On the Dangers of Poisoned LLMs in Security Automation

This paper investigates some of the risks introduced by "LLM poisoning," the intentional or unintentional introduction of malicious or biased data during model training. We demonstrate how a seemingly improved LLM, fine-tuned on a limited dataset, can introduce significant bias, to the extent tha...

Phantom Subgroup Poisoning: Stealth Attacks on Federated Recommender Systems

Federated recommender systems FedRec have emerged as a promising solution for delivering personalized recommendations while safeguarding user privacy. However, recent studies have demonstrated their vulnerability to poisoning attacks. Existing attacks typically target the entire user group, which...