SKIML-ICL/CRRAG_nq_more-train-with-predicted-answer-sentences

该数据集包含了问题（question）、答案（answers）、上下文（ctxs）、是否包含答案（hasanswer）、答案可回答性（answerable）等字段。上下文字段中包含了是否有答案、自然语言推理类型、段落ID、排名、得分、文本内容和标题等信息。此外，数据集还包含了生成答案提示、否定句提示、对抗句提示等相关字段。数据集分为测试集，测试集大小为4286676字节，共有464个示例。数据集下载大小为620703字节。

The dataset includes fields such as questions (question), answers (answers), contexts (ctxs), whether there is an answer (hasanswer), answerability (answerable), etc. The context field contains information such as whether there is an answer, natural language inference type, passage ID, ranking, score, text content, and title. In addition, the dataset also includes related fields such as prompts for answer generation, negated sentence prompts, adversarial sentence prompts, etc. The dataset is split into a test set, which is 4286676 bytes in size and contains 464 examples. The download size of the dataset is 620703 bytes.