Research map / Graph learning / Graph self-supervised learning
Multi-Domain Generalized Graph Meta Learning
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 MD-Gram to meta-train across multiple source domains and adapt to the held-out graph domain | VGAE-based link prediction · ARY-P, ARP-Y, APY-R, PRY-A; 10% known edges | AUC (%): 70.62 -> 76.79; 79.28 -> 85.30; 66.09 -> 73.00; 74.73 -> 80.50 | Helped |
| Use MD-Gram for multi-domain graph meta-learning | VGAE-based link prediction · 20% known edges, final AUC | AUC (%): 88.73 -> 89.14; 90.52 -> 90.71; 81.94 -> 82.24; 90.80 -> 88.19 | Mixed |
| Add empirical graph generalization using node and edge empirical vectors (ablation) | MD-Gram · ARY-P, ARP-Y, APY-R, PRY-A; 20% known edges, 10-epoch AUC | AUC (%): 64.57 -> 79.76; 79.70 -> 87.17; 68.96 -> 76.33; 65.09 -> 81.55 | Helped |
| Add multi-domain graphs transformation into a common domain (ablation) | MD-Gram · ARY-P, ARP-Y, APY-R, PRY-A; 20% known edges, 10-epoch AUC | AUC (%): 70.59 -> 79.76; 85.60 -> 87.17; 72.82 -> 76.33; 78.70 -> 81.55 | Helped |
| Add domain-specific GNN enhancement (ablation) | MD-Gram · ARY-P, ARP-Y, APY-R, PRY-A; 20% known edges, 10-epoch AUC | AUC (%): 76.98 -> 79.76; 88.79 -> 87.17; 74.12 -> 76.33; 76.09 -> 81.55 | Mixed |
| Add edge empirical vectors (ablation) | MD-Gram · ARY-P, ARP-Y, APY-R, PRY-A; 20% known edges, 10-epoch AUC | AUC (%): 76.41 -> 79.76; 87.06 -> 87.17; 74.47 -> 76.33; 75.76 -> 81.55 | Helped |
| Add target-graph preprocessing to transform it into the common domain (ablation) | MD-Gram during meta testing · ARY-P, ARP-Y, APY-R, PRY-A; 20% known edges, 10-epoch AUC | AUC (%): 77.78 -> 79.76; 87.04 -> 87.17; 74.72 -> 76.33; 79.19 -> 81.55 | Helped |
Closest papers by meaning
- Unified Graph Neural Networks Pre-training for Multi-domain Graphs (AAAI 2025 · similarity 0.91)
- When Sparse Graph Representation Learning Falls into Domain Shift: Data Augmentation for Cross-Domain Graph Meta-Learning (Student Abstract) (AAAI 2024 · similarity 0.89)
- How Much Can Transfer? BRIDGE: Bounded Multi-Domain Graph Foundation Model with Generalization Guarantees (ICML 2025 · similarity 0.87)
- Multi-Domain Graph Foundation Models: Robust Knowledge Transfer via Topology Alignment (ICML 2025 · similarity 0.87)
- Graph Metanetworks for Processing Diverse Neural Architectures (ICLR 2024 · similarity 0.87)
- Rethinking Propagation for Unsupervised Graph Domain Adaptation (AAAI 2024 · 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.