Research map / Label-efficient and robust learning / Self-supervised and contrastive
Self-supervised Heterogeneous Graph Pre-training Based on Structural Clustering
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 |
|---|---|---|---|
| replace K-means embedding-space clustering with structural clustering (LPA) for pseudo-labels (ablation) | SHGP · MAG classification 4% train | Micro-F1: 85.03 -> 98.23 | Helped |
| replace K-means embedding-space clustering with structural clustering (LPA) for pseudo-labels (ablation) | SHGP · IMDB classification 4% train | Macro-F1: 34.72 -> 43.36 | Helped |
| replace K-means embedding-space clustering with structural clustering (LPA) for pseudo-labels (ablation) | SHGP · MAG clustering | NMI: 42.75 -> 90.65 | Helped |
| add SHGP self-supervised pre-training on top of the HGCN encoder | HGCN · MAG classification 4% train | Micro-F1: 93.16 -> 98.23 | Helped |
| add SHGP pre-training on top of HGCN encoder | HGCN · ACM classification 8% train | Micro-F1: 78.08 -> 80.91 | Helped |
| SHGP vs HeCo on ACM 8% train | SHGP · ACM classification 8% train | Micro-F1: 80.94 -> 80.91 | No clear effect |
| SHGP vs HeCo clustering ARI on ACM | SHGP · ACM clustering | ARI: 32.69 -> 32.63 | No clear effect |
| increase warm-up epochs beyond inflection point (10-20) (ablation) | SHGP · all datasets classification | Micro-F1/Macro-F1 | Mixed |
Closest papers by meaning
- HGOT: Self-supervised Heterogeneous Graph Neural Network with Optimal Transport (ICML 2025 · similarity 0.90)
- Revisiting Self-Supervised Heterogeneous Graph Learning from Spectral Clustering Perspective (NeurIPS 2024 · similarity 0.90)
- HGPrompt: Bridging Homogeneous and Heterogeneous Graphs for Few-Shot Prompt Learning (AAAI 2024 · similarity 0.88)
- Heterogeneous Graph Neural Network on Semantic Tree (AAAI 2025 · similarity 0.87)
- HG-Adapter: Improving Pre-Trained Heterogeneous Graph Neural Networks with Dual Adapters (ICLR 2025 · similarity 0.86)
- Uncovering the Redundancy in Graph Self-supervised Learning Models (NeurIPS 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.