerNorm? Role-Aware Analysis Across Vision Topologies
- When Does Guidance Transfer? Decision-Validity Diagnostics for Scientific ML Prior-Knowledge Reuse Under Representation Shift
- When Does Hierarchy Help? Benchmarking Agent Coordination in Event-Driven Industrial Scheduling
- When Does Knowing the State Help? Diagnosing Process vs. Outcome Reward Design
- When does local learning work? The role of circuit preconfiguration
- When Does RLVR Work? Convergence Theory of GRPO and the Thinking Collapse Phase Transition
- When Does Scale-Invariant Optimization Become Unstable? An Exact Schedule Law with Weight Decay
- When Does Sequential Detection Collapse to a Scalar? A Necessary and Sufficient Characterisation
- When Does Subspace Direction Matter for LoRA? Regime Analysis of the Magnitude Principle in Few-Shot Adaptation
- When does the noise schedule matter? A spectral classification of diffusion training objectives
- When Does Trimming Help Conformal Prediction? A Retained-Law Diagnostic under Calibration Contamination
- When Do LLMs Generate Realistic Social Networks? A Multi-Dimensional Study of Culture, Language, Scale, and Method
- When Eyes Betray AI: Social Gaze Consistency as a Semantic Cue for AI-Generated Image Detection
- When Graph Anomalies Learn to Hide: Test-Time Cloaking in Graph-Level Anomaly Detection
- When Graph Structure Provably Helps Classification: Non-Asymptotic Recovery Guarantees
- When Implausible Tokens Get Rewarded: Tail-Aware Credit Calibration for LLM Reinforcement Learning
- When Instructions Retrieve Trajectories: Diagnosing Generalization Failures in VLA Models
- When is a Representation Sufficient? Probe-Based $\mathcal{V}$-Information for Neural Architecture Search
- When Judgment Becomes Noise: Detecting and Attributing Rubric Failures in LLM Judge Pipelines
- When Less is Enough: Efficient Inference via Collaborative Reasoning
- When Looking Good Isn't Good Enough: Do Rendering Metrics Predict Task Performance?
- When Matched-Rate Dropout Is the Wrong Control for Selective Sensing
- When Mechanism Transfer Becomes Readout Transfer
- When Medical VLMs Stop Understanding: MedTEC-Bench for Probing Semantic Specificity
- When Membership Signals Lie: Per-Sample Membership Inference Hidden in Your Checkpoints
- When Monitors Fail, the Model Still Knows: Probing Obfuscated Reasoning in LLMs.
- When One Sample Is Not Enough: Low-Sample Evaluation Misrepresents LLM Reliability
- When Policies Cannot Be Retrained: A Unified Closed-Form View of Post-Training Steering in Offline Reinforcement Learning
- When Predictions Take Time: Cost-Aware Predictor Selection for Learning-Augmented Scheduling
- When pre-training hurts LoRA fine-tuning: a dynamical analysis via single-index models
- When Priors Disagree: Robust Forgery Localization via Global Modeling of Cross-Prior Discrepancies
- When Properties Are Not Separated Do LLMs Fail at Judging
- When Random Number Choices Break Confidence: Restoring Reliability in LLM Judgment Guarantee
- When Refresh Metrics Disagree: Auditing Deployed Ranking-Slice Maintenance
- When Regime Change Changes the Objective: Ranking Reversal in Non-Stationary Learning
- When Representations Collapse in Logical Reasoning: Diagnosing and Mitigating Cascade Failures in LLMs
- When Right Meets Wrong: Bilateral Context Conditioning with Reward-Confidence Correction for GRPO
- When Salience Wins: Evidence Arbitration Failures in Document-Grounded Question Answering
- When Sanitization Becomes the Trigger: Defense-Triggered Backdoor Attacks
- When Scores Conflict with Preferences: Calibrated Drift Control for Heterogeneous DPO
- When Selective Reuse Breaks Structured Correctness: Lightweight Repair for Fast LLM Inference
