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

TL;DR

VReST introduces a training-free method combining tree search and self-reward to significantly improve complex reasoning in large vision-language models, achieving state-of-the-art results in multimodal mathematical reasoning benchmarks.

Contribution

The paper presents VReST, a novel approach using Monte Carlo Tree Search and self-reward mechanisms to enhance reasoning in LVLMs without additional training.

Findings

VReST outperforms existing prompting methods on three benchmarks.

It demonstrates the effectiveness of test-time scaling laws in multimodal tasks.

VReST achieves state-of-the-art performance in multimodal mathematical reasoning.

Abstract

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.