Research map / Graph learning / Graph self-supervised learning
Scalable Semi-supervised Community Search via Graph Transformer on Attributed Heterogeneous Information Networks
Linlin Ding, Zhaosong Zhao, Mo Li, Yishan Pan, Xin Wang, Renata Borovica-Gajic
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 |
|---|---|---|---|
| use SCSAH instead of strongest AHIN baseline | SCSAH · DBLP | F1 (%): 79.25 -> 87.76 | Improved |
| use SCSAH instead of strongest baseline | SCSAH · IMDB | F1 (%): 66.05 -> 75.34 | Improved |
| use SCSAH on MAG where only TransZero also runs | SCSAH · MAG | F1 (%): 32.59 -> 42.45 | Improved |
| add MvSF2Token multi-view subgraph tokens (ablation) | SCSAH · IMDB/DBLP/ACM | F1/JAC/NMI | Improved |
| add Zoom-Aware Transformer instead of standard Transformer (ablation) | SCSAH · IMDB/DBLP/ACM | F1/JAC/NMI | Improved |
| add unified contrastive loss (ablation) | SCSAH · IMDB/DBLP/ACM | average performance: 15.2% average drop when removed | Improved |
| add auxiliary node classification and semantic contrastive loss (ablation) | SCSAH · IMDB/DBLP/ACM | F1/JAC/NMI | Improved |
| set cohesion threshold theta to 0.2 (vs higher values) (ablation) | SCSAH · IMDB/DBLP | F1/JAC/NMI | Improved |
| use 1-2 Transformer layers instead of more (ablation) | SCSAH · IMDB/DBLP | F1/JAC/NMI | Improved |
| use subgraph tokens to cut training cost | SCSAH · IMDB-OAG | training time: 10.43x faster training | Improved |
Closest papers by meaning
- Heterogeneous Graph Neural Network on Semantic Tree (AAAI 2025 · similarity 0.86)
- HeterSEED: Semantics–Structure Decoupling for Heterogeneous Graph Learning under Heterophily (NeurIPS 2026 · similarity 0.86)
- Generalization Principles for Inference over Text-Attributed Graphs with Large Language Models (ICML 2025 · similarity 0.86)
- An Efficient Subgraph-Inferring Framework for Large-Scale Heterogeneous Graphs (AAAI 2024 · similarity 0.86)
- SSTAG: Structure-Aware Self-Supervised Learning Method for Text-Attributed Graphs (NeurIPS 2025 · similarity 0.85)
- Scalable Attribute-Missing Graph Clustering via Neighborhood Differentiation (ICML 2025 · similarity 0.85)
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.
