Research map / Graph learning / GNN architectures
Finding Global Homophily in Graph Neural Networks When Meeting Heterophily
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace CPGNN with GloGNN | Node classification · Texas | accuracy (%): 70.42 -> 84.32 | Helped |
| replace CPGNN with GloGNN | Node classification · Squirrel | accuracy (%): 37.03 -> 57.54 | Helped |
| replace CPGNN with GloGNN | Node classification · Chameleon | accuracy (%): 56.93 -> 69.78 | Helped |
| replace CPGNN with GloGNN | Node classification · Citeseer | accuracy (%): 72.06 -> 77.41 | Helped |
| replace CPGNN with GloGNN | Node classification · Pubmed | accuracy (%): 86.68 -> 89.62 | Helped |
| replace CPGNN with GloGNN | Node classification · Cora | accuracy (%): 83.64 -> 88.31 | Helped |
| use full GloGNN instead of H_A alone (ablation) | GloGNN · Genius | accuracy (%): 74.08 -> 90.66 | Helped |
| combine H_A and H_X instead of using either alone (ablation) | GloGNN · Penn94 | accuracy (%) | Helped |
| use H_X rather than H_A (ablation) | GloGNN · Texas | accuracy (%) | Helped |
| remove the regularization term (ablation) | GloGNN · Wisconsin, Cornell, Actor | accuracy (%) | Hurt |
| remove graph convolution (ablation) | GloGNN · Datasets shown in the hyper-parameter study | accuracy (%) | Hurt |
Closest papers by meaning
- Integrating Co-Training with Edge Discrimination to Enhance Graph Neural Networks Under Heterophily (AAAI 2025 · similarity 0.93)
- Making Classic GNNs Strong Baselines Across Varying Homophily: A Smoothness–Generalization Perspective (NeurIPS 2025 · similarity 0.93)
- Understanding Heterophily for Graph Neural Networks (ICML 2024 · similarity 0.92)
- On the Impact of Feature Heterophily on Link Prediction with Graph Neural Networks (NeurIPS 2024 · similarity 0.91)
- What Is Missing For Graph Homophily? Disentangling Graph Homophily For Graph Neural Networks (NeurIPS 2024 · similarity 0.91)
- GRAIN: Multi-Granular and Implicit Information Aggregation Graph Neural Network for Heterophilous Graphs (AAAI 2025 · similarity 0.89)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.