Research map / Graph learning / GNN architectures
Self-Supervised Graph Neural Networks via Diverse and Interactive Message Passing
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 DIMP message passing instead of existing self-supervised GNN message passing | Self-supervised node classification · Cora, Citeseer, Pubmed, Coauthor-CS, Coauthor-P | accuracy: Cora: 82.9±0.7 -> 83.3±0.5; Citeseer: 72.6±0.7 -> 73.3±0.5; Pubmed: 79.5±1.1 -> 81.4±0.5; Coauthor-CS: 91.3±0.4 -> 92.1± | Helped |
| Use DIMP message passing instead of existing self-supervised GNN message passing | Self-supervised node classification · Amazon-C, Amazon-P | accuracy: Amazon-C: 80.8±0.4 -> 83.3±0.4; Amazon-P: 87.1±1.0 -> 88.7±0.2 | Mixed |
| Use DIMP instead of the strongest reported graph-classification baseline | Graph classification · MUTAG | accuracy: 89.7±1.1 -> 91.5±1.1 | Helped |
| Use DIMP instead of the strongest reported graph-classification baseline | Graph classification · PTC-MR, IMDB-BIN, IMDB-MULTI | accuracy: PTC-MR: 66.7±5.1 -> 64.2±1.2; IMDB-BIN: 75.1±5.1 -> 74.8±0.8; IMDB-MULTI: 52.3±2.8 -> 52.0±0.6 | Hurt |
| Use DIMP instead of MVGRL for node clustering | Node clustering · Cora, Citeseer, Pubmed | NMI: Cora: 0.572 -> 0.581; Citeseer: 0.469 -> 0.471; Pubmed: 0.322 -> 0.346 | Helped |
| Use DIMP instead of MVGRL for node clustering | Node clustering · Cora, Citeseer, Pubmed | ARI: Cora: 0.495 -> 0.522; Citeseer: 0.449 -> 0.471; Pubmed: 0.296 -> 0.328 | Helped |
| Use multiple DIMP layers rather than a one-layer self-supervised encoder | DIMP · Cora, Citeseer, Pubmed node classification | accuracy | Helped |
| Increase GCN/DGI depth | GCN and DGI · Cora, Citeseer, Pubmed node classification | accuracy | Hurt |
Closest papers by meaning
- HGOT: Self-supervised Heterogeneous Graph Neural Network with Optimal Transport (ICML 2025 · similarity 0.88)
- Uncovering the Redundancy in Graph Self-supervised Learning Models (NeurIPS 2024 · similarity 0.88)
- Integrating Co-Training with Edge Discrimination to Enhance Graph Neural Networks Under Heterophily (AAAI 2025 · similarity 0.87)
- NLGT: Neighborhood-based and Label-enhanced Graph Transformer Framework for Node Classification (AAAI 2025 · similarity 0.87)
- Sign is Not a Remedy: Multiset-to-Multiset Message Passing for Learning on Heterophilic Graphs (ICML 2024 · similarity 0.87)
- Normalize Then Propagate: Efficient Homophilous Regularization for Few-Shot Semi-Supervised Node Classification (AAAI 2025 · 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.