Eurolingua/truthfulqax
收藏Hugging Face2025-05-20 更新2025-07-05 收录
下载链接：
https://hf-mirror.com/datasets/Eurolingua/truthfulqax
下载链接
链接失效反馈官方服务：
资源简介：
EU-20-TruthfulQA是一个多语言数据集，用于评估大型语言模型在回答问题时的真实性。数据集包含多种欧洲语言，规模在10K到100K之间。它通过提出一些人类可能会因错误信念或误解而给出错误答案的问题，来测试模型是否能够避免生成从模仿人类文本中学习到的错误答案。
EU-20-TruthfulQA is a multilingual dataset designed to evaluate the truthfulness of large language models in answering questions. The dataset includes various European languages and is in the size range of 10K to 100K. It tests whether models can avoid generating false answers learned from imitating human texts by crafting questions that some humans might answer falsely due to false beliefs or misconceptions.
提供机构：
Eurolingua
