adversarial-10
Backdoor Defense for Data-Free Distillation with Poisoned Teachers
3260 papers • 126 benchmarks • 313 datasets
Defend against backdoor attack from poisoned teachers.
(Image credit: Papersgraph)
Benchmarks
These leaderboards are used to track progress in backdoor-defense-for-data-free-distillation-with-poisoned-teachers-10
Trend
Dataset
Best Model
Actions
No benchmarks available.
Libraries
iUse these libraries to find backdoor-defense-for-data-free-distillation-with-poisoned-teachers-10 models and implementations
Datasets
No datasets available.
Subtasks
No subtasks available.
Most implemented papers
Revisiting Data-Free Knowledge Distillation with Poisoned Teachers
Junyuan Hong, Jiayu Zhou, L. Lyu, R. Jia, Yi Zeng, Shuyang Yu•Sat Jun 03 2023
This work proposes Anti-Backdoor Data-Free KD (ABD), the first plug-in defensive method for data-free KD methods to mitigate the chance of potential backdoors being transferred and empirically evaluates the effectiveness of the proposed ABD in diminishing transferred backdoor knowledge while maintaining compatible downstream performances as the vanilla KD.
16