Research map / Graph learning / Graph algorithms and clustering
SWING: Unlocking Implicit Graph Representations for Graph Random Features
Alessandro Manenti, Kumar Avinava Dubey, Arijit Sehanobish, Cesare Alippi, Krzysztof Marcin Choromanski
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 SWING-approximated graph-kernel soft attention mask | ViT · ImageNet val | top-1 accuracy: 80.10 -> 80.34 | Improved |
| add SWING-approximated graph-kernel soft attention mask | ViT · Places365 val | top-1 accuracy: 55.78 -> 55.97 | Improved |
| add standard GRF graph-kernel soft attention mask | ViT · ImageNet val | top-1 accuracy: 80.10 -> 80.09 | No clear effect |
| add standard GRF graph-kernel soft attention mask | ViT · Places365 val | top-1 accuracy: 55.78 -> 55.39 | Hurt |
| replace GRF node walks with SWING space walks | GRFs · Thingi10K vertex normal prediction | cosine similarity | No clear effect |
| replace GRF node walks with SWING space walks | GRFs · synthetic point clouds kernel approximation | Frobenius norm error | Improved |
| replace GRF node walks with SWING space walks | GRFs · synthetic point clouds | inference time | Improved |
| increase number of random features (ablation) | SWING · synthetic point clouds | Frobenius norm error | Improved |
Closest papers by meaning
- Computationally-efficient Graph Modeling with Refined Graph Random Features (ICML 2026 · similarity 0.88)
- General Graph Random Features (ICLR 2024 · similarity 0.88)
- Random Search Neural Networks for Efficient and Expressive Graph Learning (NeurIPS 2025 · similarity 0.86)
- Manifold Random Features (NeurIPS 2026 · similarity 0.86)
- Revisiting Random Walks for Learning on Graphs (ICLR 2025 · similarity 0.85)
- Non-convolutional graph neural networks. (NeurIPS 2024 · similarity 0.85)
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.
