ViRC：通过推理分块增强视觉交错数学思维链
ViRC: Enhancing Visual Interleaved Mathematical CoT with Reason Chunking
提出 Reason Chunking 机制，将多模态数学思维链结构化为关键推理单元（CRU），并在单元间动态调用视觉工具获取图像信息，为视觉语言模型在需要反复查看图像的推理任务中提供了可复用的分块推理与渐进训练范式。
原文依据 introducing a Reason Chunking mechanism that structures multimodal mathematical CoT into consecutive Critical Reasoning Units (CRUs)
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论思维链（CoT）显著增强了大型语言模型的推理能力，但在扩展到多模态领域时面临挑战，尤其是在数学任务中。现有的多模态大语言模型通常仅从单张静态数学图像进行文本推理，忽略了推理过程中的动态视觉获取。相比之下，人类会反复查看视觉图像并采用逐步推理来证明中间命题。这种将问题求解过程分解为关键逻辑节点的策略符合认知科学中的米勒定律。受此启发，我们提出了一个面向多模态数学任务的 ViRC 框架，引入推理分块（Reason Chunking）机制，将多模态数学思维链结构化为连续的关键推理单元（CRU），以模拟人类专家的解题模式。CRU 确保单元内文本连贯性以验证中间命题，同时跨单元整合视觉信息以生成后续命题并支持结构化推理。为此，我们提出了 CRUX 数据集，使用三种视觉工具和四种推理模式，为每个数学问题的多条推理路径提供显式标注的 CRU。利用 CRUX 数据集，我们提出了一种受人类认知学习启发的渐进式训练策略，包括指令监督微调（Instructional SFT）、练习监督微调（Practice SFT）和策略强化学习（Strategic RL），旨在进一步增强模型的推理分块能力。最终得到的 ViRC-7B 模型在多个数学基准上相比基线平均提升 18.8%。代码将公开。
英文原文摘要 1656 字符
CoT has significantly enhanced the reasoning ability of LLMs while it faces challenges when extended to multimodal domains, particularly in mathematical tasks.Existing MLLMs typically perform textual reasoning solely from a single static mathematical image, overlooking dynamic visual acquisition during reasoning.In contrast, humans repeatedly examine visual image and employ step-by-step reasoning to prove intermediate propositions.This strategy of decomposing the problem-solving process into key logical nodes adheres to Miller's Law in cognitive science.Inspired by this insight, we propose a **ViRC** framework for multimodal mathematical tasks, introducing a **Reason Chunking** mechanism that structures multimodal mathematical CoT into consecutive **Critical Reasoning Units (CRUs)** to simulate human expert problem-solving patterns.CRUs ensure intra-unit textual coherence for intermediate proposition verification while integrating visual information across units to generate subsequent propositions and support structured reasoning.To this end, we present **CRUX** dataset by using three visual tools and four reasoning patterns to provide explicitly annotated CRUs across multiple reasoning paths for each mathematical problem.Leveraging the CRUX dataset, we propose a progressive training strategy inspired by human cognitive learning, which includes Instructional SFT, Practice SFT, and Strategic RL, aimed at further strengthening the Reason Chunking ability of the model.The resulting **ViRC-7B** model achieves a 18.8% average improvement over baselines across multiple mathematical benchmarks.The codes will be made publicly available.