RAwR augments graphs with role-aware quotient graphs from approximate equitable partitions to accelerate long-range communication in GNNs, achieving SOTA results on homophilic, heterophilic, and long-range benchmarks while recovering master-node rewiring in the limit.
Geom-GCN: Geometric Graph Convolutional Networks , year =
7 Pith papers cite this work. Polarity classification is still indexing.
7
Pith papers citing it
representative citing papers
FC-GSSL improves graph SSL by generating high-frequency biased corrupted graphs via low-frequency contribution-based corruption, reconstructing low-frequency features in an autoencoder, and aligning multi-view representations to fuse frequency bands.
PLACE is a prompt-augmented graph framework for attributed community search that integrates learnable tokens with GNNs via alternating training and divide-and-conquer scaling, achieving 22% higher average F1 scores than prior methods on nine real-world graphs.
Transductive Sharpening adds an entropy-minimization term on unlabeled-node predictions to the training objective for graph node classification.
Inductive subgraphs serve as shortcuts in heterophilic graphs, and CD-GNN disentangles spurious from causal subgraphs by blocking non-causal paths to improve robustness and accuracy.
citing papers explorer
-
RAwR: Role-Aware Rewiring via Approximate Equitable Partition
RAwR augments graphs with role-aware quotient graphs from approximate equitable partitions to accelerate long-range communication in GNNs, achieving SOTA results on homophilic, heterophilic, and long-range benchmarks while recovering master-node rewiring in the limit.
-
Graph self-supervised learning based on frequency corruption
FC-GSSL improves graph SSL by generating high-frequency biased corrupted graphs via low-frequency contribution-based corruption, reconstructing low-frequency features in an autoencoder, and aligning multi-view representations to fuse frequency bands.
-
PLACE: Prompt Learning for Attributed Community Search in Large Graphs
PLACE is a prompt-augmented graph framework for attributed community search that integrates learnable tokens with GNNs via alternating training and divide-and-conquer scaling, achieving 22% higher average F1 scores than prior methods on nine real-world graphs.
-
Graph Transductive Sharpening: Leveraging Unlabeled Predictions in Node Classification
Transductive Sharpening adds an entropy-minimization term on unlabeled-node predictions to the training objective for graph node classification.
-
Inductive Subgraphs as Shortcuts: Causal Disentanglement for Heterophilic Graph Learning
Inductive subgraphs serve as shortcuts in heterophilic graphs, and CD-GNN disentangles spurious from causal subgraphs by blocking non-causal paths to improve robustness and accuracy.
- Graph Navier Stokes Networks
- A complete discussion on fully reconfigurable, digital, scalable, graph and sparsity-aware near-memory accelerator for graph neural networks