AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

HART: High-Resolution Annotation-Free Reasoning Technique through a Closed-loop Framework

Jiacheng Yang, Anqi Chen, Yunkai Dang, Qi Fan, Cong Wang, Wenbin Li, Feng Miao, Yang Gao

ECCV 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Chain-of-thought multimodal reasoning · 2 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 HART post-trainingQwen2.5-VL-7B · MME-RealWorld-Lite perceptionaccuracy: 46.4 -> 64.9Improved
apply HART post-trainingQwen2.5-VL-7B · MME-RealWorld-Lite reasoningaccuracy: 35.9 -> 58.5Improved
apply HART post-trainingQwen2.5-VL-7B · TreeBench perceptionaccuracy: 43.6 -> 57.0Improved
apply HART post-trainingInternVL3-8B · MME-RealWorld-Lite perceptionaccuracy: 51.1 -> 61.8Improved
apply HART post-trainingInternVL3-8B · TreeBench reasoningaccuracy: 34.4 -> 35.9Improved
add SFT stage after AP-GRPO (k=0.60) (ablation)HART · MME-RealWorld-Lite perceptionaccuracy: 53.3 -> 64.9Improved
run AP-GRPO before SFT rather than SFT first (ablation)HART (k=0.60) · MME-RealWorld-Lite perceptionaccuracy: 51.7 -> 64.9Improved
replace GRPO with AP-GRPO in the closed loopQwen2.5-VL-7B · TreeBench grounding (correct %)grounding accuracy: 51.0% -> 75.4%Improved
replace GRPO with AP-GRPO in the closed loopQwen2.5-VL-7B · Visual-CoT grounding (correct %)grounding accuracy: 66.1% -> 77.7%Improved
answer from cropped ROIs only (closed loop)Qwen2.5-VL-7B · Visual CoT incorrect grounding given correct answerproportion: 36.5% -> 21.5%Improved
closed loop on InternVL3-8BInternVL3-8B · Visual CoT incorrect grounding given correct answerproportion: 63.8% -> 55.9%Improved
lower k from 0.60 to 0.15 in AP-GRPO + SFT (ablation)HART · MME-RealWorld-Liteperception / reasoning accuracy: 64.9 -> 67.0 perception, 58.5 -> 56.9 reasoningMixed

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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