SKIML-ICL/med_retrieved_original

这是一个包含多个字段的数据集，其中包括问题ID、标准化问题、标准化答案、原始问题、原始答案、答案生成提示、答案句子以及上下文字段。上下文字段中又包含了是否有答案、NLI任务类型、段落ID、排名、得分、文本和标题等信息。此外，数据集还提供了测试集的split信息，包括示例数量和文件大小。数据集配置部分提供了测试数据文件的路径。

This dataset contains multiple fields, including question ID, normalized question, normalized answers, original question, original answers, prompt for answer generation, answer sentence, and context fields. The context fields include information such as whether there is an answer, NLI task type, paragraph ID, rank, score, text, and title. Moreover, the dataset provides split information for the test set, including the number of examples and file size. The dataset configuration part provides the path to the test data files.