Research map / Graph learning / GNN architectures
Rethinking Graph Transformers with Spectral Attention
Devin Kreuzer, Dominique Beaini, William L. Hamilton, Vincent Létourneau, Prudencio Tossou
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 node-wise learned positional encoding (LPE) to sparse and full attention (ablation) | SAN · ZINC and MolHIV | MAE / ROC-AUC | Improved |
| add node-wise LPE (ablation) | SAN · PATTERN and CLUSTER | accuracy | Improved |
| use full attention instead of sparse attention (gamma > 0) (ablation) | SAN · PATTERN and CLUSTER | accuracy | Improved |
| use full attention instead of sparse attention (ablation) | SAN · ZINC and MolHIV | MAE / ROC-AUC | Mixed |
| SAN vs MolHIV state of the art (PNA, DGN) on test set | SAN · MolHIV test | ROC-AUC | Hurt |
| SAN vs PNA and DGN on validation set | SAN · MolHIV validation | ROC-AUC (%): 85.30 vs 84.25 (PNA), 84.70 (DGN) | Improved |
| SAN vs other attention-based models and GT (full) | SAN · ZINC, PATTERN, CLUSTER, MolHIV | benchmark metrics | Improved |
Closest papers by meaning
- What Improves the Generalization of Graph Transformers? A Theoretical Dive into the Self-attention and Positional Encoding (ICML 2024 · similarity 0.91)
- Towards Principled Graph Transformers (NeurIPS 2024 · similarity 0.90)
- Graph Convolutions Enrich the Self-Attention in Transformers! (NeurIPS 2024 · similarity 0.89)
- Graph External Attention Enhanced Transformer (ICML 2024 · similarity 0.89)
- Learning Efficient Positional Encodings with Graph Neural Networks (ICLR 2025 · similarity 0.89)
- Supra-Laplacian Encoding for Transformer on Dynamic Graphs (NeurIPS 2024 · 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.
