PrivacyXray: Detecting Privacy Breaches in LLMs through Semantic Consistency and Probability Certainty

Large Language Models LLMs are widely used in sensitive domains, including healthcare, finance, and legal services, raising concerns about potential private information leaks during inference. Privacy extraction attacks, such as jailbreaking, expose vulnerabilities in LLMs by crafting inputs that...

Insecure Pseudo-Random Number Generator

generate-password uses insecure pseudo-random number generator to generate a random number used in password generation. This results in biasness in the probability distribution of generated characters when the character count does not divide 256...