MBZUAI/VideoMathQA

VideoMathQA是一个评估在真实世界教育视频中进行数学推理的基准。它要求模型理解和整合视觉、音频和文本三种模态的信息，并跨越时间进行推理。该基准包括三种推理类型：问题聚焦、概念迁移和深度教学理解。每个问题根据数学概念、视频时长、难度和推理类型四个维度进行评估。数据集包含420个由专家策划的问题，每个问题都有详细的解题步骤。

VideoMathQA is a benchmark designed to evaluate mathematical reasoning in real-world educational videos. It requires models to interpret and integrate information from visuals, audio, and text across time. The benchmark includes three types of reasoning: Problem Focused, Concept Transfer, and Deep Instructional Comprehension. Each question is evaluated across four dimensions: mathematical concepts, video duration, difficulty level, and reasoning type. The dataset consists of 420 expert-curated questions, each with detailed step-by-step reasoning provided.