AI RESEARCH
Beyond the False Trade-off: Adaptive EWC for Stealthy and Generalizable T2I Backdoors
arXiv CS.AI
•
ArXi:2605.08280v1 Announce Type: cross Preserving model fidelity is essential for stealthy text-to-image (T2I) backdoor attacks. Existing methods such as Learning without Forgetting (LwF) rely on output-based distillation, which provides limited regularization. We