ToxicityPrompts/PolyGuardMix
收藏Hugging Face2025-06-23 更新2025-04-12 收录
下载链接:
https://hf-mirror.com/datasets/ToxicityPrompts/PolyGuardMix
下载链接
链接失效反馈官方服务:
资源简介:
PolyGuard是一个用于保护大型语言模型生成内容的多语言安全模型,以及相应的训练和评估数据集。该数据集包含17种语言的样本,旨在提升多语言大型语言模型的全局用户安全性。
PolyGuard is a multilingual safety model for safeguarding Large Language Model generations, along with the corresponding training and evaluation datasets. The dataset contains samples across 17 languages, aiming to enhance the safety of multilingual Large Language Models for global users.
提供机构:
ToxicityPrompts



