hasan-trash/yourbench_example

该数据集包含多个配置，每个配置针对不同的文本处理任务，如文档分段、文档摄入、问题生成、多跳问题和单次问答等。每个配置都提供了文档的标识、文本内容、元数据等信息。部分配置还包括了文档摘要、问题类别、难度评估等特征。数据集分为训练集，并提供了每种配置的训练集大小和示例数量。

The dataset consists of multiple configurations, each tailored for different text processing tasks such as document chunking, document ingestion, question generation, multi-hop questions, and single-shot questions. Each configuration provides features like document identification, text content, metadata, etc. Some configurations also include document summaries, question categories, difficulty assessments, and other features. The dataset is split into training sets, with the size and number of examples provided for each configuration.