Using average controllability as edge weights or one-hot edge attributes yields small, inconsistent gains for graph anomaly detection across several GNN backbones.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Robust Anomaly Detection with Graph Neural Networks using Controllability
Using average controllability as edge weights or one-hot edge attributes yields small, inconsistent gains for graph anomaly detection across several GNN backbones.