Reinforcement Learning from LLM Feedback to Counteract Goal Misgeneralization

Published in arXiv preprint, 2024