Robust Knowledge Distillation Based on Feature Variance Against Backdoored Teacher Model

Open in new window