1
MultiTurnPSB: Evaluating Multi-Turn Jailbreak Attacks an dClassifier-Based Defenses for Medical AI Safety
医疗AI面临多轮越狱攻击威胁？本文构建新基准评估分类器防御，守护安全底线。
arXiv:2606.02630v1 Announce Type: cross Abstract: Patient-facing medical chatbots are commonly evaluated on single-turn prompts, yet real users push b…