AI RESEARCH
Chaining 2-FWL GNNs for Combinatorial Graph Alignment
arXiv CS.LG
•
ArXi:2510.03086v2 Announce Type: replace For the combinatorial graph alignment problem (GAP) -- finding the node correspondence that maximizes the number of common edges (nce) between two unlabeled graphs -- properly initialized FAQ remains a strong classical baseline, while existing GNN approaches struggle in the purely structural setting. We