mzhaoshuai/llama3-ultrafeedback-bertscore-bart-large-mnli
收藏Hugging Face2025-10-16 更新2025-10-25 收录
下载链接：
https://hf-mirror.com/datasets/mzhaoshuai/llama3-ultrafeedback-bertscore-bart-large-mnli
下载链接
链接失效反馈官方服务：
资源简介：
RefAlign是一个语言模型对齐数据集，用于从参考答案中学习，无需二元人类偏好数据即可实现多功能的语言模型对齐。该数据集通过BERTScore选择接受和拒绝的回应。
RefAlign is a language model alignment dataset used for learning from reference answers to achieve versatile language model alignment without binary human preference data. The dataset selects accepted and rejected responses using BERTScore.
提供机构：
mzhaoshuai
