# arXiv Scout Coverage — 2026-10-18_23-28

- Queries run: 14
- Total pages fetched: 22
- Total entries seen: 1300
- Total unique papers kept before cross-query dedupe: 1300
- Total duplicates skipped within queries: 0
- Queries with incomplete coverage: 0

## Per-query coverage

### primary-cs.AI
- Query: `cat:cs.AI`
- Backfill window: 72h
- Page size: 100
- Max pages: 5
- Pages fetched: 1
- Entries seen: 100
- Unique papers kept: 100
- Duplicates skipped: 0
- Newest seen: 2026-10-16T17:59:30+00:00
- Oldest seen: 2026-10-16T03:40:48+00:00
- Stop reason: cutoff_reached
- Pages:
  - page 1 start=0 entries=100 newest=2026-10-16T17:59:30+00:00 oldest=2026-10-16T03:40:48+00:00 rate_limited=False status=200

### primary-cs.LG
- Query: `cat:cs.LG`
- Backfill window: 72h
- Page size: 100
- Max pages: 5
- Pages fetched: 2
- Entries seen: 200
- Unique papers kept: 200
- Duplicates skipped: 0
- Newest seen: 2026-10-16T17:58:08+00:00
- Oldest seen: 2026-10-15T17:59:59+00:00
- Stop reason: cutoff_reached
- Pages:
  - page 1 start=0 entries=100 newest=2026-10-16T17:58:08+00:00 oldest=2026-10-16T09:23:22+00:00 rate_limited=False status=200
  - page 2 start=100 entries=100 newest=2026-10-16T09:17:26+00:00 oldest=2026-10-15T17:59:59+00:00 rate_limited=False status=200

### primary-cs.CL
- Query: `cat:cs.CL`
- Backfill window: 72h
- Page size: 100
- Max pages: 5
- Pages fetched: 1
- Entries seen: 100
- Unique papers kept: 100
- Duplicates skipped: 0
- Newest seen: 2026-10-16T17:58:08+00:00
- Oldest seen: 2026-10-15T22:53:55+00:00
- Stop reason: cutoff_reached
- Pages:
  - page 1 start=0 entries=100 newest=2026-10-16T17:58:08+00:00 oldest=2026-10-15T22:53:55+00:00 rate_limited=False status=200

### topic-world-model
- Query: `(cat:cs.AI OR cat:cs.LG OR cat:cs.CL) AND ("world model" OR world-model)`
- Backfill window: 96h
- Page size: 50
- Max pages: 2
- Pages fetched: 2
- Entries seen: 100
- Unique papers kept: 100
- Duplicates skipped: 0
- Newest seen: 2026-10-16T16:42:11+00:00
- Oldest seen: 2026-10-14T09:35:54+00:00
- Stop reason: cutoff_reached
- Pages:
  - page 1 start=0 entries=50 newest=2026-10-16T16:42:11+00:00 oldest=2026-10-15T11:27:19+00:00 rate_limited=False status=200
  - page 2 start=50 entries=50 newest=2026-10-15T10:35:47+00:00 oldest=2026-10-14T09:35:54+00:00 rate_limited=False status=200

### topic-agent
- Query: `(cat:cs.AI OR cat:cs.LG OR cat:cs.CL) AND agent`
- Backfill window: 96h
- Page size: 50
- Max pages: 2
- Pages fetched: 2
- Entries seen: 100
- Unique papers kept: 100
- Duplicates skipped: 0
- Newest seen: 2026-10-16T17:57:55+00:00
- Oldest seen: 2026-10-15T17:52:33+00:00
- Stop reason: max_pages_reached
- Pages:
  - page 1 start=0 entries=50 newest=2026-10-16T17:57:55+00:00 oldest=2026-10-16T08:43:11+00:00 rate_limited=False status=200
  - page 2 start=50 entries=50 newest=2026-10-16T08:34:20+00:00 oldest=2026-10-15T17:52:33+00:00 rate_limited=False status=200

