作者位次 #4/4ACL 2026Poster· Natural Language Processing
Reward Alignment Optimization: A Direct Point-wise Alignment Approach
Zelin Li · Jia Leng · Dawei Song · Yangen Hu
摘要
Direct Alignment Algorithms (DAAs) such as DPO simplify RLHF by optimizing policies directly from preference pairs. However, the Bradley–Terry probability-gap objective can induce likelihood displacement and, under weak KL constraints, may even reduce the probability of preferred responses, while...