PARC-DATASETS/error-detection-positives

这是一个包含数学推理问题的错误注释数据集，是PARC（前提注释推理集合）的一部分。该数据集结合了来自多个领域的正面样本，用于数学推理中的错误检测任务。每个示例包含问题的领域来源、数学问题陈述、正确答案、模型的答案（可能包含错误）、推理步骤、每个步骤的前提和结论的详细注释以及标记步骤和链路级别错误的注释。

This dataset is part of the PARC (Premise-Annotated Reasoning Collection) and contains mathematical reasoning problems with error annotations. This dataset combines positive samples from multiple domains for the task of error detection in mathematical reasoning. Each example includes the domain/source of the problem, the mathematical problem statement, the correct answer, the models answer (which may contain errors), a list of individual reasoning steps, detailed annotations of premises and conclusions for each step, and annotations marking errors at both step and chain levels.