LAReQA
收藏arXiv2025-09-30 收录
下载链接:
https://github.com/google-research-datasets/lareqa
下载链接
链接失效反馈官方服务:
资源简介:
该数据集是一个包含11种语言的跨语言答案检索数据集,它由两个检索子集组成:XQuAD-R和MLQA-R。其中,XQuAD-R是通过将SQuAD v1.1开发集中的240段文字翻译成10种语言,并将它们转换为检索任务而构建的。该任务的目标是跨语言答案检索。
This dataset is a cross-lingual answer retrieval dataset covering 11 languages, which comprises two retrieval subsets: XQuAD-R and MLQA-R. Specifically, XQuAD-R is constructed by translating 240 passages from the SQuAD v1.1 development set into 10 languages and converting them into retrieval tasks. The goal of this task is cross-lingual answer retrieval.



