Nayana-cognitivelab/nayana-beir-eval-multilang_v8

该数据集包含了三个部分：文档集合、查询与文档的相关性信息、查询信息。文档集合中包含文档id和图片；相关性信息中包含查询id、文档id、是否为答案文档、答案内容和相关性分数；查询信息中包含查询id、查询语句、查询语言、GPT-4推理结果和答案。所有数据均针对测试集。

The dataset consists of three parts: a collection of documents, information on the relevance of queries to documents, and query information. The document collection includes document IDs and images; the relevance information includes query IDs, document IDs, whether they are answer documents, answer content, and relevance scores; the query information includes query IDs, query statements, query languages, GPT-4 reasoning results, and answers. All data is for the test set.