SPA is a federated learning backdoor attack that aligns trigger features with target class features in the model's latent space, achieving high, persistent attack success rates while evading detection-based defenses.
Federated Learning for Mobile Keyboard Prediction,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CR 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
SPA: Towards More Stealth and Persistent Backdoor Attacks in Federated Learning
SPA is a federated learning backdoor attack that aligns trigger features with target class features in the model's latent space, achieving high, persistent attack success rates while evading detection-based defenses.