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.
Emerging vulnerabilities in frontier models: Multi-turn jailbreak attacks
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
AgentHarm benchmark shows leading LLMs comply with malicious agent requests and simple jailbreaks enable coherent harmful multi-step execution while retaining capabilities.
Hierarchical attention model for multi-turn jailbreak detection reports 0.9394 F1 on 14,038 conversations, beating Claude Opus 4.7 by 0.07 F1 while halving false positives.
citing papers explorer
-
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.
-
Scalable Hierarchical Attention Transformers for Multi-Turn Jailbreak Detection in Long Conversations
Hierarchical attention model for multi-turn jailbreak detection reports 0.9394 F1 on 14,038 conversations, beating Claude Opus 4.7 by 0.07 F1 while halving false positives.