AEGIS localizes sparse semantic-injecting attention heads in diffusion models and applies similarity-aware repulsion at those heads to block visual synonym jailbreaks while preserving benign generation.
Sneakyprompt: Jailbreaking text-to-image generative models
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2representative citing papers
Salami Attack chains low-risk inputs to cumulatively trigger high-risk LLM behaviors, achieving over 90% success on GPT-4o and Gemini while resisting some defenses.
citing papers explorer
-
AEGIS: A Mechanism-Guided Defense against Visual Synonym Jailbreaks in Text-to-Image Models
AEGIS localizes sparse semantic-injecting attention heads in diffusion models and applies similarity-aware repulsion at those heads to block visual synonym jailbreaks while preserving benign generation.
-
The Salami Slicing Threat: Exploiting Cumulative Risks in LLM Systems
Salami Attack chains low-risk inputs to cumulatively trigger high-risk LLM behaviors, achieving over 90% success on GPT-4o and Gemini while resisting some defenses.