3 papers
cs.CR2026
Targeted Counterfactual Fingerprinting for Black-Box LLM Ownership Verification
Yutong Wu, Xiaofan Bai, Shixin Li +10
Large language models (LLMs) are high-value assets that can be derived through redeployment, fine-tuning, quantization, or further alignment. Because deployed LLMs are commonly exp…
cs.CR2026
SoK: Robustness in Large Language Models against Jailbreak Attacks
Feiyue Xu, Hongsheng Hu, Chaoxiang He +9
Large Language Models (LLMs) have achieved remarkable success but remain highly susceptible to jailbreak attacks, in which adversarial prompts coerce models into generating harmful…
cs.CR2026
Sirens' Whisper: Inaudible Near-Ultrasonic Jailbreaks of Speech-Driven LLMs
Zijian Ling, Pingyi Hu, Xiuyong Gao +6
Speech-driven large language models (LLMs) are increasingly accessed through speech interfaces, introducing new security risks via open acoustic channels. We present Sirens' Whispe…