- When Self-Belief Misleads: Active Label Acquisition for Reinforcement Learning with Verifiable Rewards
- When Should an Audio LLM Call Its Tools? Counterfactual Reward Shaping for Selective Routing
- When Should a Routing Decision Be Re-examined? A Diagnostic Study of Model Switching During Reasoning
- When Should GNNs Look Further? A Diagnostic Framework for Non-Local Propagation
- When Should Graph Attention Be Sparse? Learning a Per-Edge Tsallis Index
- When Should OMOP Target Trial Emulation Stay Silent? Held-Out Denominator Audit of When to Report (Extended)
- When Solve-Only Misranks Failure-Conditioned Lemma Admission: A Reuse-and-Clutter Audit in Recurrence-Rich Lean 4
- When Stored Evidence Stops Being Usable: Scale-Conditioned Evaluation of Agent Memory
- When Substantial $\kappa$ Hides Systematic Disagreement: Auditing LLM-as-a-Judge Evaluations Beyond Aggregate Agreement
- When Symbol Names Should Not Matter: A Logistic Theory of Fresh-Symbol Classification
- When Tasks and Modalities Both Grow: Meta-Learned Plasticity for Multimodal Continual Learning
- When the Future Remembers: Past Reconstruction in Autoregressive Long Video Generation
- When the Mythical Man-Month Meets Multi-Agent Systems: Structured Interaction as the Key to Scalable Coordination
- When to Adopt Model Updates
- When to End, Which to Invoke Next: Transition-Aware Tokenized Procedural Memory for LLMs
- When Tokenization Premium Becomes Context Loss: An Indic-Language Audit of Modern LLM Tokenizers
- When to Restart Your LLM Session? Optimal Restarting for LLM APIs via Inventory Theory
- When to Stop Reusing: Dynamic Gradient Gating for Sample-Efficient RLVR
- When to Trust the AI Pre-decoder: Helpfulness Sparsity and Calibrated Dispatch for Surface-Code Quantum Error Correction
- When to Vote, When to Rewrite: Disagreement-Guided Strategy Routing for Test-Time Scaling
- When Trivial Baselines Win: A Cross-Benchmark Audit of Failure Attribution in Multi-Agent LLM Systems
- When Union Access Is Not a Low-Call Upper-Bound Baseline: A Controlled GQA Audit
- When Variance Control Becomes Drift: A Mechanism Analysis of GRPO-Style Policy Optimization
- Where Concepts Fire
- Where Did Sequence Go? Robust Space Reshaping via Large-Small Model Collaboration for LLM-based Sequential Recommendation
- Where Does Warm-Up Come From? Adaptive Scheduling for Norm-Constrained Optimizers
- Where Do Long Captions Fail? Position-Aware Diagnosis and Reinforcement Learning for Detailed Image Captioning
- Where Representations Converge and Where They Might Not
- Where Resource-Centric Tool Transfer Breaks:\ Audited Six-Family Boundary Study
- Where Reusable Computation Becomes Detectable: Solution-Frame Path Triage for Modular-Arithmetic Grokking
- Where Should Temporal Admissibility First Act in RAG? A Diagnostic Placement Study of the Temporal-Control Locus under Version Mixing
- Where to Look Matters: Rethinking Sub-Volume Sampling in 3D Medical Self-supervised Learning
- Where to Spend Rollouts: Hit-Utility Optimal Rollout Allocation for Group-Based RLVR
- Where You Backpropagate Matters: Token Hypothesis for Memory-Efficient Fine-Tuning
- Which and When to Admit: Gradient Admission for Data-Centric Small Language Model Finetuning
- Which Foundation Backbone Survives the Wild? A Laundering Atlas for Face-Deepfake Detection
- Which Geometry on Which Layer? A Principled Criterion for Mixed-Optimizer Training
- Which Graph Shift Operator? A Spectral Answer to an Empirical Question
- Who Grades the Graders? Validating Automated Evaluators in Adversarial AI Safety Benchmarks
- Who Made This LLM? Black-Box Family Provenance for Large Language Models
- Who Needs a Cherry When the Icing Is Robust? Distributionally Robust Fine-Tuning Is KTO in Disguise.
