Axioms for AI Alignment from Human Feedback

Open in new window