Research map / Vision-language and multimodal / Multimodal LLMs
VisReason: A Large-Scale Dataset for Visual Chain-of-Thought Reasoning
Lingxiao Li, Yifan Wang, Xinyan Gao, Chen Tang, Xiangyu Yue, Chenyu You
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| fine-tune on VisReason-Pro with adaptive zoom-in | Qwen2.5-VL-7B · Visual-CoT average | accuracy: 0.770 -> 0.802 | Improved |
| fine-tune on VisReason instead of no fine-tuning (ablation) | Qwen2.5-VL-7B · Visual-CoT relation reasoning | accuracy: 0.598 -> 0.678 | Improved |
| fine-tune on VisReason instead of no fine-tuning (ablation) | Qwen2.5-VL-7B · Visual-CoT Doc/Text | accuracy: 0.920 -> 0.864 | Hurt |
| add adaptive zoom-in strategy (ablation) | VisReason-7B · Visual-CoT average | accuracy: 0.777 -> 0.789 | Improved |
| add adaptive zoom-in strategy (ablation) | VisReason-7B + Pro · Visual-CoT average | accuracy: 0.780 -> 0.802 | Improved |
| add VisReason-Pro data to VisReason (ablation) | VisReason-7B · Visual-CoT fine-grained | accuracy: 0.798 -> 0.809 | Improved |
| add VisReason-Pro data to VisReason (ablation) | VisReason-7B · Visual-CoT average | accuracy: 0.777 -> 0.780 | No clear effect |
| add VisReason-Pro data to VisReason (ablation) | VisReason-7B with AZ · Visual-CoT Doc/Text | accuracy: 0.882 -> 0.892 | Improved |
| evaluate with direct QA instead of multi-round CoT | VisReason-Pro-7B · Visual-CoT Text/Doc | accuracy: 0.892 -> 0.916 | Improved |
| evaluate with direct QA instead of multi-round CoT | VisReason-Pro-7B · Visual-CoT spatial relation | accuracy: 0.722 -> 0.599 | Hurt |
| evaluate with direct QA instead of multi-round CoT | VisReason-Pro-7B · V*Bench | accuracy: 0.603 -> 0.791 | Improved |
| evaluate with direct QA instead of multi-round CoT | VisReason-Pro-7B · MME | score: 0.777 -> 0.856 | Improved |
Closest papers by meaning
- MMFineReason: Closing the Multimodal Reasoning Gap via Open Data-Centric Methods (NeurIPS 2026 · similarity 0.93)
- VGR: Visual Grounded Reasoning (ICLR 2026 · similarity 0.93)
- TVI-CoT: Text-Visual Interleaved Chain-of-Thought Reasoning for Multimodal Understanding (ICML 2026 · similarity 0.93)
- Multi-step Visual Reasoning with Visual Tokens Scaling and Verification (NeurIPS 2025 · similarity 0.93)
- Imagine While Reasoning in Space: Multimodal Visualization-of-Thought (ICML 2025 · similarity 0.92)
- VLM-R³: Region Recognition, Reasoning, and Refinement for Enhanced Multimodal Chain-of-Thought (NeurIPS 2025 · similarity 0.92)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
