illuin-conteb/covid-qa

ConTEB数据集是ConTEB（上下文感知文本嵌入基准）的一部分，旨在评估上下文嵌入模型的能力。该数据集关注健康护理主题，特别是源自关于COVID-19疫情的文章。数据集基于COVID-QA数据集构建，包含经过精心挑选的原始文档、从这些文档中提取的片段以及查询。共有115个文档，3351个片段和1111个查询，平均每个文档有153.9个标记。

ConTEB dataset is part of ConTEB (Context-aware Text Embedding Benchmark), designed to evaluate the capabilities of contextual embedding models. The dataset focuses on the theme of Healthcare, particularly from articles about the COVID-19 pandemic. It is built upon the COVID-QA dataset and includes a curated set of original documents, chunks derived from them, and queries. There are 115 documents, 3351 chunks, and 1111 queries, with an average of 153.9 tokens per document.