paper-with-me

Papers

VR-Thinker: Boosting Video Reward Models through Thinking-with-Image Reasoning

2025-10-12 · Qunzhong Wang, Jie Liu, Jiajun Liang, Yilei Jiang, Yuanxing Zhang, Yaozhi Zheng, Xintao Wang, Pengfei Wan, Xiangyu Yue, Jiaheng Liu arxiv

Recent advancements in multimodal reward models (RMs) have substantially improved post-training for visual generative models. However, current RMs face inherent limitations: (1) visual inputs consume large context budgets, forcing fewer frames and causing loss of fine-grained details; and (2) all visual information is packed into the initial prompt, exacerbating hallucination and forgetting during chain-of-thought reasoning. To overcome these issues, we introduce VideoReward Thinker (VR-Thinker), a thinking-with-image framework that equips the RM with visual reasoning operations (e.g., select frame) and a configurable visual memory window. This allows the RM to actively acquire and update visual evidence within context limits, improving reasoning fidelity and reliability. We activate visual reasoning via a reinforcement fine-tuning pipeline: (i) Cold Start with curated visual chain-of-thought data to distill basic reasoning skills and operation formatting; (ii) select samples whose per-dimension and overall judgments are all correct, then conduct Rejection sampling Fine-Tuning on these high-quality traces to further enhance reasoning; and (iii) apply Group Relative Policy Optimization (GRPO) to strengthen reasoning. Our approach delivers state-of-the-art accuracy among open-source models on video preference benchmarks, especially for longer videos: a 7B VR-Thinker achieves 80.5% on VideoGen Reward, 82.3% on GenAI-Bench, and 75.6% on MJ-Bench-Video. These results validate the effectiveness and promise of thinking-with-image multimodal reward modeling.

📄 PDF Abstract BibTeX arXiv:2510.10518

Code (0)

등록된 구현이 없습니다.

Tasks

Visual Reasoning

Similar Papers 제목 키워드 기반

Video-Thinker: Sparking "Thinking with Videos" via Reinforcement Learning

2025-10-27 · Shijian Wang, Jiarui Jin, Xingjian Wang, Linxin Song 외 arxiv

Recent advances in image reasoning methods, particularly "Thinking with Images", have demonstrated remarkable success in Multimodal Large Language Models (MLLMs); however, this dynamic reasoning paradigm has not yet been…

Reinforcement Learning

FrameThinker: Learning to Think with Long Videos via Multi-Turn Frame Spotlighting

2025-09-29 · Zefeng He, Xiaoye Qu, Yafu Li, Siyuan Huang 외 arxiv

While Large Vision-Language Models (LVLMs) have achieved substantial progress in video understanding, their application to long video reasoning is hindered by uniform frame sampling and static textual reasoning, which ar…

Reinforcement Learning

VLA-Thinker: Boosting Vision-Language-Action Models through Thinking-with-Image Reasoning

2026-03-15 · Chaoyang Wang, Wenrui Bao, Sicheng Gao, Bingxin Xu 외 arxiv

Vision-Language-Action (VLA) models have shown promising capabilities for embodied intelligence, but most existing approaches rely on text-based chain-of-thought reasoning where visual inputs are treated as static contex…

Reinforcement Learning

4DThinker: Thinking with 4D Imagery for Dynamic Spatial Understanding

2026-05-07 · Zhangquan Chen, Manyuan Zhang, Xinlei Yu, Xiang An 외 arxiv

Dynamic spatial reasoning from monocular video is essential for bridging visual intelligence and the physical world, yet remains challenging for vision-language models (VLMs). Prior approaches either verbalize spatial-te…

Reinforcement LearningSpatial Reasoning

ProxyThinker: Test-Time Guidance through Small Visual Reasoners

2025-05-30 · Zilin Xiao, Jaywon Koo, Siru Ouyang, Jefferson Hernandez 외

Recent advancements in reinforcement learning with verifiable rewards have pushed the boundaries of the visual reasoning capabilities in large vision-language models (LVLMs). However, training LVLMs with reinforcement fi…

Visual Reasoning