AI Paper Atlas

Research map / Video / Video-language understanding

VisionCoach: Reinforcing Grounded Video Reasoning via Visual-Perception Prompting

Daeun Lee, Shoubin Yu, Yue Zhang, Mohit Bansal

ECCV 2026 · Video › Video-language understanding · approach: Long video understanding · 3 citations (OpenAlex)

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 changedWhereMeasuredEffect
apply VisionCoach training to Qwen2.5-VL-7BQwen2.5-VL-7B · V-STAR WhatAcc: 33.5 -> 61.1Improved
apply VisionCoach training to Qwen2.5-VL-7BQwen2.5-VL-7B · V-STARmAM: 19.3 -> 34.3 (+15.0)Improved
apply VisionCoach training to Qwen2.5-VL-7BQwen2.5-VL-7B · V-STARmLGM: 22.4 -> 47.5 (+25.1)Improved
use VisionCoach instead of Open-o3-videoOpen-o3-video · V-STARmAM: 33.4 -> 34.3Improved
use VisionCoach instead of Open-o3-videoOpen-o3-video · V-STARmLGM: 46.0 -> 47.5Improved
use VisionCoach instead of Open-o3-videoOpen-o3-video · V-STAR Where Chain2Spatial IoU: 5.9 -> 5.3Hurt
use VisionCoach instead of Open-o3-videoOpen-o3-video-7B · WorldSenseOverall accuracy: 37.5 -> 43.8Improved
use VisionCoach instead of Open-o3-videoOpen-o3-video-7B · VideoMMEOverall accuracy: 63.1 -> 63.3No clear effect
use VisionCoach instead of Open-o3-videoOpen-o3-video-7B · VideoMMMUOverall accuracy: 52.3 -> 54.4Improved
use VisionCoach instead of Open-o3-videoOpen-o3-video-7B · PerceptionTestOverall accuracy: 67.5 -> 68.7Improved
apply VisionCoach training to Qwen2.5-VL-7BQwen2.5-VL-7B · VideoMMEOverall accuracy: 62.4 -> 63.3Improved

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.

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

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy