FedGSA aggregates differentially private federated LoRA updates as subspaces on a Grassmann manifold, improving GLUE and E2E NLG accuracy over Euclidean averaging baselines.
Improvedgradientinversionattacksand defenses in federated learning.IEEE Transactions on Big Data, 10(6): 839–850
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
FedGSA: Geometry-Consistent Subspace Aggregation for Differentially Private Federated LoRA
FedGSA aggregates differentially private federated LoRA updates as subspaces on a Grassmann manifold, improving GLUE and E2E NLG accuracy over Euclidean averaging baselines.