Shentao Yang

model-based reinforcement learning (2) offline reinforcement learning (2) preference learning (1) text generation (1) policy learning (1) continuous control (1) stationary distribution (1) lower bound (1) language model (1) dynamics model (1) language model fine-tuning (1) dynamic model (1) expected return (1) distribution mismatch (1) token-level guidance (1) model-based policy (1) offline model-based reinforcement learning (1) model-policy learning (1) sequence-level preference (1) policy optimization (1)