FinancialQA

该数据集包含了基于一家全球领先会计事务所关于收入确认的文件综合生成的提问，这为问答语境检索带来了重大的挑战。该文件可在以下公开链接获取：https://kpmg.com/kpmg-us/content/dam/kpmg/frv/pdf/2024/handbook-revenue-recognition-1224.pdf。数据集规模超过1000页文本，共生成了50个问题（其中30个用于训练，20个用于测试），任务内容为问答检索。

This dataset consists of questions comprehensively generated based on a revenue recognition handbook from a leading global accounting firm, which presents substantial challenges for contextual question answering retrieval. The source document is accessible through the following public link: https://kpmg.com/kpmg-us/content/dam/kpmg/frv/pdf/2024/handbook-revenue-recognition-1224.pdf. The dataset spans over 1,000 pages of text, with a total of 50 questions generated, among which 30 are allocated for training and 20 for testing. The core task of this dataset is question answering retrieval.