Research map / Graph learning / Graph self-supervised learning
Label Efficient Regularization and Propagation for Graph Node Classification
Tian Xie, Rajgopal Kannan, C.‐C. Jay Kuo
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 |
|---|---|---|---|
| replace GraphHop with LERP (reliable pseudo-labels, more label iterations) | GraphHop · Cora (1 label/class) | test accuracy (%): 59.12 -> 72.47 | Improved |
| replace GraphHop with LERP | GraphHop · Amazon Photo (2 labels/class) | test accuracy (%): 70.86 -> 83.10 | Improved |
| replace GraphHop with LERP | GraphHop · Coauthor CS (1 label/class) | test accuracy (%): 65.03 -> 82.46 | Improved |
| replace GraphHop with LERP at standard label rate | GraphHop · Cora (20 labels/class) | test accuracy (%): 81.05 -> 82.65 | Improved |
| LERP versus APPNP at 16 labels/class | APPNP · PubMed (16 labels/class) | test accuracy (%): 79.17 -> 77.20 | Hurt |
| add LR classifier training (LERP vs LERP-V with U_alpha=0) (ablation) | LERP · five datasets, various label rates | test accuracy (%) | Improved |
| increase label-embedding iterations max_iter from 0 (GraphHop-like) to up to 20 (ablation) | LERP · five datasets, various label rates | test accuracy (%) | Improved |
| increase sharpening temperature T above 1 (ablation) | LERP · CiteSeer (1 label/class) | test accuracy (%) | No clear effect |
| small sharpening temperature T (ablation) | LERP · CiteSeer (1 label/class) | test accuracy (%) | Hurt |
| slightly increase alpha and beta (ablation) | LERP · CiteSeer (1 label/class) | test accuracy (%) | Improved |
| use 20 alternate rounds instead of 100 (ablation) | LERP · five datasets (20 labels/class) | test accuracy (%) | No clear effect |
| apply LERP to kNN graph of images instead of GCN | GCN · COIL20 (1 label/class) | test accuracy (%): 61.05 -> 86.75 | Improved |
Closest papers by meaning
- Posterior Label Smoothing for Node Classification (AAAI 2026 · similarity 0.89)
- Normalize Then Propagate: Efficient Homophilous Regularization for Few-Shot Semi-Supervised Node Classification (AAAI 2025 · similarity 0.89)
- NLGT: Neighborhood-based and Label-enhanced Graph Transformer Framework for Node Classification (AAAI 2025 · similarity 0.88)
- HGDL: Heterogeneous Graph Label Distribution Learning (NeurIPS 2024 · similarity 0.87)
- Multi-Label Node Classification with Label Influence Propagation (ICLR 2025 · similarity 0.86)
- Unsupervised Neighborhood Propagation Kernel Layers for Semi-supervised Node Classification (AAAI 2024 · similarity 0.86)
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.
