passdrill
AI & LLM Engineering · Fine-tuning & Model Customisation · Card 008/010 medium

Gathering large volumes of high-quality human preference labels for RLHF is expensive, so Lee et al. studied replacing the human-labeled preference data used to train a reward model with preference labels generated by an off-the-shelf large language model instead (RLAIF), then compared the resulting policies across summarization and dialogue tasks using human evaluators. What did the study actually find?

  1. RLAIF only matched human-evaluated RLHF performance when the LLM generating the preference labels was substantially larger and more capable than the policy model being trained, and performed far worse whenever the labeler was the same size as the policy
  2. RLAIF consistently and substantially underperformed both RLHF and a plain supervised-fine-tuned baseline across every task tested, showing that AI-generated preference labels cannot substitute for human-labeled ones
  3. RLAIF achieved performance comparable to RLHF as judged by human evaluators across the tasks tested, including cases where the labeling model was the same size as the policy being trained or even the exact same checkpoint, and RLAIF outperformed the supervised-fine-tuned baseline in these comparisons
  4. RLAIF eliminated the need for a reward model and the reinforcement-learning optimization step entirely in every configuration tested, since the AI-generated labels were used to edit the policy's weights directly without any RL algorithm
Next card → Shuffle