- Who Needs Labels? Adapting Vision Foundation Models With the Metadata You Already Have
- Who Verifies the Agents? Toward Reliable Agent Development
- Who Wrote This Paper? Autonomous Scientific Discovery for 3DGS Research
- Why Adam Works Better with $\beta_1 = \beta_2$: The Missing Gradient Scale Invariance Principle
- Why Adaptive Sampling Fails: The Projection Bottleneck in Diffusion Inference
- Why Alignment Fails: A Geometric Theory of Structural Hallucination in LLMs
- Why Can Sparse Mixture-of-Experts Generalize Well? An Information-Theoretic Perspective
- Why Clipping Matters in AdaGrad? Toward a High-Probability Theory under Generalized Smoothness
- Why Confidence-Based Early Exit Fails on Reasoning Traces: A Projection Bottleneck Analysis
- Why Copy Others? Insights into Social Learning from Multi-Agent Reinforcement Learning
- Why Deterministic PRM Guidance Underperforms in Discrete Diffusion Reasoning
- Why Do Bad Demonstrations Break In-Context Learning? Label-Channel Cascade Lock-In
- Why Does Relational Drift Occur? The Attention Sink Phenomenon and Mechanisms of Perspective Information Loss in Transformers
- Why Efficient Scaling of Any-to-Any MLLMs Requires NPUs
- Why Engagement Optimization is a Risky Slippery Slope
- Why Fair Dataset Distillation Fails on Complex Distributions: A Geometric Analysis and a Wasserstein Remedy
- Why Geometric Continuity Emerges in Deep Neural Networks: Residual Connections and Rotational Symmetry Breaking
- Why In-Sample Diagnostics Cannot Certify Cross-Task Foundation-Model Transfer
- Why LLMs Should Be Reasonably Morally Inconsistent
- Why PCA Feature Alignment is Important for Building Graph Foundation Models
- Why Quantum Natural Gradients Fail?
- Why Routers Freeze: Infinite Width Learning Dynamics for Mixture of Experts
- Why Self-Consistency Fails: Impossibility Results and Predictive Theory for Multi-View LLM Uncertainty
- Why Some Models Resist Unlearning: A Linear Stability Perspective
- Why Static Probes Fail: Context Determines Deception Feature Composition
- Why Transformer-Based Language Models Need Explicit Mechanisms of Cognitive Control
- WiCER: Wiki-memory Compile, Evaluate, Refine Iterative Knowledge Compilation for LLM Wiki Systems
- Width-Independent Compressibility of Deep Neural Networks
- WiFi-TTA-Bench: Test-Time Adaptation Under Physics-Structured Wireless Shift
- WikiDel-10M: A Stress-Test Benchmark for Dense Retrieval Under High-Churn Revocation and Tight Handling Windows
- WildDet3D: Scaling Promptable 3D Detection in the Wild
- WILDER-Net: Invertible Wavelet-Layered Audio Steganography Towards Robust High-Capacity Hiding and Authentication
- WILD: Widely Linear Conditioning for Time Series Forecasting
- Winners Take All: Persistent Activation Outliers in Pre-Norm Transformers
- Winner-Take-All bottlenecks enforce disentangled symbolic representations in multi-task learning
- WINR: Neural Implicit Representations for Weight Space
- WiREBench: Evaluating AI Agents' Capabilities in Reverse-Engineering Black-Box Applications in the Real World
- Wireless Planning Benchmark for Reward-Guided Network Deployment
- Within-Model vs Between-Prompt Variability in Large Language Models for Creative Tasks
- WMEval: Evaluating Interactive World Models Grounded on Physical Simulations
- WM-R1: Training GUI Agents to Reason with World Models via Reinforcement Learning
- Woodbury Influence: Fast and Effective Evaluation of Unseen Sample Familiarity Before Training: A Closer Look
- WOODELF-HD: Efficient Background SHAP for High-Depth Decision Trees
- WordEval: Evaluating Word-Native Operation Fidelity in Document Editing
- Workshop for Autonomous Machine Learning Research
- Workshop on Evaluation of Interactive Agents
- Workshop on Resource-Aware Agentic AI
- Workshop on the Linguistic Principles for Foundation Models
- Workshop on Towards Test-Time Continual Learning Agents
- WorldAct: Activating Monolithic 3D Worlds into Interactive-Ready Object-Centric Scenes
- WorldCraft: From Camera Navigation to Object Manipulation in Interactive Video World Models
- WorldDirector: Building Controllable World Simulators with Persistent Dynamic Memory
- World-Ego Modeling for Long-Horizon Evolution in Hybrid Embodied Tasks
