首页 > AI前沿 > The Role of Fine-grained Harm Signals in LLM Safety

The Role of Fine-grained Harm Signals in LLM Safety

arXiv自然语言 2026-09-17 03:40 4 阅读 查看原文

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 harmfulness representation, yielding a category residual that is orthogonal to general harmfulness at every layer.

Using activation steering with category residuals across 11 risk categories in 3 instruction-tuned LLMs, we find that whether category residuals encode harmfulness varies across categories, and that this category-wise pattern is similar across models.

Whether category residuals induce refusal also varies across categories, but this category-wise pattern is more model-dependent.

We also find that category residuals increase LLMs' downstream internal alignment with shared general harmfulness representation.

Together, these findings demonstrate that more fine-grained category residuals should also be considered beyond shared general harmfulness representation to fully understand LLM safety.

More broadly, our findings show that even a direction orthogonal to a concept at one layer can contribute to the concept's downstream amplification.