Research map / Graph learning / Graph self-supervised learning
DGTF: Cross-Domain Decentralized Graph Learning with Topology-Aware Knowledge Fusion
Ruisheng Zheng, Mingyi Li, Xiao Zhang, Hongjian Shi, Yanjie Fu, Yuan Yuan, Dongxiao Yu
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 Topological Representation Module (ablation) | DGTF · Graph classification (SM-BIO) | accuracy (%): 75.48 -> 83.39 | Improved |
| add Topological Representation Module (ablation) | DGTF · Graph classification (SM-BIO-CV-SN) | accuracy (%): 69.07 -> 76.48 | Improved |
| add Semantic Context Module (ablation) | DGTF · Graph classification (SM-BIO-CV) | accuracy (%): 62.85 -> 78.77 | Improved |
| add Semantic Context Module (ablation) | DGTF · Graph classification (SM-BIO-CV-SN) | accuracy (%): 61.92 -> 76.48 | Improved |
| add cross client attention instead of uniform weighting (ablation) | DGTF · Graph classification (SM-BIO) | accuracy (%): 79.97 -> 83.39 | Improved |
| add cross client attention instead of uniform weighting (ablation) | DGTF · Graph classification (SM-BIO-CV-SN) | accuracy (%): 73.55 -> 76.48 | Improved |
| use self-supervised instead of supervised local training (ablation) | DGTF · Graph classification (SM-BIO) | accuracy (%): 76.81 -> 83.39 | Improved |
| use self-supervised instead of supervised local training (ablation) | DGTF · Graph classification (SM-BIO-CV-SN) | accuracy (%): 67.77 -> 76.48 | Improved |
| increase local epochs (ablation) | DGTF · Graph classification (three cross-domain settings) | accuracy (%) | Hurt |
| share node-feature-based models (GCN, GCFL) instead of topology-only (FedStar, FedSSP) | baselines with model-heterogeneity methods · Graph classification (cross-domain) | accuracy (%) | Hurt |
Closest papers by meaning
- Multi-Domain Graph Foundation Models: Robust Knowledge Transfer via Topology Alignment (ICML 2025 · similarity 0.90)
- Federated Graph Learning under Domain Shift with Generalizable Prototypes (AAAI 2024 · similarity 0.86)
- TFGDA: Exploring Topology and Feature Alignment in Semi-supervised Graph Domain Adaptation through Robust Clustering (NeurIPS 2024 · similarity 0.86)
- Towards Effective Federated Graph Foundation Model via Mitigating Knowledge Entanglement (NeurIPS 2025 · similarity 0.86)
- Toward Data-centric Directed Graph Learning: An Entropy-driven Approach (ICML 2025 · similarity 0.86)
- GraphKeeper: Graph Domain-Incremental Learning via Knowledge Disentanglement and Preservation (NeurIPS 2025 · 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.
