seonjeongh/ODQA_colbert_top5_100words

该数据集包含三个子集：nq、triviaqa和hotpotqa，每个子集包含多个字段，如id、question、answers、top5等。top5字段是一个列表，包含rank、docID、colbertscore、has_answer和contents等子字段。数据集的主要特征包括问题的回答情况（top1_answerable和top5_answerable）以及每个问题的前五个相关文档的评分和内容。数据集的总下载大小为60329738字节，总存储大小为104904545字节。

This dataset includes three subsets: nq, triviaqa, and hotpotqa, each containing multiple fields such as id, question, answers, top5, etc. The top5 field is a list that includes subfields like rank, docID, colbertscore, has_answer, and contents. The main features of the dataset include the answerability of questions (top1_answerable and top5_answerable) and the scores and contents of the top five related documents for each question. The total download size of the dataset is 60329738 bytes, and the total storage size is 104904545 bytes.