offline online alignment

preference sampling (1) offline alignment (1) large language model (1) llm alignment (1) language model alignment (1) preference alignment (1) direct preference optimization (1) data transformation (1) offline reinforcement learning (1) reward modeling (1)