Research map / Graph learning / GNN architectures
LGAN: An Efficient High-Order Graph Neural Network via the Line Graph Aggregation
Lin Du, Lu Bai, Jincheng Li, Lixin Cui, Hangyuan Du, Lichi Zhang, Yuting Chen, Zhao Li
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 line graph aggregation over node-centered subgraphs (LGAN vs GIN) | GIN · MUTAG | accuracy (%): 89.4±5.6 -> 92.5±6.3 | Improved |
| add line graph aggregation (LGAN vs GIN) | GIN · IMDB-B | accuracy (%): 75.1±5.1 -> 76.7±4.0 | Improved |
| add line graph aggregation (LGAN vs GIN) | GIN · COLLAB | accuracy (%): 80.2±1.9 -> 82.8±1.5 | Improved |
| add line graph aggregation (LGAN vs GIN) | GIN · PTC(MR) | accuracy (%): 64.6±7.0 -> 67.4±6.2 | Improved |
| replace 2-FWL-style k-WL GNN PPGN by LGAN | PPGN · COLLAB | accuracy (%): 81.4±1.4 -> 82.8±1.5 | Improved |
| replace CW Networks by LGAN | CW Networks · IMDB-B | accuracy (%): 75.6±3.7 -> 76.7±4.0 | Improved |
| replace CW Networks by LGAN | CW Networks · PTC(MR) | accuracy (%): 68.2±5.6 -> 67.4±6.2 | Hurt |
| replace CW Networks by LGAN | CW Networks · MUTAG | accuracy (%): 92.7±6.1 -> 92.5±6.3 | No clear effect |
| add residual connection (LGAN-res vs LGAN) | LGAN · PROTEINS | accuracy (%): 77.3±3.7 -> 76.8±3.6 | No clear effect |
| add residual connection (LGAN-res vs LGAN) | LGAN · IMDB-M | accuracy (%): 53.3±3.2 -> 53.5±2.7 | No clear effect |
| add residual connection (LGAN-res vs LGAN) | LGAN · PTC(MR) | accuracy (%): 67.4±6.2 -> 66.6±7.0 | Hurt |
Closest papers by meaning
- A New Perspective on "How Graph Neural Networks Go Beyond Weisfeiler-Lehman?" (ICLR 2022 · similarity 0.91)
- A Practical, Progressively-Expressive GNN (NeurIPS 2022 · similarity 0.90)
- Weisfeiler and Leman Go Neural: Higher-Order Graph Neural Networks (AAAI 2019 · similarity 0.90)
- Graph Neural Networks Are More Than Filters: Revisiting and Benchmarking from A Spectral Perspective (ICLR 2025 · similarity 0.89)
- Integrating Co-Training with Edge Discrimination to Enhance Graph Neural Networks Under Heterophily (AAAI 2025 · similarity 0.88)
- NLGT: Neighborhood-based and Label-enhanced Graph Transformer Framework for Node Classification (AAAI 2025 · similarity 0.88)
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.
