Nayana-cognitivelab/nayana-beir-eval-multilang_v7_crosslingual_only
收藏Hugging Face2025-10-19 更新2025-10-25 收录
下载链接：
https://hf-mirror.com/datasets/Nayana-cognitivelab/nayana-beir-eval-multilang_v7_crosslingual_only
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含三个部分：corpus、qrels和queries。corpus部分包含文档标识、图像和文档ID；qrels部分包含查询ID、文档ID、是否可回答的标记、答案和分数；queries部分包含查询ID、查询内容、语言类型、GPT-4O推理结果和答案。数据集提供了测试分割的数据，可用于测试和评估相关模型。
The dataset consists of three parts: corpus, qrels, and queries. The corpus part includes document identification, images, and document IDs; the qrels part includes query IDs, document IDs, answerability marks, answers, and scores; the queries part includes query IDs, query content, language type, GPT-4O reasoning results, and answers. The dataset provides test split data for testing and evaluating related models.
提供机构：
Nayana-cognitivelab
