AI RESEARCH
Operational Feature Fingerprints of Graph Datasets via a White-Box Signal-Subspace Probe
arXiv CS.LG
•
ArXi:2604.22676v1 Announce Type: new Graph neural networks achieve strong node-classification accuracy, but their learned message passing entangles ego attributes, neighborhood smoothing, high-pass graph differences, class geometry, and classifier boundaries in an opaque representation. This obscures why a node is classified and what feature-level graph-learning mechanisms a dataset requires. We propose WG-SRC, a white-box signal-subspace probe for prediction and graph dataset diagnosis.