Research map / Graph learning / GNN architectures
Clarify Confused Nodes via Separated Learning
Jiajun Zhou, Shengbo Gong, Xuanze Chen, Chenxuan Xie, Shanqing Yu, Qi Xuan, Xiaoniu Yang
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 separation strategies (target and source masking) (ablation) | NCGCN · Cora Full | accuracy (%): 69.70 -> 73.42 | Improved |
| add separation strategies (target and source masking) (ablation) | NCGCN · Chameleon Filtered | accuracy (%): 36.49 -> 46.84 | Improved |
| add separation strategies (target and source masking) (ablation) | NCGCN · Pubmed | accuracy (%): 90.73 -> 91.64 | Improved |
| add source masking to restrict inter-group message passing (ablation) | NCGCN · Squirrel Filtered | accuracy (%): 36.72 -> 54.15 | Improved |
| add source masking to restrict inter-group message passing (ablation) | NCGCN · Cora Full | accuracy (%): 67.98 -> 73.42 | Improved |
| guide separation by NC instead of NH metric (ablation) | NCGCN · Squirrel Filtered | accuracy (%): 31.81 -> 54.15 | Improved |
| guide separation by NC instead of NH metric (ablation) | NCGCN · Cora Full | accuracy (%): 66.17 -> 73.42 | Improved |
| guide separation by NH metric instead of no separation (ablation) | NCGCN · Pubmed | accuracy (%): 90.73 -> 90.88 | Improved |
| guide separation by NH metric instead of no separation (ablation) | NCGCN · Chameleon Filtered | accuracy (%): 36.49 -> 39.24 | Hurt |
| use 3 hops for NC instead of 1 or 2 (ablation) | NCGCN · Cora Full | accuracy (%): 73.42 -> 56.33 | Hurt |
| use 3 hops for NC instead of 1 or 2 (ablation) | NCGCN · Pubmed | accuracy (%): 91.64 -> 91.17 | No clear effect |
| replace GCN backbone with NC-guided separated learning | NCGCN · 10 benchmarks | accuracy (%): average improvement of 8.04% | Improved |
Closest papers by meaning
- Is Homophily a Necessity for Graph Neural Networks? (ICLR 2022 · similarity 0.89)
- ES-GNN: Generalizing Graph Neural Networks Beyond Homophily With Edge Splitting (TPAMI 2024 · similarity 0.89)
- Integrating Co-Training with Edge Discrimination to Enhance Graph Neural Networks Under Heterophily (AAAI 2025 · similarity 0.89)
- Bag of Tricks for Training Deeper Graph Neural Networks: A Comprehensive Benchmark Study (TPAMI 2022 · similarity 0.88)
- What Is Missing For Graph Homophily? Disentangling Graph Homophily For Graph Neural Networks (NeurIPS 2024 · similarity 0.88)
- Understanding Heterophily for Graph Neural Networks (ICML 2024 · 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.
