AI Paper Atlas

Research map / Video / Video-language understanding

Thinking With Videos: Multimodal Tool-Augmented Reinforcement Learning for Long Video Reasoning

Haoji Zhang, Xin Gu, Jiawen Li, Chixiang Ma, Sule Bai, Chubin Zhang, Bowen Zhang, Zhichao Zhou, Dongliang He, Yansong Tang

CVPR 2026 · Video › Video-language understanding · approach: Video LLMs with fine-grained understanding

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
add DGRPO (difficulty-aware rewards) in place of GRPO (ablation)VITAL-7B (SFT, think, no tools) · LVR / VidCh / MMMU / Cha averageaverage score: 50.3 -> 52.1Improved
add DGRPO in place of GRPO (ablation)VITAL-7B · LongVideo-Reasonaccuracy: 66.0 -> 70.2Improved
add DGRPO in place of GRPO (ablation)VITAL-7B · Charades-STAmIoU: 57.2 -> 57.1No clear effect
add tool-augmented SFT and DGRPO stages (video clip tool) (ablation)VITAL-7B · LVR / VidCh / MMMU / Cha averageaverage score: 52.1 -> 57.1Improved
add toolbox at inference (ablation)VITAL-7B · LongVideo-Reasonaccuracy: 73.2 -> 79.3Improved
add toolbox at inference (ablation)VITAL-7B · VidChapters-7MR@0.5: 25.8 -> 34.7Improved
add thinking (reasoning) style to SFT+GRPO (ablation)VITAL-7B · LVR / VidCh / MMMU / Cha averageaverage score: 48.3 -> 50.3Improved
add GRPO after SFT (ablation)VITAL-7B (think) · LVR / VidCh / MMMU / Cha averageaverage score: 43.9 -> 50.3Improved
train on temporal grounding + reasoning VQA + grounded VQA instead of a single task (same 73k size) (ablation)VITAL-7B · LVR / VidCh / MMMU / Cha averageaverage score: 45.4 (TG) / 42.4 (RQA) -> 53.9Improved
scale training data from 73k to 182k (ablation)VITAL-7B · LVR / VidCh / MMMU / Cha averageaverage score: 53.9 -> 57.1Improved
use clip caption or clip QA tool (Gemini-2.5-Pro) instead of video clip tool (ablation)VITAL-7B · LVR / VidCh / MMMU / Cha averageaverage score: 57.1 -> 51.8 (caption) / 53.0 (QA)Hurt
add clip caption or clip QA tool zero-shot to GPT-4.1 (ablation)GPT-4.1 · VidChapters-7M temporal grounding (zero-shot)mIoU: 28.4 -> 2.0Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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