mehuldamani/hotpot_qa_for_multi

这是一个包含问题和答案的数据集，每个问题都有相应的类型和难度等级。数据集中的答案可能包含来自上下文的证据，这些证据标有相应的句子ID和标题。上下文字段包含一系列句子和标题。此外，数据集还记录了问题的来源，是否有 golden 答案被移除，以及移除的标题。数据集分为训练集和测试集。

This is a dataset containing questions and answers, with each question having a corresponding type and difficulty level. The answers in the dataset may contain evidence from the context, which are marked with sentence IDs and titles. The context field includes a series of sentences and a title. Additionally, the dataset records the source of the questions, whether a golden answer has been removed, and the titles that were removed. The dataset is divided into a training set and a test set.