Research map / Graph learning / Graph algorithms and clustering
Interpretable Neural Subgraph Matching for Graph Retrieval
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 edge alignment instead of node alignment (ablation) | ISONET · PTC-FR, PTC-FM, PTC-MM, PTC-MR, MUTAG, AIDS | MAP: Node-align (Node loss): 0.832, 0.859, 0.862, 0.838, 0.886, 0.849; ISONET: 0.857, 0.901, 0.906, 0.894, 0.920, 0.894 | Helped |
| use edge alignment instead of node alignment with edge loss (ablation) | ISONET · PTC-FR, PTC-FM, PTC-MM, PTC-MR, MUTAG, AIDS | MAP: Node-align (Edge loss): 0.803, 0.847, 0.828, 0.814, 0.815, 0.869; ISONET: 0.857, 0.901, 0.906, 0.894, 0.920, 0.894 | Helped |
| use asymmetric rather than symmetric scoring (ablation) | ISONET · PTC-FR, PTC-FM, PTC-MM, PTC-MR, MUTAG, AIDS | MAP: Sym: 0.858, 0.839, 0.867, 0.804, 0.891, 0.884; Asym: 0.857, 0.901, 0.906, 0.894, 0.920, 0.894 | Mixed |
| use edge-level rather than graph-level alignment | ISONET · PTC-FR, PTC-FM, PTC-MM, PTC-MR, MUTAG, AIDS | MAP: ISONET: 0.857, 0.901, 0.906, 0.894, 0.920, 0.894; GMN-match: 0.774, 0.783, 0.837, 0.805, 0.909, 0.835 | Helped |
| use asymmetric scoring rather than symmetric scoring (ablation) | ISONET and GMN variants · all six datasets | MAP: ISONET, GMN-embed and GMN-match asymmetric scoring significantly outperforms the symmetric model. | Helped |
Closest papers by meaning
- Iteratively Refined Early Interaction Alignment for Subgraph Matching based Graph Retrieval (NeurIPS 2024 · similarity 0.92)
- Charting the Design Space of Neural Graph Representations for Subgraph Matching (ICLR 2025 · similarity 0.90)
- RAGraph: A General Retrieval-Augmented Graph Learning Framework (NeurIPS 2024 · similarity 0.88)
- Contextual Tokenization for Graph Inverted Indices (NeurIPS 2025 · similarity 0.86)
- G-Retriever: Retrieval-Augmented Generation for Textual Graph Understanding and Question Answering (NeurIPS 2024 · similarity 0.86)
- How Interpretable Are Interpretable Graph Neural Networks? (ICML 2024 · 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.