Research map / Graph learning / GNN architectures
Automatic Relation-Aware Graph Network Proliferation
Check 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 |
|---|---|---|---|
| Add hierarchical relation learning to the search space (ablation) | ARGNP · CLUSTER | AA (%): 71.92_{\pm0.29} -> 77.35_{\pm0.05} | Helped |
| Add hierarchical relation learning to the search space (ablation) | ARGNP · ZINC | MAE: 0.221_{\pm0.004} -> 0.136_{\pm0.002} | Helped |
| Add hierarchical relation learning to the search space (ablation) | ARGNP · CIFAR10 | OA (%): 67.10_{\pm0.51} -> 73.90_{\pm0.15} | Helped |
| Add hierarchical relation learning to the search space (ablation) | ARGNP · TSP | F1: 0.684_{\pm0.002} -> 0.855_{\pm0.001} | Helped |
| Add hierarchical relation learning to the search space (ablation) | ARGNP · ModelNet10 | OA (%): 94.23_{\pm 0.22} -> 95.87_{\pm0.22} | Helped |
| Add hierarchical relation learning to the search space (ablation) | ARGNP · ModelNet40 | OA (%): 91.85_{\pm 0.18} -> 93.33_{\pm0.15} | Helped |
| Use network proliferation search instead of a single global supernet (ablation) | R-space with DARTS · ZINC | MAE: 0.157_{\pm0.008} -> 0.150_{\pm0.006} | Helped |
| Use network proliferation search instead of a single global supernet (ablation) | R-space with SGAS · ZINC | MAE: 0.161_{\pm0.008} -> 0.155_{\pm0.003} | Helped |
| Use network proliferation search instead of a single global supernet (ablation) | R-space with DARTS at size 16 · ZINC | MAE | Helped |
| Use network proliferation search instead of a single global supernet (ablation) | R-space with SGAS at size 16 · ZINC | MAE | Helped |
| Remove the cell trick (ablation) | R-space with DARTS at size 8 · ZINC | MAE: 0.160_{\pm0.005} -> 0.157_{\pm0.008} | Helped |
| Remove the cell trick (ablation) | R-space with SGAS at size 8 · ZINC | MAE: 0.165_{\pm0.008} -> 0.161_{\pm0.008} | Helped |
Closest papers by meaning
- AutoSGNN: Automatic Propagation Mechanism Discovery for Spectral Graph Neural Networks (AAAI 2025 · similarity 0.87)
- RelGNN: Composite Message Passing for Relational Deep Learning (ICML 2025 · similarity 0.86)
- Charting the Design Space of Neural Graph Representations for Subgraph Matching (ICLR 2025 · similarity 0.86)
- Towards Dynamic Message Passing on Graphs (NeurIPS 2024 · similarity 0.86)
- EiG-Search: Generating Edge-Induced Subgraphs for GNN Explanation in Linear Time (ICML 2024 · similarity 0.86)
- GNNs Meet Sequence Models Along the Shortest-Path: an Expressive Method for Link Prediction (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.