STAGE: Tackling Semantic Drift in Multimodal Federated Graph Learning

TL;DR

STAGE is a framework for multimodal federated graph learning that enhances semantic calibration across clients and reduces inconsistency amplification, leading to improved performance and communication efficiency.

Contribution

It introduces a novel protocol-first approach that translates heterogeneous multimodal features into comparable representations for federated graph learning.

Findings

Achieves state-of-the-art results on 8 multimodal graph datasets.

Reduces communication payload during training.

Improves cross-client semantic calibration.

Abstract

Federated graph learning (FGL) enables collaborative training on graph data across multiple clients. As graph data increasingly contain multimodal node attributes such as text and images, multimodal federated graph learning (MM-FGL) has become an important yet substantially harder setting. The key challenge is that clients from different modality domains may not share a common semantic space: even for the same concept, their local encoders can produce inconsistent representations before collaboration begins. This makes direct parameter coordination unreliable and further causes two downstream problems: forcing heterogeneous client representations into a naively shared semantic space may create false semantic agreement, and graph message passing may amplify residual inconsistency across neighborhoods. To address this issue, we propose \textbf{STAGE}, a protocol-first framework for MM-FGL. Instead of relying on direct parameter averaging, STAGE builds a shared semantic space that first translates heterogeneous multimodal features into comparable representations and then regulates how these representations propagate over local graph structures. In this way, STAGE not only improves cross-client semantic calibration, but also reduces the risk of inconsistency amplification during graph learning. Extensive experiments on 8 multimodal-attributed graphs across 5 graph-centric and modality-centric tasks show that STAGE consistently achieves state-of-the-art performance while reducing per-round communication payload.