AI RESEARCH
PEANUT: Perturbations by Eigenvalue Alignment for Attacking GNNs Under Topology-Driven Message Passing
arXiv CS.LG
•
ArXi:2603.26136v1 Announce Type: new Graph Neural Networks (GNNs) have achieved remarkable performance on tasks involving relational data. However, small perturbations to the graph structure can significantly alter GNN outputs, raising concerns about their robustness in real-world deployments.