---
title: LLaVA’s IT
author: Anonym
date: 2023-04-17 00:00:00 +0800
categories: [NeurIPS 2023]
tags: [IT Dataset]
math: true
pin: false
---

- Paper: [Visual Instruction Tuning](https://arxiv.org/abs/2304.08485)
- [GitHub Link](https://llava-vl.github.io)
- Publisher: `NeurIPS 2023`
- Author Affiliation: `University of Wisconsin–Madison & Microsoft Research & Columbia University`
- Type
  + [x] SFT
  + [ ] RLHF
- Multi-turn
  + [x] ✔
  + [ ] ✖
- Input Modalities $\rightarrow$ Output Modalities <br />(I: Image, V: Video, A: Audio, 3D: Point Cloud, T: Text, B: Bounding box, Tab: Table, Web: Web page)
  + I+T $\rightarrow$ T
- Source
  + `MS-COCO`
- Method
  + `Auto.`
- I/V/A Scale
  + I
    * `81K`
  + V
    * `Not report`
  + A
    * `Not report`
- Dialog Turn
  + `2.29`
- Instance Scale
  + `150K`