4 papers
Bleeding Pathways: Vanishing Discriminability in LLM Hidden States Fuels Jailbreak Attacks
Yingjie Zhang, Tong Liu, Zhe Zhao +2
LLMs remain vulnerable to jailbreak attacks that exploit adversarial prompts to circumvent safety measures. Current safety fine-tuning approaches face two critical limitations. Fir…
The Art of Hide and Seek: Making Pickle-Based Model Supply Chain Poisoning Stealthy Again
Tong Liu, Guozhu Meng, Peng Zhou +3
Pickle deserialization vulnerabilities have persisted throughout Python's history, remaining widely recognized yet unresolved. Due to its ability to transparently save and restore…
Understanding and Enhancing the Transferability of Jailbreaking Attacks
Runqi Lin, Bo Han, Fengwang Li +1
Jailbreaking attacks can effectively manipulate open-source large language models (LLMs) to produce harmful responses. However, these attacks exhibit limited transferability, faili…
Demystifying RCE Vulnerabilities in LLM-Integrated Apps
Tong Liu, Zizhuang Deng, Guozhu Meng +2
LLMs show promise in transforming software development, with a growing interest in integrating them into more intelligent apps. Frameworks like LangChain aid LLM-integrated app dev…