Research map / Video / Video-language understanding
VisionCoach: Reinforcing Grounded Video Reasoning via Visual-Perception Prompting
Daeun Lee, Shoubin Yu, Yue Zhang, Mohit Bansal
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 |
|---|---|---|---|
| apply VisionCoach training to Qwen2.5-VL-7B | Qwen2.5-VL-7B · V-STAR What | Acc: 33.5 -> 61.1 | Improved |
| apply VisionCoach training to Qwen2.5-VL-7B | Qwen2.5-VL-7B · V-STAR | mAM: 19.3 -> 34.3 (+15.0) | Improved |
| apply VisionCoach training to Qwen2.5-VL-7B | Qwen2.5-VL-7B · V-STAR | mLGM: 22.4 -> 47.5 (+25.1) | Improved |
| use VisionCoach instead of Open-o3-video | Open-o3-video · V-STAR | mAM: 33.4 -> 34.3 | Improved |
| use VisionCoach instead of Open-o3-video | Open-o3-video · V-STAR | mLGM: 46.0 -> 47.5 | Improved |
| use VisionCoach instead of Open-o3-video | Open-o3-video · V-STAR Where Chain2 | Spatial IoU: 5.9 -> 5.3 | Hurt |
| use VisionCoach instead of Open-o3-video | Open-o3-video-7B · WorldSense | Overall accuracy: 37.5 -> 43.8 | Improved |
| use VisionCoach instead of Open-o3-video | Open-o3-video-7B · VideoMME | Overall accuracy: 63.1 -> 63.3 | No clear effect |
| use VisionCoach instead of Open-o3-video | Open-o3-video-7B · VideoMMMU | Overall accuracy: 52.3 -> 54.4 | Improved |
| use VisionCoach instead of Open-o3-video | Open-o3-video-7B · PerceptionTest | Overall accuracy: 67.5 -> 68.7 | Improved |
| apply VisionCoach training to Qwen2.5-VL-7B | Qwen2.5-VL-7B · VideoMME | Overall accuracy: 62.4 -> 63.3 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- Open-o3-video-7B — Open-o3-Video: Grounded Video Reasoning with Explicit Spatio-Temporal Evidence (ICML 2026) · better on 3 of 3 shared results
- Video-o3 — Video-o3: Native Interleaved Clue Seeking for Long Video Multi-Hop Reasoning (ICML 2026) · better on 1 of 1 shared result
- VideoZoomer — VideoZoomer: Reinforcement-Learned Temporal Focusing for Long Video Reasoning (ICLR 2026) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Grounded Reinforcement Learning for Visual Reasoning (NeurIPS 2025 · similarity 0.90)
- Motion-Grounded Video Reasoning: Understanding and Perceiving Motion at Pixel Level (CVPR 2025 · similarity 0.90)
- When Thinking Drifts: Evidential Grounding for Robust Video Reasoning (NeurIPS 2025 · similarity 0.90)
- VideoRFT: Incentivizing Video Reasoning Capability in MLLMs via Reinforced Fine-Tuning (NeurIPS 2025 · similarity 0.90)
- Video-R1: Reinforcing Video Reasoning in MLLMs (NeurIPS 2025 · similarity 0.90)
- Look, Remember and Reason: Grounded Reasoning in Videos with Language Models (ICLR 2024 · similarity 0.90)
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.
