AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

VisRL: Intention-Driven Visual Perception via Reinforced Reasoning

Zhangquan Chen, Xufang Luo, Dongsheng Li

ICCV 2025 · Vision-language and multimodal › Multimodal LLMs · approach: Reinforcement learning for visual reasoning · 41 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
add RL1 (bounding box DPO) after SFT (ablation)VisRL on LLaVA-1.5-7B · Visual CoT benchmark (11 datasets, average)average accuracy: 0.538 -> 0.568Improved
add RL2 (joint box and response DPO) after RL1 (ablation)VisRL on LLaVA-1.5-7B · Visual CoT benchmark (11 datasets, average)average accuracy: 0.568 -> 0.605Improved
add RL1 and RL2 to SFT on Qwen2.5-VL-7B (ablation)VisRL · Visual CoT benchmark (11 datasets, average)average accuracy: 0.730 -> 0.796Improved
add RL1 before RL2 (instead of RL2-only) (ablation)VisRL on Qwen2.5-VL-7B · Visual CoT benchmark, final-response training strategies (average)average accuracy: 0.759 -> 0.796Improved
use bounding box CoT RL instead of direct response DPO (ablation)VisRL on Qwen2.5-VL-7B · Visual CoT benchmark, final-response training strategies (average)average accuracy: 0.656 -> 0.796Improved
use bounding box CoT RL instead of PPO on final response (ablation)VisRL on Qwen2.5-VL-7B · Visual CoT benchmark, final-response training strategies (average)average accuracy: 0.661 -> 0.796Improved
SFT 30k instead of 438k bounding box data (ablation)LLaVA-1.5-7B · Visual CoT benchmark (11 datasets, average)average accuracy: 0.547 -> 0.538No clear effect
add RL after SFT for bounding box accuracy (ablation)VisRL LLaVA-1.5-7B · Visual CoT benchmark bounding box detection (Acc@0.5, average, LLaVA-1.5-7B)Top-1 Accuracy@0.5: 0.374 -> 0.437Improved
one iteration of data regeneration and retrainingVisRL on LLaVA-1.5 · MMEscore: 1526.3 -> 1560.0Improved
add diversity controller to data generation (ablation)VisRL data pipeline · Data generation pipeline (Qwen2.5-VL-7B)WP-LN proportion: 12.62% reduction without itImproved
add bounding box evaluation to data generation (ablation)VisRL data pipeline · Data generation pipeline (Qwen2.5-VL-7B)WP-LN proportion: 33.62% drop without itImproved
use pre-SFT model as critic instead of SFTed model (ablation)VisRL data pipeline · Data generation pipeline (Qwen2.5-VL-7B)WP-LN proportion: 54.68% -> 64.64%Improved

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