Research map / Graph learning / GNN architectures
Local Vertex Colouring Graph Neural Networks
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 |
|---|---|---|---|
| Use BFS-guided SGN rather than the strongest listed baseline | Vertex classification · Homophilic datasets: Computers, Photo, and Pubmed | accuracy: 90.7 vs 87.5; 96.1 vs 93.9; 90.2 vs 89.6 | Helped |
| Use BFS-guided SGN rather than the strongest listed baseline | Vertex classification · Homophilic datasets: [cite] and Cora | accuracy: 78.0 vs 80.5; 88.7 vs 88.6 | Mixed |
| Use DFS-guided SGN rather than the strongest listed baseline | Vertex classification · Homophilic datasets: Computers, Photo, [cite], and Cora | accuracy: 90.9 vs 87.5; 95.2 vs 93.9; 79.7 vs 80.5; 89.5 vs 88.6 | Mixed |
| Use BFS-guided SGN rather than the strongest listed baseline | Vertex classification · Heterophilic datasets: Wisconsin, Chameleon, and Squirrel | accuracy: 91.2 vs 86.7; 72.8 vs 67.3; 59.0 vs 50.2 | Helped |
| Use BFS-guided SGN rather than the strongest listed baseline | Vertex classification · Heterophilic datasets: Cornell and Texas | accuracy: 89.5 vs 91.8; 88.7 vs 93.0 | Hurt |
| Use DFS-guided SGN rather than the strongest listed baseline | Vertex classification · Heterophilic datasets: Wisconsin, Cornell, Texas, Chameleon, and Squirrel | accuracy: 84.1 vs 86.7; 83.2 vs 91.8; 86.8 vs 93.0; 56.6 vs 67.3; 47.0 vs 50.2 | Hurt |
| Use DFS-guided SGN rather than the strongest listed baseline | Graph classification · NCI1, PROTEINS, and ENZYMES | accuracy: 81.0 vs 80.0; 76.1 vs 75.8; 66.9 vs 65.2 | Helped |
| Use DFS-guided SGN rather than the strongest listed baseline | Graph classification · D&D and IMDB-BINARY | accuracy: 78.01 vs 78.4; 72.3 vs 73.0 | Hurt |
| Use BFS-guided SGN rather than the strongest listed baseline | Graph classification · D&D, NCI1, PROTEINS, ENZYMES, and IMDB-BINARY | accuracy: 76.3 vs 78.4; 78.8 vs 80.0; 74.0 vs 75.8; 64.8 vs 65.2; 71.4 vs 73.0 | Mixed |
Closest papers by meaning
- The Expressive Power of Path-Based Graph Neural Networks (ICML 2024 · similarity 0.87)
- Homomorphism Expressivity of Spectral Invariant Graph Neural Networks (ICLR 2025 · similarity 0.87)
- Towards Bridging Generalization and Expressivity of Graph Neural Networks (ICLR 2025 · similarity 0.87)
- Beyond Weisfeiler-Lehman: A Quantitative Framework for GNN Expressiveness (ICLR 2024 · similarity 0.87)
- Weisfeiler-Leman at the margin: When more expressivity matters (ICML 2024 · similarity 0.87)
- On the Rademacher Complexity of Graph Neural Networks: Unifying Expressivity and Geometry (NeurIPS 2026 · similarity 0.86)
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.