mteb/NeuCLIR2022Retrieval_fas_top_250_only_w_correct-v2

该数据集包含三个不同的配置：文本语料（corpus）、默认配置（default）和查询配置（queries）。文本语料配置包含文档的唯一标识符和文本内容；默认配置包含查询的唯一标识符、对应语料的唯一标识符和分数；查询配置包含查询的唯一标识符和查询文本。每个配置都有测试分片，分别包含不同数量的示例和字节数。总下载大小和总数据集大小也提供了数据集规模的信息。The dataset consists of three different configurations: text corpus (corpus), default configuration (default), and query configuration (queries). The text corpus configuration includes a unique identifier and text content for documents; the default configuration includes a unique identifier for the query, a unique identifier for the corresponding corpus, and a score; the query configuration includes a unique identifier and the query text. Each configuration has a test split, which contains a different number of examples and byte sizes. The total download size and total dataset size also provide information about the scale of the dataset.