lightonai/nv-embed-supervised-distill-dedup

该数据集是一个用于问答系统的数据集，包含文档、查询和分数信息。它包括fiqa、hotpotqa、msmarco、nq、fever、squadv2和trivia等多个数据分割，每个分割都有大量的示例。数据集的文档部分包含文档ID和文本，查询部分包含查询ID和查询文本，分数部分包含查询ID、相关文档ID列表和相关性分数。

This dataset is a question-answering system dataset, containing document, query, and score information. It includes multiple data splits such as fiqa, hotpotqa, msmarco, nq, fever, squadv2, and trivia, each with a large number of examples. The document part of the dataset contains document IDs and text, the query part contains query IDs and query text, and the score part contains query IDs, related document ID lists, and relevance scores.