Research map / Graph learning / Graph self-supervised learning
A Multi-Scale Approach for Graph Link Prediction
Lei Cai, Shuiwang Ji
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 multi-scale aggregated subgraphs (mLink) vs single-scale SEAL | SEAL/DGCNN · 14 datasets, 50% training links | AUC: 85.10 -> 87.50 (BUP) | Improved |
| add multi-scale aggregated subgraphs vs SEAL | SEAL/DGCNN · 14 datasets, 80% training links | AUC: 81.37 -> 83.14 (Power) | Improved |
| combine all three scales vs only original scale (ablation) | mLink · GRQ (50% links) | AUC: 90.07 -> 92.44 | Improved |
| use only aggregated Scale-3 graph instead of original (ablation) | mLink · C.ele (50% links) | AUC: 81.23 -> 77.71 | Hurt |
| use only aggregated Scale-2 graph instead of original (ablation) | mLink · SMG (50% links) | AUC: 86.56 -> 87.06 | No clear effect |
| use mLink vs PageRank heuristic | link prediction · C.ele (80% links) | AUC: 89.14 vs 89.08 | Hurt |
Closest papers by meaning
- Line Graph Neural Networks for Link Prediction (TPAMI 2021 · similarity 0.91)
- Structural Information Enhanced Graph Representation for Link Prediction (AAAI 2024 · similarity 0.90)
- GNNs Meet Sequence Models Along the Shortest-Path: an Expressive Method for Link Prediction (NeurIPS 2026 · similarity 0.89)
- A Topological Perspective on Demystifying GNN-Based Link Prediction Performance (ICLR 2024 · similarity 0.88)
- Subgraph Invariant Learning Towards Large-Scale Graph Node Classification (AAAI 2025 · similarity 0.88)
- Pure Message Passing Can Estimate Common Neighbor for Link Prediction (NeurIPS 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.
