STAGE builds a shared semantic space through feature translation and controlled graph propagation to reduce semantic drift in multimodal federated graph learning, delivering state-of-the-art results with lower communication cost.
Subgraph federated learning via spectral methods
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
citation-role summary
baseline 1
citation-polarity summary
fields
cs.LG 4years
2026 4verdicts
UNVERDICTED 4roles
baseline 1polarities
baseline 1representative citing papers
FedMGS formalizes modality-imbalanced MM-FGL as latent semantic synthesis and uses availability-aware encoding, prototype-guided synthesis, and reliability-calibrated fusion to recover missing modalities, reporting up to 17.41% gains on four tasks.
CE-FedGNN enables federated GNN training on coupled distributed graphs via infrequent aggregated representation exchange, moving-average estimation for staleness, and metric-DP, with O(1/sqrt(T)) convergence and O(T^{3/4}) communication.
PRISM proposes a topology-aware cross-modal imputation framework for client-level modality-deficient multimodal federated graph learning that improves deficient clients by 4.48% on average over baselines across six datasets.
citing papers explorer
-
STAGE: Tackling Semantic Drift in Multimodal Federated Graph Learning
STAGE builds a shared semantic space through feature translation and controlled graph propagation to reduce semantic drift in multimodal federated graph learning, delivering state-of-the-art results with lower communication cost.