mteb/cosqa

该数据集是一个包含文本和查询信息的语料库，分为四个部分：corpus, default, qrels, queries。corpus部分包含文本和标题，用于文本分析；default和qrels部分包含查询ID、语料库ID和分数，可能用于信息检索任务；queries部分仅包含文本，可能用于查询构建。数据集包含训练集、测试集和验证集。

The dataset is a corpus containing texts and query information, divided into four parts: corpus, default, qrels, queries. The corpus part contains text and title for text analysis; the default and qrels parts contain query-id, corpus-id, and score, which may be used for information retrieval tasks; the queries part contains only text, which may be used for query construction. The dataset includes training set, test set, and validation set.