yue1045/rag-mini-wikipedia

该数据集是一个基于维基百科的小型开放域问答数据集，适用于问答和句子相似性任务。它包含两个配置：文本语料（text-corpus）和问答对（question-answer），用于支持信息检索、RAG（检索增强生成）和DPR（密集段落检索）等应用。数据集源自Kaggle的questionanswer-dataset，并通过generate.py生成了子集，规模较小（少于1K样本）。

This dataset is a small open-domain question-answering dataset based on Wikipedia, suitable for question-answering and sentence-similarity tasks. It includes two configurations: text-corpus and question-answer, designed to support applications such as information retrieval, RAG (Retrieval-Augmented Generation), and DPR (Dense Passage Retrieval). The dataset is derived from the Kaggle questionanswer-dataset, with a subset generated using generate.py, and has a small size (less than 1K samples).