Sparse Mixture-of-Experts Reward Models Learn Interpretable and Specialized Experts for Personalized Preference Modeling 文章

ArXiv CS.CL2026-06-04NEWSen作者: Yifan Wang, Jinyi Mu, Mayank Jobanputra, Yu Wang, Ji-Ung Lee, Soyoung Oh, Isabel Valera, Vera Demberg
查看原文 →
关系图谱

Sparse Mixture-of-Experts Reward Models Learn Interpretable and Specialized Experts for Personalized Preference Modeling · 相关技术

相关技术

reinforcement learning from human feedbacksparse Mixture-of-Experts (MoE) reward modellarge language models