AStar: Boosting Multimodal Reasoning with Automated Structured Thinking
Jinyang Wu, Mingkuan Feng, Guocheng Zhai, Shuai Zhang, Zheng Lian, Fangrui Lv, Pengpeng Shao, Ruihan Jin
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 **AStar**, a training-free, **A**utomatic **S**tructured **t**hinking paradigm for multimod**a**l **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
BibTeX
@inproceedings{aaai2026_astarboostingmul,
title = {AStar: Boosting Multimodal Reasoning with Automated Structured Thinking},
author = {Jinyang Wu and Mingkuan Feng and Guocheng Zhai and Shuai Zhang and Zheng Lian and Fangrui Lv and Pengpeng Shao and Ruihan Jin and Zhengqi Wen and Jianhua Tao},
booktitle = {AAAI 2026},
year = {2026}
}