RaguTeam/ragu_benchmarks
收藏Hugging Face2025-10-18 更新2025-11-01 收录
下载链接：
https://hf-mirror.com/datasets/RaguTeam/ragu_benchmarks
下载链接
链接失效反馈官方服务：
资源简介：
MultiQ数据集：一个俄语问答和多云文档信息检索任务的数据集，包含169个问题，每个问题都有一个正确答案和一组相关的维基百科文章作为上下文。NSU QA数据集：一个基于NSU官方材料的问答和多云文档检索基准，包含207个问题，每个问题都有一个精确答案和相关的文档ID列表。CheGeKa数据集：一个为了评估模型回答复杂智力问题能力的数据集，包含104个问题，每个问题都配有一个精确答案和相关的维基百科文档。
MultiQ Dataset: A Russian-language question answering and multi-document information retrieval dataset containing 169 questions, each with a correct answer and a set of relevant Wikipedia articles serving as context. NSU QA Dataset: A question answering and multi-document retrieval benchmark based on NSU official materials, containing 207 questions, each paired with a precise answer and a list of relevant document IDs. CheGeKa Dataset: A benchmark dataset designed to evaluate models ability to answer complex intellectual questions, containing 104 questions, each with a precise answer and references to related Wikipedia documents.
提供机构：
RaguTeam
