3 papers
cs.LG2025
A Common Pitfall of Margin-based Language Model Alignment: Gradient Entanglement
Hui Yuan, Yifan Zeng, Yue Wu +3
Reinforcement Learning from Human Feedback (RLHF) has become the predominant approach for language model (LM) alignment. At its core, RLHF uses a margin-based loss for preference o…
cs.CL2024
MaxMin-RLHF: Alignment with Diverse Human Preferences
Souradip Chakraborty, Jiahao Qiu, Hui Yuan +5
Reinforcement Learning from Human Feedback (RLHF) aligns language models to human preferences by employing a singular reward model derived from preference data. However, such an ap…
stat.ML2024
Gradient Guidance for Diffusion Models: An Optimization Perspective
Yingqing Guo, Hui Yuan, Yukang Yang +2
Diffusion models have demonstrated empirical successes in various applications and can be adapted to task-specific needs via guidance. This paper studies a form of gradient guidanc…