llamastack/mmmu

该数据集包含多个特定领域的数据集，如会计、农业、建筑与工程、艺术、艺术理论、基础医学、生物学、化学、临床医学、计算机科学、设计、诊断与实验室医学、经济学、电子学、能源与动力、金融、地理、历史、文学和管理等。每个数据集包含的特征有：id、问题、选项、解释、多张图片、图片类型、答案、题目难度、问题类型、子领域、聊天完成输入、输入查询和预期答案。数据集被分为开发集、验证集和测试集，每个集都有指定的样本数量和字节大小。还提供了下载大小和数据集大小。

This dataset includes multiple datasets corresponding to specific fields such as Accounting, Agriculture, Architecture_and_Engineering, Art, Art_Theory, Basic_Medical_Science, Biology, Chemistry, Clinical_Medicine, Computer_Science, Design, Diagnostics_and_Laboratory_Medicine, Economics, Electronics, Energy_and_Power, Finance, Geography, History, Literature, and Manage. Each dataset contains features like id, question, options, explanation, multiple images, img_type, answer, topic_difficulty, question_type, subfield, chat_completion_input, input_query, and expected_answer. The datasets are divided into dev, validation, and test sets with specified numbers of examples and bytes. The download and dataset sizes are also provided.