mteb/MIRACLRetrieval_de_top_250_only_w_correct-v2

该数据集包含三个部分：corpus、default和queries。corpus部分包含带有标题和文本的语料库，default部分包含查询、语料库ID和分数信息，queries部分则包含查询文本。每个部分都有测试集split，并且提供了每个split的字节大小和示例数量。

The dataset consists of three parts: corpus, default, and queries. The corpus section includes a collection of texts with titles, the default section contains queries, corpus IDs, and score information, and the queries section includes query texts. Each part has a test set split, and the byte size and number of examples for each split are provided.