Research map / Graph learning / Graph algorithms and clustering
Collapsed Effective Operators for Higher-order Structures
Maximilian Krahn, Lennart Bastian, Vikas K Garg, Björn Schuller, Tolga Birdal
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 |
|---|---|---|---|
| replace graph Laplacian with collapsed operator S_eps | spectral clustering · Topotein HKS clustering | accuracy: 46.9% -> 70.9% | Improved |
| replace graph Laplacian with collapsed operator S_eps | normalized cuts · SBM Dense-Cliques | clustering accuracy: +3% | Improved |
| replace graph Laplacian with collapsed operator S_eps | normalized cuts · SBM Unequal-Sizes | clustering accuracy: +6% | Improved |
| replace graph Laplacian with collapsed operator S_eps | normalized cuts · College Football | clustering accuracy: +3% | Improved |
| replace graph Laplacian with collapsed operator S_eps | normalized cuts · Les Miserables | clustering accuracy: +2% | Improved |
| replace graph Laplacian with collapsed operator S_eps | normalized cuts · Karate | clustering accuracy: 0.97 -> 0.97 | No clear effect |
| replace graph Laplacian with collapsed operator S_eps | normalized cuts · SBM Planted-Clique | clustering accuracy: .96 -> .96 | No clear effect |
| replace graph Laplacian with collapsed operator S_eps | manifold denoising · random geometric graphs, 200 shortcut edges | MSE: 0.099 -> 0.073 | Improved |
| use collapsed operator spectrum instead of graph Laplacian spectrum | MLP classifier · MANTRA beta1 | accuracy: 0.92 -> 0.95 | Improved |
| use collapsed operator spectrum instead of multi-order Laplacian spectrum | MLP classifier · MANTRA beta1 | accuracy: 0.97 -> 0.95 | Hurt |
| replace Laplacian PE with SchurPE | GraphTransformer · protein node classification Contact | accuracy: 0.551 -> 0.573 | Improved |
| replace Laplacian PE with SchurPE | GCN · protein node classification ResType | accuracy: 0.582 -> 0.666 | Improved |
Closest papers by meaning
- Higher-Order Graph Convolutional Network with Flower-Petals Laplacians on Simplicial Complexes (AAAI 2024 · similarity 0.85)
- GIST: Gauge-Invariant Spectral Transformers for Scalable Graph Neural Operators (NeurIPS 2026 · similarity 0.84)
- Dynamics of the Transformer Residual Stream: Coupling Spectral Geometry to Network Topology (NeurIPS 2026 · similarity 0.84)
- Schur Nets: exploiting local structure for equivariance in higher order graph neural networks (NeurIPS 2024 · similarity 0.84)
- Topological Blindspots: Understanding and Extending Topological Deep Learning Through the Lens of Expressivity (ICLR 2025 · similarity 0.83)
- MANTRA: The Manifold Triangulations Assemblage (ICLR 2025 · similarity 0.83)
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.
