Research map / Graph learning / Graph algorithms and clustering
Deep Latent Graph 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 |
|---|---|---|---|
| add latent topology learning (deterministic, DLGM-D) instead of fixed Delaunay topology | BBGM pipeline · Pascal VOC | accuracy (inliers): 80.1 -> 82.9 | Helped |
| add generative latent topology learning (DLGM-G) instead of fixed Delaunay topology | BBGM pipeline · Pascal VOC | accuracy (inliers): 80.1 -> 83.8 | Helped |
| add generative latent topology learning (DLGM-G) | BBGM pipeline · SPair-71K | accuracy: 78.9 -> 82.0 | Helped |
| add deterministic latent topology learning (DLGM-D) | BBGM pipeline · SPair-71K | accuracy: 78.9 -> 81.3 | Helped |
| add generative latent topology learning (DLGM-G) | BBGM pipeline · Pascal VOC | F1-score (inliers and outliers): 61.4 -> 64.8 | Helped |
| add deterministic latent topology learning (DLGM-D) | BBGM pipeline · Pascal VOC | F1-score (inliers and outliers): 61.4 -> 63.9 | Helped |
| remove consistency loss LC (keep matching and locality losses) in DLGM-D (ablation) | DLGM-D · Pascal VOC | accuracy (inliers): 82.9 -> 79.5 when removed | Hurt |
| remove locality loss LL (keep matching and consistency losses) in DLGM-D (ablation) | DLGM-D · Pascal VOC | accuracy (inliers): 82.9 -> 79.8 when removed | Hurt |
| remove consistency loss LC in DLGM-G (ablation) | DLGM-G · Pascal VOC | accuracy (inliers): 83.8 -> 80.4 when removed | Hurt |
| remove locality loss LL in DLGM-G (ablation) | DLGM-G · Pascal VOC | accuracy (inliers): 83.8 -> 80.9 when removed | Hurt |
| increase generator sample size from 1 to 16 in DLGM-G (ablation) | DLGM-G · Pascal VOC | accuracy (inliers): 82.5 -> 83.8 (plateaus beyond 16; 32 gives 83.7) | Helped |
| generative (DLGM-G) vs deterministic (DLGM-D) latent topology (ablation) | DLGM-G vs DLGM-D · Pascal VOC | accuracy (inliers): 82.9 -> 83.8 | Helped |
Closest papers by meaning
- Learning Partial Graph Matching via Optimal Partial Transport (ICLR 2025 · similarity 0.89)
- Position: Graph Matching Systems Deserve Better Benchmarks (ICML 2025 · similarity 0.87)
- Unifying Generation and Prediction on Graphs with Latent Graph Diffusion (NeurIPS 2024 · similarity 0.87)
- Charting the Design Space of Neural Graph Representations for Subgraph Matching (ICLR 2025 · similarity 0.87)
- Iteratively Refined Early Interaction Alignment for Subgraph Matching based Graph Retrieval (NeurIPS 2024 · similarity 0.87)
- Discrete Cycle-Consistency Based Unsupervised Deep Graph Matching (AAAI 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.