vidore/vidore_v3_hr_mteb_format
收藏Hugging Face2025-11-05 更新2025-11-15 收录
下载链接：
https://hf-mirror.com/datasets/vidore/vidore_v3_hr_mteb_format
下载链接
链接失效反馈官方服务：
资源简介：
Vidore3HrRetrieval是一个多语言数据集，用于评估文本嵌入模型的检索任务。它包含德语、英语、法语、意大利语、葡萄牙语和西班牙语等多种语言的数据，并源自Vidore/vidore_v3_hr数据集。数据集支持视觉文档检索、图像到文本和文本到图像等任务。数据集使用CC BY 4.0许可证，并提供了使用mteb库评估模型的指南。
Vidore3HrRetrieval is a multilingual dataset used for evaluating text embedding models on retrieval tasks. It includes data in German, English, French, Italian, Portuguese, and Spanish, derived from the Vidore/vidore_v3_hr dataset. The dataset supports tasks such as visual-document retrieval, image-to-text, and text-to-image. It is licensed under CC BY 4.0 and provides guidelines for evaluating models using the mteb library.
提供机构：
vidore
