AStar: Boosting Multimodal Reasoning with Automated Structured Thinking

TL;DR

AStar is a training-free, structured thinking framework that enhances multimodal reasoning by adaptively using thought cards, leading to improved accuracy and transferability without extensive search or retraining.

Contribution

It introduces a novel thought card library and an adaptive retrieval method, enabling efficient, plug-and-play multimodal reasoning without additional training.

Findings

Achieves 53.9% accuracy on MathVerse, surpassing GPT-4o.

Attains 32.7% accuracy on MathVision, outperforming GPT-4o.

Thought cards transfer effectively across different reasoning tasks.

Abstract

Multimodal large language models excel across diverse domains but struggle with complex visual reasoning tasks. To enhance their reasoning capabilities, current approaches typically rely on explicit search or post-training techniques. However, search-based methods suffer from computational inefficiency due to extensive solution space exploration, while post-training methods demand substantial data, computational resources, and often exhibit training instability. To address these challenges, we propose \textbf{AStar}, a training-free, \textbf{A}utomatic \textbf{S}tructured \textbf{t}hinking paradigm for multimod\textbf{a}l \textbf{r}easoning. Specifically, we introduce novel ``thought cards'', a lightweight library of high-level reasoning patterns abstracted from prior samples. For each test problem, AStar adaptively retrieves the optimal thought cards and seamlessly integrates these external explicit guidelines with the model's internal implicit reasoning capabilities. Compared to previous methods, AStar eliminates computationally expensive explicit search and avoids additional complex post-training processes, enabling a more efficient reasoning approach. Extensive experiments demonstrate that our framework achieves 53.9\% accuracy on MathVerse (surpassing GPT-4o's 50.2\%) and 32.7\% on MathVision (outperforming GPT-4o's 30.4\%). Further analysis reveals the remarkable transferability of our method: thought cards generated from mathematical reasoning can also be applied to other reasoning tasks, even benefiting general visual perception and understanding. AStar serves as a plug-and-play test-time inference method, compatible with other post-training techniques, providing an important complement to existing multimodal reasoning approaches.