oat
Public? OAT: A research-friendly framework for LLM online alignment, including preference learning, reinforcement learning, etc.
Creat:2024-10-15T13:53:45
Update:2025-03-27T08:59:26
431
Stars
3
Stars Increase
? OAT: A research-friendly framework for LLM online alignment, including preference learning, reinforcement learning, etc.