seongil-dn/mteb-kor-retrieval-full_naive

该数据集是一个包含查询和对应正负例文本的数据集，适用于文本相似度或检索任务。每个查询都有一组正例和负例，每个例子都有一个唯一的标识符（id），文本内容（text）和一个分数（score），负例还有一个topk_rank字段表示其在查询结果中的排名。

The dataset is a collection of queries along with their corresponding positive and negative text examples, suitable for text similarity or retrieval tasks. Each query has a set of positive and negative examples, each with a unique identifier (id), text content (text), and a score (score), while negative examples also have a topk_rank field indicating their rank in the query results.