RAGEVALUATION-HJKMY/TSBC_100row_mistake_added-Evaluated-2025-04-10-15-04-40-gpt-4o-2024-08-06

该数据集包含了一系列的问题和对应的回答，同时还包含了用于评估回答准确性和完整性的多个评分指标。数据集中的字段包括问题文本、回答文本、相关文档、错误数量、错误分布、错误位置、关键点以及针对回答不正确时的事实准确性、关键点完整性、关键点虚构性、上下文相关性、连贯性和参与度等方面的评分。数据集分为训练集，适用于自然语言处理任务中的模型训练和评估。

The dataset consists of a series of questions and corresponding answers, along with multiple scoring metrics for evaluating the accuracy and completeness of the answers. The fields in the dataset include question text, answer text, related documents, number of mistakes, mistake distribution, locations of errors, key points, and scores for factual accuracy, key point completeness, key point hallucination, context relevance, coherence, and engagement in cases of incorrect answers. The dataset is split into a training set, suitable for model training and evaluation in natural language processing tasks.