FedMPO recovers missing modalities via topology-aware generation, filters noisy recoveries with missing-aware routing, and uses reliability-aware aggregation to achieve up to 5.65% gains over baselines in high-missing and non-IID federated graph settings.
arXiv preprint arXiv:2602.05576 , year =
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
citation-role summary
dataset 1
citation-polarity summary
years
2026 5verdicts
UNVERDICTED 5roles
dataset 1polarities
use dataset 1representative citing papers
RoleMAG learns neighbor roles in multimodal graphs to route shared, complementary, and heterophilous signals through separate channels, improving propagation without modality interference.
SMGFM decomposes multimodal node signals on graphs into frequency bands via Chebyshev filters, routes them by topology reliability, and aligns objectives to preserve modality-specific semantics while achieving SOTA on MAG tasks.
GraphMNL applies negative learning as cross-branch guidance in multimodal graphs to mitigate semantic imbalance without propagating bias from dominant branches.
CAMPA resolves modal conflicts in decoupled multimodal GNNs via cross-modal aligned propagation and trajectory aligned aggregation, outperforming coupled and decoupled baselines on benchmarks while retaining efficiency.
citing papers explorer
-
Towards Robust Federated Multimodal Graph Learning under Modality Heterogeneity
FedMPO recovers missing modalities via topology-aware generation, filters noisy recoveries with missing-aware routing, and uses reliability-aware aggregation to achieve up to 5.65% gains over baselines in high-missing and non-IID federated graph settings.
-
RoleMAG: Learning Neighbor Roles in Multimodal Graphs
RoleMAG learns neighbor roles in multimodal graphs to route shared, complementary, and heterophilous signals through separate channels, improving propagation without modality interference.
-
SMGFM: Spectral Multimodal Graph Pretraining for Multimodal-Attributed Graphs
SMGFM decomposes multimodal node signals on graphs into frequency bands via Chebyshev filters, routes them by topology reliability, and aligns objectives to preserve modality-specific semantics while achieving SOTA on MAG tasks.
-
Multimodal Graph Negative Learning
GraphMNL applies negative learning as cross-branch guidance in multimodal graphs to mitigate semantic imbalance without propagating bias from dominant branches.
-
CAMPA: Efficient and Aligned Multimodal Graph Learning via Decoupled Propagation and Aggregation
CAMPA resolves modal conflicts in decoupled multimodal GNNs via cross-modal aligned propagation and trajectory aligned aggregation, outperforming coupled and decoupled baselines on benchmarks while retaining efficiency.