sumuks/yourbench_example

该数据集包含了多种配置，每种配置都有不同的特征。主要特征包括文档ID、文档文本、文档文件名、文档元数据（如文件大小）、文档摘要、摘要模型、文本块信息（如块ID、块文本）、多跳块信息、块信息指标（如平均令牌长度、困惑度等）、分块模型等。此外，数据集还包含了问题生成、问题回答、问题难度评估等特征。数据集被划分为训练集，并提供了不同配置的训练数据。

The dataset consists of multiple configurations, each with different features. Main features include document ID, document text, document filename, document metadata (such as file size), document summary, summarization model, text chunk information (such as chunk ID, chunk text), multi-hop chunk information, chunk information metrics (such as average token length, perplexity), chunking model, etc. Additionally, the dataset includes features for question generation, question answering, and question difficulty assessment. The dataset is split into training sets and provides training data for different configurations.