Research map / Graph learning / Graph self-supervised learning
Deformable Graph Transformer
Jinyoung Park, Seongjun Yun, Hyeonjin Park, Jaewoo Kang, Jisu Jeong, Kyung-Min Kim, Jung-Woo Ha, Hyunwoo J. Kim
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 relative ordering with BFS instead of absolute BFS ordering (ablation) | DGT · Cora | accuracy (%): 81.22 -> 86.60 | Improved |
| use relative ordering with BFS instead of absolute BFS ordering (ablation) | DGT · Chameleon | accuracy (%): 71.95 -> 73.04 | Improved |
| use absolute BFS ordering instead of random permuted absolute ordering (ablation) | DGT · Cora | accuracy (%): 81.53 -> 81.22 | No clear effect |
| add multiple sorting criteria (BFS, PPR, Feat) to relative ordering (ablation) | DGT · Cora | accuracy (%): 86.60 -> 87.55 | Improved |
| add multiple sorting criteria (BFS, PPR, Feat) to relative ordering (ablation) | DGT · Squirrel | accuracy (%): 62.58 -> 63.78 | Improved |
| add multiple sorting criteria to absolute ordering (ablation) | DGT · Citeseer | accuracy (%): 72.28 -> 70.20 | Hurt |
| use multiple node sequences instead of Feature criterion only (ablation) | DGT · Cora | accuracy (%): 84.09 -> 87.55 | Improved |
| add Katz positional encoding (ablation) | DGT · Squirrel | accuracy (%): 46.59 -> 63.78 | Improved |
| add Katz positional encoding (ablation) | DGT · Chameleon | accuracy (%): 59.44 -> 73.48 | Improved |
| add Katz positional encoding (ablation) | Transformer · Chameleon | accuracy (%): 45.93 -> 72.13 | Improved |
| add Laplacian Eigvecs positional encoding (ablation) | DGT · Cora | accuracy (%): 86.32 -> 83.94 | Hurt |
| use kernel-based instead of bilinear interpolation (ablation) | DGT · Cora | accuracy (%): 86.02 -> 87.55 | 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 0.90)
- Generalizable Insights for Graph Transformers in Theory and Practice (NeurIPS 2025 · similarity 0.89)
- Towards Principled Graph Transformers (NeurIPS 2024 · similarity 0.88)
- Unifying and Enhancing Graph Transformers via a Hierarchical Mask Framework (NeurIPS 2025 · similarity 0.88)
- Even Sparser Graph Transformers (NeurIPS 2024 · similarity 0.88)
- Cluster-wise Graph Transformer with Dual-granularity Kernelized Attention (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.
