ACE adds a heterophily-aware auxiliary loss to coarsening-based GNN training, recovering discarded node-level information and improving accuracy on heterophilic graphs by up to ~15 points.
Graph coarsening via supervised granular-ball for scalable graph neural network training
1 Pith paper cite this work, alongside 7 external citations. Polarity classification is still indexing.
1
Pith paper citing it
7
external citations · OpenAlex
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Remedying Coarsening-Based GNN Training under Heterophily via Adaptive Complementary Enhancement
ACE adds a heterophily-aware auxiliary loss to coarsening-based GNN training, recovering discarded node-level information and improving accuracy on heterophilic graphs by up to ~15 points.