- World-Grounded Camera Trajectory Generation
- World-HOI: World Model for 4D Hand-Object Interaction
- WorldKV: Training-Free KV Cache Retrieval and Compression for Efficient World Memory
- World-Model-Inspired Flicker State Modeling for Burst Flicker Removal
- World Model Self-Distillation: Training World Models to Solve General Tasks
- World Models for High-Stakes Health: Reliable Clinical Trial Simulation and Intervention-Aware Reasoning
- World Models in Physical AI
- World of Workflows: a Benchmark for Bringing World Models to Enterprise Systems
- WorldReasonBench: Human-Aligned Stress Testing of Video Generators as Future World-State Predictors
- WorldREPA:Extending Representation Alignment to Multi-source World Knowledge Learning
- WorldSR: Harnessing World Knowledge Search for Grounded Image Super-Resolution
- Write-Before-Query Memory: Separating Stored Burden from Local Service
- Wrong-Object Diagnostics for Revocation
- Wrong-Physics Backdoors in Neural PDE Operators
- WTF?! Simulation-Free Reinforcement Learning with Wasserstein-Tilted Flow Maps
- WURI: Watching Unfolding Risk in Agent Interactions
- xAdvCool: 3D Conjugate Heat Transfer Dataset and Benchmarks for Cold-Plate Surrogate Modeling
- XAI4science: Knowledge Discovery and Trust through Interpretable Foundation Models
- X-AVDD: Cross-Attentive Audio-Visual Dataset Distillation
- XBRIDGE: Entity-Grounded Latent Bridge for Heterogeneous LLM Communication
- XFactors: Disentangled Information Bottleneck via Contrastive Supervision
- X-Humanoid: Robotize Human Videos to Generate Humanoid Videos at Scale
- xMIx: High-Performance Serving-Time Platform for Mechanistic Interpretability Apps
- Xolver: Generalist Reasoning and Problem Solving through Federated Multi-Agent Dynamics and Holistic Experience Learning
- X-Palm: Paired Multispectral-to-Smartphone Dataset for Cross-Domain Palmprint Authentication
- xVGAE: A Hierarchical Variational Graph Autoencoder for Exchangeable Graphs
- XWOD: A Real-World Benchmark for Object Detection under Extreme Weather Conditions
- You Don't Have to Be the One Doing Evil! Locate a Scapegoat within Multi-Agent Systems for Executing Covert Attacks
- You Only Need Minimal RLVR Training: Extrapolating LLMs via Rank-1 Trajectories
- Your Code Agent Can Grow Alongside You with Structured Memory
- Your GFlowNet Secretly Learns an Optimal Transport Plan
- Your Hypergradient is Skewed: Antithetic Neumann Estimation for Bilevel Optimization
- Your Language Model is Its Own Critic: Reinforcement Learning with Value Estimation from Actor’s Internal States
- Your Own Words Against You: Generated-Text KV as an Inference-Time Prior in Streaming VLMs
- Your Pre-trained Diffusion Model Secretly Knows Restoration
- Your Teacher Can’t Help You Here: Combating Supervision Fidelity Decay in On-Policy Distillation
- Your Transformer Can Hold Two Thoughts at Once: Evidence of Linear Superposition in LLMs
- yugi-bench: A Long-Horizon Tool-Use Benchmark and Frontier-Model Capability Analysis
- ZEBRA: Zeroth-order Energy-Based Robust Adaptation for the Last Mile
- ZEMM: Fusing Lossless Decompression with GEMM for Flexible and Efficient LLM Serving
- Zen: A Framework for Zero-Shot, Across-Subject Neural Decoding
- ZeroCode: On-demand Error-Correcting Code Construction from Zero-Matrix via Reinforcement Learning
- Zero knowledge verification for frontier AI training is possible
- Zero-Shot Burst Restoration via Diffusion MAP Inference with Poisson-Gaussian Noise Likelihood
- Zero-Shot Coordination among LLM Agents
- Zero-Shot VLM Ranking through Spectral Analysis of Unforgettable Representations
- Zeroth-Order Optimization under Rotating Intrinsic Geometry
- Zeroth-Order Stackelberg Control in Combinatorial Congestion Games
- Zero-Violation Regret for Cooperative Markov Games with Coupled Instantaneous Hard Constraints
- ZetaEvolve: Learning to Search through History-Conditioned Potential Value
- zkFU: Verifiable Federated Unlearning via Trace-Consistent Subtraction and Zero-Knowledge Proofs
- ZOLD: A Zeroth-Order Langevin Dynamics Framework for LLM Fine-Tuning
- α-Robust SGD: Oracle-Free Adaptive Gradient Clipping under Heavy-Tailed Noise
- κ-LoRA: Condition Numbers Reveal Which LoRA Matrices Worth Updating
Successful Page Load