Research map / Graph learning / GNN architectures
Restructuring Graph for Higher Homophily via Adaptive Spectral Clustering
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 |
|---|---|---|---|
| Restructure the graph with adaptive spectral clustering | GCN · Actor | accuracy: 30.7\pm0.5 -> 36.2\pm1.0 | Helped |
| Restructure the graph with adaptive spectral clustering | GCN · Chameleon | accuracy: 59.8\pm2.6 -> 66.9\pm3.1 | Helped |
| Restructure the graph with adaptive spectral clustering | GCN · Squirrel | accuracy: 36.9\pm1.3 -> 55.7\pm2.4 | Helped |
| Restructure the graph with adaptive spectral clustering | GCN · Wisconsin | accuracy: 64.1\pm6.3 -> 83.1\pm3.2 | Helped |
| Restructure the graph with adaptive spectral clustering | GCN · Cornell | accuracy: 59.2\pm3.2 -> 79.2\pm6.3 | Helped |
| Restructure the graph with adaptive spectral clustering | GCN · Texas | accuracy: 64.1\pm4.9 -> 78.4\pm5.4 | Helped |
| Add node features to the random-signal input (ablation) | GCN with graph restructuring · Actor | accuracy: 36.1\pm0.9 -> 36.2\pm1.0 | No clear effect |
| Add node features to the random-signal input (ablation) | GCN with graph restructuring · Chameleon | accuracy: 53.5\pm6.1 -> 66.9\pm3.1 | Helped |
| Add node features to the random-signal input (ablation) | GCN with graph restructuring · Squirrel | accuracy: 47.4\pm4.1 -> 55.7\pm2.4 | Helped |
| Add node features to the random-signal input (ablation) | GCN with graph restructuring · Wisconsin | accuracy: 78.4\pm4.2 -> 83.1\pm3.2 | Helped |
| Add node features to the random-signal input (ablation) | GCN with graph restructuring · Cornell | accuracy: 71.1\pm6.0 -> 79.2\pm6.3 | Helped |
| Add node features to the random-signal input (ablation) | GCN with graph restructuring · Texas | accuracy: 67.8\pm8.3 -> 78.4\pm5.4 | Helped |
Closest papers by meaning
- What Is Missing For Graph Homophily? Disentangling Graph Homophily For Graph Neural Networks (NeurIPS 2024 · similarity 0.90)
- Homophily-Related: Adaptive Hybrid Graph Filter for Multi-View Graph Clustering (AAAI 2024 · similarity 0.90)
- Homomorphism Expressivity of Spectral Invariant Graph Neural Networks (ICLR 2025 · similarity 0.89)
- Making Classic GNNs Strong Baselines Across Varying Homophily: A Smoothness–Generalization Perspective (NeurIPS 2025 · similarity 0.89)
- Unifying Homophily and Heterophily for Spectral Graph Neural Networks via Triple Filter Ensembles (NeurIPS 2024 · similarity 0.89)
- AutoSGNN: Automatic Propagation Mechanism Discovery for Spectral Graph Neural Networks (AAAI 2025 · similarity 0.88)
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.