PolicyAlign: Direct Policy-Based Safety Alignment for Large Language Models

arXiv:2606.25442v1 Announce Type: new Abstract: Safety alignment of large language models (LLMs) typically depends on high-quality supervision data, such as safe demonstrations or preference pairs. However, in real-world deployment, emerging safety requirements are often specified as natural-language policies, while corresponding supervision data may be costly, delayed, or unavailable. This creates a mismatch between rapidly evolving safety policies and conventional data-driven alignment methods...