AproMFL enables mixed-modality federated learning with heterogeneous tasks by adaptively building and aggregating prototypes without unified labels.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Adaptive Prototype Knowledge Transfer for Federated Learning with Mixed Modalities and Heterogeneous Tasks
AproMFL enables mixed-modality federated learning with heterogeneous tasks by adaptively building and aggregating prototypes without unified labels.