作者位次 #9/13ICML 2026Poster· Deep Learning / Large Language Models
GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimization
Shih-Yang Liu · Xin Dong · Ximing Lu · Shizhe Diao · Peter Belcak · Mingjie Liu · Min-Hung Chen · Hongxu (Danny) Yin · Yu-Chiang Wang · Kwang-Ting Cheng · Yejin Choi · Jan Kautz · Pavlo Molchanov
强化学习与后训练OpenReview