### topic-tool-use
- Query: `(cat:cs.AI OR cat:cs.LG OR cat:cs.CL) AND ("tool use" OR tool-use)`
- Backfill window: 96h
- Page size: 50
- Max pages: 2
- Pages fetched: 1
- Entries seen: 50
- Unique papers kept: 50
- Duplicates skipped: 0
- Newest seen: 2026-10-16T17:48:26+00:00
- Oldest seen: 2026-10-15T03:32:39+00:00
- Stop reason: cutoff_reached
- Pages:
  - page 1 start=0 entries=50 newest=2026-10-16T17:48:26+00:00 oldest=2026-10-15T03:32:39+00:00 rate_limited=False status=200

### topic-memory
- Query: `(cat:cs.AI OR cat:cs.LG OR cat:cs.CL) AND memory`
- Backfill window: 96h
- Page size: 50
- Max pages: 2
- Pages fetched: 2
- Entries seen: 100
- Unique papers kept: 100
- Duplicates skipped: 0
- Newest seen: 2026-10-16T17:49:43+00:00
- Oldest seen: 2026-10-14T15:40:05+00:00
- Stop reason: cutoff_reached
- Pages:
  - page 1 start=0 entries=50 newest=2026-10-16T17:49:43+00:00 oldest=2026-10-15T15:29:57+00:00 rate_limited=False status=200
  - page 2 start=50 entries=50 newest=2026-10-15T15:28:13+00:00 oldest=2026-10-14T15:40:05+00:00 rate_limited=False status=200

### topic-reasoning
- Query: `(cat:cs.AI OR cat:cs.LG OR cat:cs.CL) AND reasoning`
- Backfill window: 96h
- Page size: 50
- Max pages: 2
- Pages fetched: 2
- Entries seen: 100
- Unique papers kept: 100
- Duplicates skipped: 0
- Newest seen: 2026-10-16T17:52:58+00:00
- Oldest seen: 2026-10-15T07:16:25+00:00
- Stop reason: max_pages_reached
- Pages:
  - page 1 start=0 entries=50 newest=2026-10-16T17:52:58+00:00 oldest=2026-10-16T01:21:47+00:00 rate_limited=False status=200
  - page 2 start=50 entries=50 newest=2026-10-16T01:13:23+00:00 oldest=2026-10-15T07:16:25+00:00 rate_limited=False status=200

### topic-llm
- Query: `(cat:cs.AI OR cat:cs.LG OR cat:cs.CL) AND (LLM OR "large language model" OR "large language models")`
- Backfill window: 96h
- Page size: 50
- Max pages: 2
- Pages fetched: 2
- Entries seen: 100
- Unique papers kept: 100
- Duplicates skipped: 0
- Newest seen: 2026-10-16T17:58:08+00:00
- Oldest seen: 2026-10-16T01:24:42+00:00
- Stop reason: max_pages_reached
- Pages:
  - page 1 start=0 entries=50 newest=2026-10-16T17:58:08+00:00 oldest=2026-10-16T10:26:23+00:00 rate_limited=False status=200
  - page 2 start=50 entries=50 newest=2026-10-16T10:23:41+00:00 oldest=2026-10-16T01:24:42+00:00 rate_limited=False status=200

### topic-quantization
- Query: `(cat:cs.AI OR cat:cs.LG OR cat:cs.CL) AND quantization`
- Backfill window: 96h
- Page size: 50
- Max pages: 2
- Pages fetched: 1
- Entries seen: 50
- Unique papers kept: 50
- Duplicates skipped: 0
- Newest seen: 2026-10-16T17:22:01+00:00
- Oldest seen: 2026-10-12T07:16:40+00:00
- Stop reason: cutoff_reached
- Pages:
  - page 1 start=0 entries=50 newest=2026-10-16T17:22:01+00:00 oldest=2026-10-12T07:16:40+00:00 rate_limited=False status=200

