CodePercept: 基于代码的MLLM视觉STEM感知
CodePercept: Code-Grounded Visual STEM Perception for MLLMs
AI总结：
本文发现多模态大模型在STEM视觉推理中感知能力是瓶颈，提出通过代码作为感知媒介，构建ICC-1M数据集和STEM2Code-Eval基准来增强和评估感知能力。
摘要：
当多模态大模型（MLLM）在科学、技术、工程和数学（STEM）视觉推理中失败时，一个基本问题出现了：是由于感知缺陷还是推理限制？通过独立缩放感知和推理组件的系统缩放分析，我们揭示了一个关键见解：缩放感知始终优于缩放推理。这表明感知是限制当前STEM视觉推理的真正杠杆。受此启发，我们的工作专注于通过将代码建立为强大的感知媒介——可执行代码提供精确的语义，自然与STEM视觉的结构化性质对齐——来系统增强MLLM的感知能力。具体来说，我们构建了ICC-1M，一个包含100万图像-标题-代码三元组的大规模数据集，通过两种互补方法实现了这种代码即感知范式：（1）基于代码的标题生成将可执行代码视为图像标题的真实标准，消除了现有知识蒸馏方法固有的幻觉；（2）STEM图像到代码翻译提示模型生成重建代码，减轻了自然语言在感知增强中的歧义。为了验证这一范式，我们进一步引入了STEM2Code-Eval，一个直接评估STEM领域视觉感知的新基准。与依赖问题解决准确性作为代理（仅测量问题相关理解）的现有工作不同，我们的基准通过可执行代码生成进行图像重建，要求全面的视觉理解，提供确定性和可验证的评估。代码可在 https://this URL 获取。
英文摘要：
When MLLMs fail at Science, Technology, Engineering, and Mathematics (STEM) visual reasoning, a fundamental question arises: is it due to perceptual deficiencies or reasoning limitations? Through systematic scaling analysis that independently scales perception and reasoning components, we uncover a critical insight: scaling perception consistently outperforms scaling reasoning. This reveals perception as the true lever limiting current STEM visual reasoning. Motivated by this insight, our work focuses on systematically enhancing the perception capabilities of MLLMs by establishing code as a powerful perceptual medium--executable code provides precise semantics that naturally align with the structured nature of STEM visuals. Specifically, we construct ICC-1M, a large-scale dataset comprising 1M Image-Caption-Code triplets that materializes this code-as-perception paradigm through two complementary approaches: (1) Code-Grounded Caption Generation treats executable code as ground truth for image captions, eliminating the hallucinations inherent in existing knowledge distillation methods; (2) STEM Image-to-Code Translation prompts models to generate reconstruction code, mitigating the ambiguity of natural language for perception enhancement. To validate this paradigm, we further introduce STEM2Code-Eval, a novel benchmark that directly evaluates visual perception in STEM domains. Unlike existing work relying on problem-solving accuracy as a proxy that only measures problem-relevant understanding, our benchmark requires comprehensive visual comprehension through executable code generation for image reconstruction, providing deterministic and verifiable assessment. Code is available at https://github.com/TongkunGuan/Qwen-CodePercept.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
