Bilingual graduate textbook on LLM post-training: SFT, RLHF, DPO, GRPO, RLVR, reasoning, agents, systems, evaluation, and safety.
machine-learning latex reinforcement-learning deep-learning alignment chinese textbook ai-safety reasoning post-training sft dpo large-language-models llm
8000
rlhf llm-training agentic-ai grpo rlvr agentic-rl
-
Updated
Aug 1, 2026 - TeX