Research map / Graph learning / Graph self-supervised learning
Generative Subgraph Contrast for Self-Supervised Graph Representation Learning
Yuehui Han, Le Hui, Haobo Jiang, Jianjun Qian, Jin Xie
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 OT distance contrastive loss instead of readout similarity (ablation) | GSC · Cora | accuracy: 83.6 -> 84.6 | Improved |
| add OT distance contrastive loss instead of readout similarity (ablation) | GSC · Citeseer | accuracy: 72 -> 73.7 | Improved |
| add OT distance contrastive loss instead of readout similarity (ablation) | GSC · Pubmed | accuracy: 78.5 -> 82.1 | Improved |
| add subgraph generation module instead of perturbation (ablation) | GSC · Cora | accuracy: 84.1 -> 84.6 | Improved |
| add subgraph generation module instead of perturbation (ablation) | GSC · Citeseer | accuracy: 72.3 -> 73.7 | Improved |
| add subgraph generation module instead of perturbation (ablation) | GSC · Pubmed | accuracy: 80.4 -> 82.1 | Improved |
| use BFS subgraph sampling instead of importance score (ablation) | GSC · Cora | accuracy: 83.5 -> 84.6 | Improved |
| use BFS subgraph sampling instead of importance score (ablation) | GSC · Citeseer | accuracy: 71.68 -> 73.7 | Improved |
| use BFS subgraph sampling instead of importance score (ablation) | GSC · Pubmed | accuracy: 80.6 -> 82.1 | Improved |
| use BFS subgraph sampling instead of random walk (ablation) | GSC · Cora | accuracy: 84.2 -> 84.6 | Improved |
| use BFS subgraph sampling instead of random walk (ablation) | GSC · Citeseer | accuracy: 72.4 -> 73.7 | Improved |
| use BFS subgraph sampling instead of random walk (ablation) | GSC · Pubmed | accuracy: 81.2 -> 82.1 | Improved |
Closest papers by meaning
- Improving Graph Contrastive Learning via Adaptive Positive Sampling (CVPR 2024 · similarity 0.90)
- Hierarchical Topology Isomorphism Expertise Embedded Graph Contrastive Learning (AAAI 2024 · similarity 0.89)
- Does GCL Need a Large Number of Negative Samples? Enhancing Graph Contrastive Learning with Effective and Efficient Negative Sampling (AAAI 2025 · similarity 0.89)
- ReGCL: Rethinking Message Passing in Graph Contrastive Learning (AAAI 2024 · similarity 0.89)
- A New Mechanism for Eliminating Implicit Conflict in Graph Contrastive Learning (AAAI 2024 · similarity 0.89)
- Graph Structure Refinement with Energy-based Contrastive Learning (AAAI 2025 · similarity 0.89)
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.
