sail-sg/oat · AI Repo StatsSAsail-sg/oat
🌾 OAT: A research-friendly framework for LLM online alignment, including reinforcement learning, preference learning, etc.
LLM & foundation modelsPython Apache-2.0 64created 2024-10-15 #alignment#distributed-rl#distributed-training#dpo#dueling-bandits#grpo#llm#llm-aligment#llm-exploration#online-alignment#online-rl#ppo#r1-zero#reasoning#rlhf#thompson-sampling