Research map / Graph learning / GNN architectures
Scalable Graph Neural Networks via Bidirectional Propagation
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 localized bidirectional propagation | GBP · Cora | accuracy: 81.5 ± 0.6 -> 83.9 ± 0.7 | Helped |
| Use localized bidirectional propagation | GBP · Citeseer | accuracy: 71.3 ± 0.4 -> 72.9 ± 0.5 | Helped |
| Use localized bidirectional propagation | GBP · Pubmed | accuracy: 80.1 ± 0.2 -> 80.6 ± 0.4 | Helped |
| Use GBP rather than GraphSAINT | PPI, 6-layer · PPI | Time (s): 1961 -> 167 | Helped |
| Use GBP rather than GraphSAINT | Yelp, 4-layer · Yelp | F1-score: 64.7 ± 0.08 -> 65.4 ± 0.03 | Helped |
| Use GBP rather than GraphSAINT | Amazon, 4-layer · Amazon | F1-score: 91.5 ± 0.01 -> 91.5 ± 0.01 | No clear effect |
| Use GBP rather than SGC | Friendster, feature dimension 100 · Friendster | F1-score: 40.2 ± 0.01 -> 79.7 ± 0.32 | Helped |
| Use transition-probability weights instead of Personalized PageRank weights (ablation) | GBP on Friendster, feature dimension 100 · Friendster | F1-score: 7.3 ± 0.12 -> 79.7 ± 0.32 | Helped |
| Use GBP rather than SGC | PPI, 6-layer · PPI | F1-score: 62.4 ± 0.01 -> 99.3 ± 0.03 | Helped |
| Use GBP rather than PPRGo | Yelp, 6-layer · Yelp | Time (s): 1215 -> 30 | Helped |
Closest papers by meaning
- A Comprehensive Study on Large-Scale Graph Training: Benchmarking and Rethinking (NeurIPS 2022 · similarity 0.89)
- Can Classic GNNs Be Strong Baselines for Graph-level Tasks? Simple Architectures Meet Excellence (ICML 2025 · similarity 0.89)
- Generative Graph Pattern Machine (NeurIPS 2025 · similarity 0.89)
- A Scalable and Effective Alternative to Graph Transformers (AAAI 2025 · similarity 0.88)
- Classic GNNs are Strong Baselines: Reassessing GNNs for Node Classification (NeurIPS 2024 · similarity 0.88)
- Improving Graph Neural Networks by Learning Continuous Edge Directions (ICLR 2025 · similarity 0.88)
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.