EasyDeepLearn

Describe the full RLHF pipeline in three stages.

hard

Answer

  • (1) SFT: fine-tune base LM on instruction data.
  • (2) Reward Model: collect preference pairs (prompt, chosen, rejected) from human annotators; train an RM (typically the SFT model with a scalar head) to predict which of two completions humans prefer via Bradley-Terry / log-sigmoid loss.
  • (3) RL fine-tune: use PPO to maximize the RM's score with a KL penalty against the SFT reference to prevent reward hacking.
  • Output: a preference-aligned model.
  • Introduced in InstructGPT (Ouyang 2022).
Check yourself — multiple choice
  • One-step SFT
  • SFT → train Reward Model on preference pairs → PPO with KL to SFT ref → preference-aligned model
  • RLHF has no SFT step
  • Only fine-tune with cross-entropy

RLHF: SFT → RM (preference pairs) → PPO with KL against SFT.

#alignment#rlhf

Practise LLMs & GenAI

214 interview questions in this topic.

Related questions