Research map / Graph learning / GNN architectures
T2-GNN: Graph Neural Networks for Graphs with Incomplete Features and Structure via Teacher-Student Distillation
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 |
|---|---|---|---|
| Add feature and structure teachers with dual distillation to GCN | GCN · Eight benchmark datasets | accuracy: +9.12% | Helped |
| Add feature and structure teachers with dual distillation to GAT | GAT · Eight benchmark datasets | accuracy: +4.51% | Helped |
| Add feature and structure teachers with dual distillation to GraphSAGE | GraphSAGE · Eight benchmark datasets | accuracy: +5.21% | Helped |
| Add feature and structure teachers with dual distillation to APPNP | APPNP · Eight benchmark datasets | accuracy: +6.77% | Helped |
| Train feature and structure teachers separately rather than combining them into one teacher (ablation) | T2-GCN · Node classification, most datasets | accuracy | Helped |
| Train teachers separately rather than jointly end-to-end with the student (ablation) | T2-GCN · Node classification, most datasets | accuracy | Helped |
| Add the structure teacher (ablation) | T2-GNN · Node classification across eight datasets | accuracy | Helped |
| Add the feature teacher (ablation) | T2-GNN · Node classification across eight datasets | accuracy | Helped |
| Add logit-based distillation (ablation) | T2-GNN · Node classification across eight datasets | accuracy | Helped |
| Add intermediate-representation distillation (ablation) | T2-GNN · Node classification across eight datasets | accuracy | Helped |
| Use feature-teacher guidance instead of structure-teacher guidance (ablation) | T2-GNN · Chameleon and Squirrel | accuracy | Mixed |
| Use too few or too many PPR-enhanced neighbors (ablation) | T2-GNN structure teacher · Wisconsin and Squirrel | accuracy | Hurt |
Closest papers by meaning
- Graph-less Neural Networks: Teaching Old MLPs New Tricks Via Distillation (ICLR 2022 · similarity 0.91)
- Demystifying Structural Disparity in Graph Neural Networks: Can One Size Fit All? (NeurIPS 2023 · similarity 0.91)
- OpenGSL: A Comprehensive Benchmark for Graph Structure Learning (NeurIPS 2023 · similarity 0.91)
- TINED: GNNs-to-MLPs by Teacher Injection and Dirichlet Energy Distillation (ICML 2025 · similarity 0.90)
- On the Interplay between Graph Structure and Learning Algorithms in Graph Neural Networks (ICML 2025 · similarity 0.89)
- Can Classic GNNs Be Strong Baselines for Graph-level Tasks? Simple Architectures Meet Excellence (ICML 2025 · similarity 0.89)
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.