sohey1024/miracl-id-seed42

该数据集包含开发集（dev）、测试集（test）和训练集（train）三个部分。每个部分都包含了query_id、doc_id、relevance和text等字段。训练集额外包含了正例段落（positive_passages）和负例段落（negative_passages），每个段落包括doc_id、title和text信息。数据集大小从566465字节到83680044字节不等，示例数量从8292到3256不等。

The dataset consists of development (dev), test (test), and training (train) sets. Each set includes fields such as query_id, doc_id, relevance, and text. The training set additionally contains positive passages (positive_passages) and negative passages (negative_passages), with each passage including doc_id, title, and text information. The size of the dataset ranges from 566465 bytes to 83680044 bytes, and the number of examples varies from 8292 to 3256.