Bridging Computational Social Science and Deep Learning: Cultural Dissemination-Inspired Graph Neural Networks

Researchers introduce AxelGNN, a new Graph Neural Network architecture inspired by cultural dissemination theory that addresses key limitations of existing GNNs including oversmoothing and poor handling of heterogeneous relationships. The model demonstrates superior performance in node classification and influence estimation while maintaining computational efficiency across both homophilic and heterophilic graphs.

• →AxelGNN introduces similarity-gated interactions that adaptively promote convergence or divergence based on feature similarity.
• →The architecture uses segment-wise feature copying for fine-grained aggregation instead of monolithic vector processing.
• →Global polarization maintains multiple distinct representation clusters to prevent oversmoothing in deep architectures.
• →The model handles both homophilic and heterophilic graphs within a single architecture without specialized model selection.
• →AxelGNN achieves competitive or superior performance in node classification and influence estimation tasks.