patrickmlml/yourbench_example

该数据集包含了文档的片段、摘要以及与问题回答相关的多种信息。具体包括文档ID、文档文本、文件名、文档元数据、摘要、问题、答案、问题类别、难度评估、引用、多跳片段信息等。数据集分为多个配置，每个配置都有其特定的特征和应用场景，如文档分段、问题评估、多跳和单次提问处理以及文档摘要生成。

The dataset includes various information related to document chunks, summaries, and question answering. It specifically contains document ID, document text, filename, document metadata, summaries, questions, answers, question categories, difficulty assessment, citations, multi-hop chunk information, and more. The dataset is divided into multiple configurations, each with its specific features and application scenarios, such as document chunking, question evaluation, multi-hop and single-shot question processing, and document summary generation.