Mengdi Li

Curriculum-RLAIF: Curriculum Alignment with Reinforcement Learning from AI Feedback

Efficient and Stable Grouped RL Training for Large Language Models

Towards User-level Private Reinforcement Learning with Human Feedback

Understanding Reasoning in Chain-of-Thought from the Hopfieldian View

What Makes Your Model a Low-empathy or Warmth Person: Exploring the Origins of Personality in LLMs