mlfoundations-dev/addition_pdf_flow

这是一个包含网页提取的问题和回答的数据集，适用于问答系统的研究和开发。数据集中的每个样本都包含了网页的URL、文件名、成功提取状态等元数据，以及从网页中提取的问题和回答文本。数据集分为训练集，可通过配置文件指定训练数据的位置。

This dataset consists of questions and answers extracted from web pages, suitable for research and development of question-answering systems. Each sample in the dataset includes metadata such as the URL of the web page, filename, extraction success status, and the text of the extracted questions and answers. The dataset is split into a training set, and the location of the training data can be specified through a configuration file.