Madjakul/HALvest-Contrastive-R

该数据集是一个文本数据集，包含了查询及其相关属性，如查询ID、查询内容、年份、所属领域、从属关系和作者ID等。同时，对于每个查询，还提供了与之相关的正例和负例信息。数据集分为多个配置版本，每个版本包含的训练数据量不同。

The dataset is a text dataset containing queries and their related attributes such as query ID, query content, year, domain, affiliations, and author IDs. For each query, it also provides positive and negative example information. The dataset is divided into multiple configuration versions, each containing a different amount of training data.