HeRB combines heterophily-lessening edge augmentation with homophilic knowledge transfer to improve node classification for low-degree tail nodes.
Representation learning for scale-free networks,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
HeRB: Heterophily-Resolved Structure Balancer for Graph Neural Networks
HeRB combines heterophily-lessening edge augmentation with homophilic knowledge transfer to improve node classification for low-degree tail nodes.