conftrace_

refusal generation

2 papers

Explore in graph

Co-occurring keywords · papers mentioning both

vision-language model (2) continual unlearning (1) safety awareness (1) jailbreak defense (1) concept decomposition (1) prompt tuning (1) mixture of expert (1) adversarial attack (1)

Papers

SafetyReminder: Reviving Delayed Safety Awareness of Vision-Language Models to Defend Against Jailbreak Attacks AAAI 2026 Which Concepts to Forget and How to Refuse? Decomposing Concepts for Continual Unlearning in Large Vision-Language Models CVPR 2026