WG-SRC is a white-box signal-subspace classifier that decomposes graph node classification into interpretable components to produce operational fingerprints distinguishing dataset behaviors like low-pass dominance or high-pass noise.
NodeFormer: A scalable graph structure learning transformer for node classification
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Operational Feature Fingerprints of Graph Datasets via a White-Box Signal-Subspace Probe
WG-SRC is a white-box signal-subspace classifier that decomposes graph node classification into interpretable components to produce operational fingerprints distinguishing dataset behaviors like low-pass dominance or high-pass noise.