isaacus/mteb-barexam-qa

Bar Exam QA是一个用于评估信息检索模型在法律条文检索任务上的数据集，采用了MTEB格式。该数据集包括三种划分：默认划分将查询与相关条文配对，条文划分包含相关的法律条文，查询划分包含查询语句。这个数据集旨在为法律文本嵌入模型提供一个大规模、多样化、全面的基准。

Bar Exam QA is a dataset for evaluating information retrieval models on the task of legal provision retrieval, formatted in the MTEB format. The dataset includes three splits: the default split pairs queries with relevant provisions, the provisions split contains relevant legal provisions, and the queries split contains the query statements. This dataset aims to provide a large-scale, diverse, and comprehensive benchmark for legal text embedding models.