AI Paper Atlas

Research map / Video / Video-language understanding

VideoTIR: Accurate Understanding for Long Videos with Efficient Tool-Integrated Reasoning

Zhe Gao, Shiyu Shen, Taifeng Chai, Weinong Wang, Haotian Xu, Xing W, Wenbin Li, Qi Fan, Yang Gao, Dacheng Tao

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
replace GRPO with TAGPO (ablation)VideoTIR 7B · Validation at 30 training stepsval accuracy: 21.1 -> 24.6Improved
replace GRPO with TAGPO (ablation)VideoTIR 7B (8-frame Low) · Video-MME w/o sub, MVBench, LVB valaccuracy: Video-MME 54.1 -> 54.2; MVBench 56.9 -> 56.0; LVB 50.9 -> 51.3Mixed
add RL (GRPO) on top of zero-shot tool use (ablation)Qwen2.5-VL-7B · Video-MME w/o sub (8-frame Low)accuracy: 42.9 -> 54.1Improved
add tool calling (zero-shot) to thinking Qwen2.5-VL-7B at low res (ablation)Qwen2.5-VL-7B · Video-MME w/o sub (8-frame Low)accuracy: 35.0 -> 42.9Improved
increase input to 16 frames at high resolution (ablation)VideoTIR (TAGPO) · Video-MME w/o sub (Mean)accuracy: 54.2 -> 57.9Improved
enable thinking mode on Qwen2.5-VL-7B (ablation)Qwen2.5-VL-7B · Video-MME w/o sub (Mean)accuracy: 53.8 -> 49.5Hurt
use Random-Noised SFT (vision masked) instead of normal SFT for cold start (ablation)Qwen2.5-VL-3B · MVBench (10% subset)accuracy: 36.3 -> 40.7Improved
use Random-Noised SFT instead of normal SFT (ablation)Qwen2.5-VL-3B · Video-MME (10% subset)accuracy: 35.0 -> 38.5Improved
add SFT cold start with synthetic trajectories (ablation)Qwen2.5-VL-3B · Format qualityformat quality: <=1.0 -> 90.1 (Long prompt)Improved
use internal toolkit instead of external tool pipelinesVideoTIR · Efficiencylatency (s): 68.3s (COLT) -> 4.6sImproved
VideoTIR 16-frame High vs base Qwen2.5-VL-7BQwen2.5-VL-7B · MVBenchaccuracy: 60.6 -> 64.3Improved
Zero-RL directly on 3B modelQwen2.5-VL-3B · Zero-RL trainingvalid tool callsHurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 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