Optimizing Against Safety Representations: Activation-Guided Adversarial Suffixes and the Geometry of Refusal
Behavioral alignment in large language models often masks fragile internal safety representations. Recent work suggests that refusal behavior is mediated by low-dimensional directions in activation space. This raises questions about how such representations are structured, localized, and accessed by optimization. We study adversarial suffix attacks as a probe of representational alignment. We introduce Activation-Guided GCG, which replaces output-based objectives with losses that directly target a model's internal refusal direction. Across several objective variants, we find that suppressing refusal globally across all layers and positions is more effective than targeting a single layer-p...
| Authors | Ege Çakar, Hannah Guan, Kayden Kehe |
|---|---|
| Primary category | cs.LG |
| Published | 9 July 2026 |
| Last updated at source | 9 July 2026 |
Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.