Research map / Graph learning / GNN architectures
Learning-Based Efficient Graph Similarity Computation via Multi-Scale Convolutional Set Matching
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 GraphSim instead of the strongest listed baseline | graph similarity prediction · Aids (GED) | mse (10^-3): 1.189 -> 0.787 | Helped |
| Use GraphSim instead of the strongest listed baseline | graph similarity prediction · Linux (GED) | mse (10^-3): 1.027 -> 0.058 | Helped |
| Use GraphSim instead of the strongest listed baseline | graph similarity prediction · Imdb (GED) | mse (10^-3): 1.264 -> 0.743 | Helped |
| Use GraphSim instead of the strongest listed baseline | graph similarity prediction · Ptc (GED) | mse (10^-3): 0.850 -> 0.749 | Helped |
| Use GraphSim instead of the strongest listed baseline | graph similarity prediction · Aids (MCS) | mse (10^-3): 1.750 -> 2.402 | Hurt |
| Use GraphSim instead of the strongest listed baseline | graph similarity prediction · Linux (MCS) | mse (10^-3): 0.598 -> 0.164 | Helped |
| Use GraphSim instead of the strongest listed baseline | graph similarity prediction · Imdb (MCS) | mse (10^-3): 0.590 -> 0.307 | Helped |
| Use GraphSim instead of the strongest listed baseline | graph similarity prediction · Ptc (MCS) | mse (10^-3): 3.781 -> 2.268 | Helped |
| Use max padding plus resizing rather than dataset-wide zero padding (ablation) | GraphSim · Imdb (GED) | mse (10^-3): 6.455 -> 0.743 | Helped |
| Use multi-scale embeddings rather than only last-layer embeddings (ablation) | GraphSim with max padding and resizing · Imdb (GED) | mse (10^-3): 0.896 -> 0.743 | Helped |
| Add BFS node ordering (ablation) | GraphSim · Imdb (GED) | mse (10^-3): 1.105 -> 0.743 | Helped |
| Replace GCN embeddings with GMN node embeddings (ablation) | GraphSim · Aids (GED) | mse (10^-3): 0.787 -> 0.557 | Helped |
Closest papers by meaning
- Efficient Graph Similarity Computation with Alignment Regularization (NeurIPS 2022 · similarity 0.90)
- Position: Graph Matching Systems Deserve Better Benchmarks (ICML 2025 · similarity 0.89)
- Exploring Consistency in Graph Representations: from Graph Kernels to Graph Neural Networks (NeurIPS 2024 · similarity 0.88)
- GraSP: Simple Yet Effective Graph Similarity Predictions (AAAI 2025 · similarity 0.87)
- Graph Classification via Reference Distribution Learning: Theory and Practice (NeurIPS 2024 · similarity 0.87)
- NoiseHGNN: Synthesized Similarity Graph-Based Neural Network for Noised Heterogeneous Graph Representation Learning (AAAI 2025 · similarity 0.87)
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.