AdNGCL introduces loss-gated adaptive negative scheduling across hardness strata to improve accuracy on graph contrastive learning benchmarks while controlling computational cost.
Adversarial contrastive graph augmentation with counterfactual regularization
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
-
Adaptive Negative Scheduling for Graph Contrastive Learning
AdNGCL introduces loss-gated adaptive negative scheduling across hardness strata to improve accuracy on graph contrastive learning benchmarks while controlling computational cost.