qualifire/our-financebench-hallucination-dataset

该数据集包含知识、问题、答案、标签和全部文本数据五个字段，分为通过（PASS）和未通过（FAIL）两部分，每部分大约包含7275和7271个样本。具体内容可能是关于某种问答或判断任务的数据集，但详细描述未在README中给出。The dataset includes fields for knowledge, question, answer, label, and all text data, split into two parts: PASS with 7275 examples and FAIL with 7271 examples. It might be a question answering or judgment task dataset, but the detailed description is not provided in the README.