### topic-open-source
- Query: `(cat:cs.AI OR cat:cs.LG OR cat:cs.CL) AND ("open source" OR open-source)`
- Backfill window: 96h
- Page size: 50
- Max pages: 2
- Pages fetched: 1
- Entries seen: 50
- Unique papers kept: 50
- Duplicates skipped: 0
- Newest seen: 2026-10-16T15:48:00+00:00
- Oldest seen: 2026-10-13T22:01:49+00:00
- Stop reason: cutoff_reached
- Pages:
  - page 1 start=0 entries=50 newest=2026-10-16T15:48:00+00:00 oldest=2026-10-13T22:01:49+00:00 rate_limited=False status=200

### topic-self-improvement
- Query: `(cat:cs.AI OR cat:cs.LG OR cat:cs.CL) AND ("self-improvement" OR "self improvement")`
- Backfill window: 96h
- Page size: 50
- Max pages: 2
- Pages fetched: 1
- Entries seen: 50
- Unique papers kept: 50
- Duplicates skipped: 0
- Newest seen: 2026-10-16T17:56:35+00:00
- Oldest seen: 2026-10-09T17:34:44+00:00
- Stop reason: cutoff_reached
- Pages:
  - page 1 start=0 entries=50 newest=2026-10-16T17:56:35+00:00 oldest=2026-10-09T17:34:44+00:00 rate_limited=False status=200

### topic-fine-tuning
- Query: `(cat:cs.AI OR cat:cs.LG OR cat:cs.CL) AND ("fine-tuning" OR finetuning OR finetune)`
- Backfill window: 96h
- Page size: 50
- Max pages: 2
- Pages fetched: 2
- Entries seen: 100
- Unique papers kept: 100
- Duplicates skipped: 0
- Newest seen: 2026-10-16T17:58:08+00:00
- Oldest seen: 2026-10-14T09:48:15+00:00
- Stop reason: cutoff_reached
- Pages:
  - page 1 start=0 entries=50 newest=2026-10-16T17:58:08+00:00 oldest=2026-10-15T13:52:20+00:00 rate_limited=False status=200
  - page 2 start=50 entries=50 newest=2026-10-15T12:56:57+00:00 oldest=2026-10-14T09:48:15+00:00 rate_limited=False status=200

### topic-benchmark
- Query: `(cat:cs.AI OR cat:cs.LG OR cat:cs.CL) AND (benchmark OR benchmarks)`
- Backfill window: 96h
- Page size: 50
- Max pages: 2
- Pages fetched: 2
- Entries seen: 100
- Unique papers kept: 100
- Duplicates skipped: 0
- Newest seen: 2026-10-16T17:57:55+00:00
- Oldest seen: 2026-10-15T20:47:46+00:00
- Stop reason: max_pages_reached
- Pages:
  - page 1 start=0 entries=50 newest=2026-10-16T17:57:55+00:00 oldest=2026-10-16T09:38:11+00:00 rate_limited=False status=200
  - page 2 start=50 entries=50 newest=2026-10-16T09:32:22+00:00 oldest=2026-10-15T20:47:46+00:00 rate_limited=False status=200

