Abstract: Backdoor attacks threaten federated learning (FL) models, where malicious participants embed hidden triggers into local models during training. These triggers can compromise crucial ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback