AI RESEARCH
SCL-GNN: Towards Generalizable Graph Neural Networks via Spurious Correlation Learning
arXiv CS.LG
•
ArXi:2603.08270v1 Announce Type: new Graph Neural Networks (GNNs) have nstrated remarkable success across diverse tasks. However, their generalization capability is often hindered by spurious correlations between node features and labels in the graph. Our analysis reveals that GNNs tend to exploit imperceptible statistical correlations in