ORPO: Monolithic Preference Optimization without Reference Model2024-01-010 PRE AI DOI:10.18653/v1/2024.emnlp-main.626原文链接原文求助分享收藏摘要 En