A multi-space GNN with learnable curvature and distance-aware propagation that outperforms existing node anomaly detectors on 9 benchmark datasets under limited supervision.
Let X and Y denote the distribution of the normal node and the informati on over Rd, respectively
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
-
SpaceGNN: Multi-Space Graph Neural Network for Node Anomaly Detection with Extremely Limited Labels
A multi-space GNN with learnable curvature and distance-aware propagation that outperforms existing node anomaly detectors on 9 benchmark datasets under limited supervision.