FelixFester/NuminaMath-CoT

NuminaMath CoT数据集包含约86万个数学问题，每个问题的解决方案都以思维链（CoT）的形式呈现。数据来源广泛，包括中国高中数学练习题、美国和国际数学奥林匹克竞赛问题等。数据集主要通过在线考试试卷PDF和数学讨论论坛收集，并经过OCR、分割、翻译、重新对齐和最终答案格式化等处理步骤。数据源详细分类包括aops_forum、amc_aime、cn_k12、gsm8k、math、olympiads、orca_math、synthetic_amc和synthetic_math，总样本数为859,608个。

The NuminaMath CoT dataset contains approximately 860k math problems, where each solution is formatted in a Chain of Thought (CoT) manner. The sources of the dataset range from Chinese high school math exercises to US and international mathematics olympiad competition problems. The data were primarily collected from online exam paper PDFs and mathematics discussion forums, and processed through steps including OCR, segmentation into problem-solution pairs, translation into English, realignment to produce a CoT reasoning format, and final answer formatting. The source breakdown includes aops_forum, amc_aime, cn_k12, gsm8k, math, olympiads, orca_math, synthetic_amc, and synthetic_math, with a total of 859,608 samples.