Active membership inference attacks on federated vision models keep high success rates under LDP when the privacy budget is large, and the noise that would stop them destroys model utility.
Exploring homomorphic encryption and differential privacy techniques towards secure federated learning paradigm
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Theoretically Unmasking Inference Attacks Against LDP-Protected Clients in Federated Vision Models
Active membership inference attacks on federated vision models keep high success rates under LDP when the privacy budget is large, and the noise that would stop them destroys model utility.