3 papers
cs.CL2025
SafeTy Reasoning Elicitation Alignment for Multi-Turn Dialogues
Martin Kuo, Jianyi Zhang, Aolin Ding +12
Malicious attackers can exploit large language models (LLMs) by engaging them in multi-turn dialogues to achieve harmful objectives, posing significant safety risks to society. To…
cs.LG2025
FedProphet: Memory-Efficient Federated Adversarial Training via Robust and Consistent Cascade Learning
Minxue Tang, Yitu Wang, Jingyang Zhang +5
Federated Adversarial Training (FAT) can supplement robustness against adversarial examples to Federated Learning (FL), promoting a meaningful step toward trustworthy AI. However,…
cs.CL2025
Proactive Privacy Amnesia for Large Language Models: Safeguarding PII with Negligible Impact on Model Utility
Martin Kuo, Jingyang Zhang, Jianyi Zhang +9
With the rise of large language models (LLMs), increasing research has recognized their risk of leaking personally identifiable information (PII) under malicious attacks. Although…