Reward Modeling - Tag - lilfry's library

Reward Modeling - Tag - lilfry's libraryhttps://lilfry09.github.io/en/tags/reward-modeling/Reward Modeling - Tag - lilfry's libraryHugo -- gohugo.ioenlilfry@sjtu.edu.cn (lilfry)lilfry@sjtu.edu.cn (lilfry)All rights reserved.Wed, 08 Apr 2026 14:20:00 +0800High frequency interview: Bradley-Terry vs Plackett-Luce, what is the difference between reward modeling?https://lilfry09.github.io/en/ai/bt-vs-pl-ranking-in-alignment/Wed, 08 Apr 2026 14:20:00 +0800fryhttps://lilfry09.github.io/en/ai/bt-vs-pl-ranking-in-alignment/In a word: BT is a special case of PL when the number of candidates is N=2. BT is lighter, more stable, and more suitable for existing RLHF/DPO; PL has more complete information, but the labeling and training costs are significantly higher.