hotchpotch/hpprc_emb-scores

该数据集包含多个子数据集，主要用于问答系统和信息检索任务。每个子数据集包含原始行ID、锚点文本、答案、段落ID、正负样本ID、相似度评分等特征字段。数据集涵盖了多种语言和领域，训练集的大小和示例数量也被详细列出。

This dataset contains multiple sub-datasets primarily used for question answering and information retrieval tasks. Each sub-dataset includes features such as original row ID, anchor text, answers, passage ID, positive and negative sample IDs, and similarity scores. The datasets cover multiple languages and domains, with detailed information on the size and number of examples in the training sets.