MMTutorBench：首个多模态AI数学辅导基准
MMTutorBench: The First Multimodal Benchmark for AI Math Tutoring
AI总结：
MMTutorBench首次提出多模态AI数学辅导基准，包含685个围绕教学关键步骤构建的问题，通过六维度细粒度评估和三个任务（洞察发现、操作构建、操作执行）评估12个顶级MLLMs，揭示了专有与开源系统间的性能差异及OCR、少样本提示等对辅导质量的影响。
摘要：
有效的数学辅导不仅需要解决问题，还需诊断学生困难并逐步引导。尽管多模态大语言模型（MLLMs）表现出潜力，但现有基准大多忽视了这些辅导技能。我们引入MMTutorBench，首个AI数学辅导基准，包含685个围绕教学关键步骤构建的问题。每个问题配以问题特定的评分标准，可在六个维度上进行细粒度评估，并分为三个任务：洞察发现、操作构建和操作执行。我们评估了12个领先的MLLMs，发现专有与开源系统之间存在明显的性能差距，与人类导师相比仍有显著提升空间，并在输入变体上呈现出一致的趋势：OCR流程会降低辅导质量，少样本提示带来的增益有限，而基于评分标准的LLM-as-a-Judge证明了高度可靠性。这些结果突显了MMTutorBench在推动AI辅导方面的难度和诊断价值。
英文摘要：
Effective math tutoring requires not only solving problems but also diagnosing students' difficulties and guiding them step by step. While multimodal large language models (MLLMs) show promise, existing benchmarks largely overlook these tutoring skills. We introduce MMTutorBench, the first benchmark for AI math tutoring, consisting of 685 problems built around pedagogically significant key-steps. Each problem is paired with problem-specific rubrics that enable fine-grained evaluation across six dimensions, and structured into three tasks-Insight Discovery, Operation Formulation, and Operation Execution. We evaluate 12 leading MLLMs and find clear performance gaps between proprietary and open-source systems, substantial room compared to human tutors, and consistent trends across input variants: OCR pipelines degrade tutoring quality, few-shot prompting yields limited gains, and our rubric-based LLM-as-a-Judge proves highly reliable. These results highlight both the difficulty and diagnostic value of MMTutorBench for advancing AI tutoring.
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。
