VReST: Enhancing Reasoning in Large Vision-Language Models through Tree Search and Self-Reward Mechanism

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Congzhi, Peng, Jiawei, Wang, Zhenglin, Lai, Yilong, Sun, Haowen, Chang, Heng, Ma, Fei, Yu, Weijiang
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913887949946880
author Zhang, Congzhi
Peng, Jiawei
Wang, Zhenglin
Lai, Yilong
Sun, Haowen
Chang, Heng
Ma, Fei
Yu, Weijiang
author_facet Zhang, Congzhi
Peng, Jiawei
Wang, Zhenglin
Lai, Yilong
Sun, Haowen
Chang, Heng
Ma, Fei
Yu, Weijiang
contents Large Vision-Language Models (LVLMs) have shown exceptional performance in multimodal tasks, but their effectiveness in complex visual reasoning is still constrained, especially when employing Chain-of-Thought prompting techniques. In this paper, we propose VReST, a novel training-free approach that enhances Reasoning in LVLMs through Monte Carlo Tree Search and Self-Reward mechanisms. VReST meticulously traverses the reasoning landscape by establishing a search tree, where each node encapsulates a reasoning step, and each path delineates a comprehensive reasoning sequence. Our innovative multimodal Self-Reward mechanism assesses the quality of reasoning steps by integrating the utility of sub-questions, answer correctness, and the relevance of vision-language clues, all without the need for additional models. VReST surpasses current prompting methods and secures state-of-the-art performance across three multimodal mathematical reasoning benchmarks. Furthermore, it substantiates the efficacy of test-time scaling laws in multimodal tasks, offering a promising direction for future research.
format Preprint
id arxiv_https___arxiv_org_abs_2506_08691
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle VReST: Enhancing Reasoning in Large Vision-Language Models through Tree Search and Self-Reward Mechanism
Zhang, Congzhi
Peng, Jiawei
Wang, Zhenglin
Lai, Yilong
Sun, Haowen
Chang, Heng
Ma, Fei
Yu, Weijiang
Computer Vision and Pattern Recognition
Large Vision-Language Models (LVLMs) have shown exceptional performance in multimodal tasks, but their effectiveness in complex visual reasoning is still constrained, especially when employing Chain-of-Thought prompting techniques. In this paper, we propose VReST, a novel training-free approach that enhances Reasoning in LVLMs through Monte Carlo Tree Search and Self-Reward mechanisms. VReST meticulously traverses the reasoning landscape by establishing a search tree, where each node encapsulates a reasoning step, and each path delineates a comprehensive reasoning sequence. Our innovative multimodal Self-Reward mechanism assesses the quality of reasoning steps by integrating the utility of sub-questions, answer correctness, and the relevance of vision-language clues, all without the need for additional models. VReST surpasses current prompting methods and secures state-of-the-art performance across three multimodal mathematical reasoning benchmarks. Furthermore, it substantiates the efficacy of test-time scaling laws in multimodal tasks, offering a promising direction for future research.
title VReST: Enhancing Reasoning in Large Vision-Language Models through Tree Search and Self-Reward Mechanism
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.08691