Yi-Chen Li

ADM-v2: Pursuing Full-Horizon Roll-out in Dynamics Models for Offline Policy Learning and Evaluation

EMFuse: Energy-based Model Fusion for Decision Making

Provably Efficient Policy-Reward Co-Pretraining for Adversarial Imitation Learning

REAR: Scalable Test-time Preference Realignment through Reward Decomposition

Q-Adapter: Customizing Pre-trained LLMs to New Preferences with Forgetting Mitigation

Any-step Dynamics Model Improves Future Predictions for Online and Offline Reinforcement Learning

Multi-Agent Imitation by Learning and Sampling from Factorized Soft Q-Function

Controlling Large Language Model with Latent Action

Learning to Reuse Policies in State Evolvable Environments

Boosting Offline Multi-Objective Reinforcement Learning via Preference Conditioned Diffusion Models