## Priority summary
- High-priority papers: 478
- Priority threshold: 4
- High-similarity papers: 0
- Similarity separation threshold: 0.15
- Topic hits:
  - agent: 100
  - benchmark: 100
  - fine-tuning: 100
  - llm: 100
  - memory: 100
  - reasoning: 100
  - world-model: 100
  - open-source: 50
  - quantization: 50
  - self-improvement: 50
  - tool-use: 50
- Top papers:
  - 22.0 / rank=22.00 :: From Trajectories to Differentiated Skills: Cognitive Differentiation for Self-Improving Agents [agent, benchmark, llm, reasoning, self-improvement, tool-use] sim+=0.26 sim-=0.06 sep=0.00
  - 20.0 / rank=20.00 :: Darwin-Claw: Environment-Adaptive Experience Consolidation for Recursively Self-Evolving Agents [agent, llm, memory, reasoning, world-model] sim+=0.20 sim-=0.04 sep=0.00
  - 19.0 / rank=19.00 :: MetaCurate: Evolving Curation Filters for High-Quality Agentic Coding Data [agent, fine-tuning, llm, reasoning, tool-use] sim+=0.21 sim-=0.06 sep=0.00
  - 17.0 / rank=17.00 :: MS-Eval: Benchmarking Sub-Agent Delegation and Orchestration in MCP-Intensive Working Scenarios [agent, benchmark, llm, open-source, world-model] sim+=0.18 sim-=0.05 sep=0.00
  - 15.0 / rank=15.00 :: SkillBrowser: Selective Skill Exposure for Scalable Skill-Using Agents [agent, llm, reasoning, tool-use] sim+=0.19 sim-=0.04 sep=0.00
  - 15.0 / rank=15.00 :: Execution Before Memory: Diagnosing Memory Failures in Video World Models [agent, benchmark, memory, world-model] sim+=0.36 sim-=0.11 sep=0.00
  - 15.0 / rank=15.00 :: GatewayBench: Unified Capability and Security Evaluation Platform for MCP Gateways [agent, benchmark, tool-use, world-model] sim+=0.19 sim-=0.04 sep=0.00
  - 15.0 / rank=15.00 :: SAFE-Mem: Secure Agentic Memory with Fine-Grained Authority and Governance-Aware Retrieval [agent, llm, memory, reasoning] sim+=0.20 sim-=0.05 sep=0.00
  - 15.0 / rank=15.00 :: 3DCodeVerse: Exploring 3D Worlds as Code [agent, fine-tuning, open-source, reasoning] sim+=0.21 sim-=0.06 sep=0.00
  - 14.0 / rank=14.00 :: OfficeTown: Generative Organizations for Evolving Agent Intelligence [agent, benchmark, fine-tuning, tool-use] sim+=0.21 sim-=0.05 sep=0.00
  - 13.0 / rank=13.00 :: GLOBE: Evolving Structured Agentic Memory for E-commerce Product Optimization in Generative Search [agent, memory, self-improvement] sim+=0.16 sim-=0.02 sep=0.00
  - 13.0 / rank=13.00 :: RL-Guided Temporal Localization for Dual-Channel Retrieval in Long-Horizon Agent Memory [agent, benchmark, llm, memory] sim+=0.18 sim-=0.06 sep=0.00
  - 13.0 / rank=13.00 :: QD-Rec: Open-Ended Evolution of Recommendations via Quality-Diversity [agent, reasoning, self-improvement] sim+=0.16 sim-=0.04 sep=0.00
  - 13.0 / rank=13.00 :: Language Models Learn to Improve Language Models [llm, self-improvement, world-model] sim+=0.15 sim-=0.04 sep=0.00
  - 13.0 / rank=13.00 :: MobileInject: Evaluating Reasoning-to-Action Security of MLLM-Based Mobile Agents under Cross-Modal Consistency Injection [agent, benchmark, llm, reasoning] sim+=0.18 sim-=0.04 sep=0.00
  - 13.0 / rank=13.00 :: Beyond Marginal Influence: Interaction-Aware Memory Selection for LLM Agents [agent, benchmark, llm, memory] sim+=0.17 sim-=0.06 sep=0.00
  - 13.0 / rank=13.00 :: Ignored, Not Refused: What Low Prompt-Injection Success Rates Hide in Small Tool-Using Agents [agent, benchmark, llm, tool-use] sim+=0.20 sim-=0.07 sep=0.00
  - 13.0 / rank=13.00 :: Not All Test-Time Compute Is Evidence: Model-Specific Bias in LLM Reasoning [agent, benchmark, llm, reasoning] sim+=0.23 sim-=0.14 sep=0.00
  - 13.0 / rank=13.00 :: ParetoChat: An LLM-Based Interface for Interactive Multi-Objective Optimization [agent, benchmark, llm, tool-use] sim+=0.18 sim-=0.05 sep=0.00
  - 13.0 / rank=13.00 :: Prism Attention: Proposal-Refined Index Sharing Mechanism for Efficient LLMs Inference [benchmark, llm, memory, reasoning] sim+=0.18 sim-=0.05 sep=0.00