EasyDeepLearn

RL for reasoning (o1, DeepSeek-R1) — the paradigm shift.

hard

Answer

  • Train LLM to generate long chain-of-thought reasoning by RL with automatic reward from verifier (math answer correct, code passes tests).
  • Model learns to think longer, retry, self-correct.
  • DeepSeek-R1: pure RL from base model (no SFT step) → emergent reasoning behaviors.
  • Uses GRPO (group-normalized rewards).
  • Test-time compute becomes a knob.
  • Marks shift from imitation-based to verification-based post-training.
Check yourself — multiple choice
  • Random
  • RL with automatic verifier reward (math correct / code passes) → model learns to think longer, retry, self-correct; DeepSeek-R1 pure RL from base; GRPO; shift from imitation to verification
  • Same as RLHF
  • Not real

RL reasoning: verifier reward + long CoT; DeepSeek-R1 GRPO; verification paradigm.

#llm#alignment

Practise Reinforcement Learning

214 interview questions in this topic.

Related questions