cl-nagoya/ruri-v3-dataset-reranker

这是一个用于文本排名和文本检索任务的数据集，包含了经过reranker模型评分筛选后的正负样本。数据集特征包括样本标签、评分和来源信息。通过筛选高评分的正样本和低评分的负样本，该数据集能够提高模型性能。

This is a dataset for text ranking and text retrieval tasks, containing positive and negative samples filtered by reranker model scores. The dataset features include sample labels, scores, and source information. By filtering high-scoring positive samples and low-scoring negative samples, this dataset can improve model performance.