Backdoor Evaluation
Cross-source consensus on Backdoor Evaluation from 1 sources and 4 claims.
1 sources · 4 claims
Benefits
Comparisons
Evidence quality
Highlighted claims
- Lower backdoor accuracy indicates better removal of backdoor behavior. — Asynchronous Federated Unlearning with Invariance Calibration for Medical Imaging
- Unlearning completeness was evaluated with a backdoor proxy where the target client poisons part of its local data. — Asynchronous Federated Unlearning with Invariance Calibration for Medical Imaging
- Projected-GA and FedRecovery initially lowered backdoor accuracy but reverted strongly after continued training. — Asynchronous Federated Unlearning with Invariance Calibration for Medical Imaging
- AFU-IC's backdoor accuracy closely tracked retraining over 10 post-learning rounds on all three datasets. — Asynchronous Federated Unlearning with Invariance Calibration for Medical Imaging