The Role of Fine-grained Harm Signals in LLM Safety

Excerpt

arXiv:2609.19366v1 Announce Type: new Abstract: Prior work has shown that internal harmfulness representations in large language models vary across risk categories, while sharing a common general harm representation component. This raises a question about the role of the category-specific component beyond general harm representation in LLM safety. To answer this question, we isolate the category-specific component by removing shared general harmfulness representation from each categorical harmfu