Research map / Probabilistic and scientific modeling / Differential equations and PDEs
Adaptive Mamba Neural Operators
Zeyuan Song, Zheyu Jiang
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 adaptive (input-dependent) poles instead of 32 static random poles (ablation) | AMO · Airfoil | relative L2 error: 0.0067 -> 0.0020 | Improved |
| use adaptive poles instead of static poles (ablation) | AMO · Darcy | relative L2 error: 0.0174 -> 0.0021 | Improved |
| use adaptive poles instead of static poles (ablation) | AMO · N-S | relative L2 error: 0.1103 -> 0.0278 | Improved |
| use adaptive poles instead of static poles (ablation) | AMO · Elasticity | relative L2 error: 0.0097 -> 0.0043 | Improved |
| use orthogonal TM bases instead of non-orthogonal kernels (ablation) | AMO · Airfoil | relative L2 error: 0.0083 -> 0.0020 | Improved |
| use orthogonal TM bases instead of non-orthogonal kernels (ablation) | AMO · Elasticity | relative L2 error: 0.0094 -> 0.0043 | Improved |
| use orthogonal TM bases instead of non-orthogonal kernels (ablation) | AMO · Training time per epoch | training time: 50.3% | Improved |
| increase number of adaptive poles from 4 to 32 (ablation) | AMO · Darcy | relative L2 error: 0.0057 -> 0.0021 | Improved |
| increase number of adaptive poles from 32 to 64 (ablation) | AMO · Airfoil | relative L2 error: 0.0020 -> 0.0036 | Hurt |
| use bidirectional SSM instead of unidirectional or multidirectional SSM (ablation) | AMO · seven benchmarks | relative L2 error | Improved |
| replace LaMO with AMO | LaMO · Airfoil | relative L2 error: 0.0041 -> 0.0020 | Improved |
| replace LaMO with AMO | LaMO · Pipe | relative L2 error: 0.0026 -> 0.0023 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Transolver — Transolver: A Fast Transformer Solver for PDEs on General Geometries (ICML 2024) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Latent Mamba Operator for Partial Differential Equations (ICML 2025 · similarity 0.91)
- Alias-Free Mamba Neural Operator (NeurIPS 2024 · similarity 0.87)
- Beyond Regular Grids: Fourier-Based Neural Operators on Arbitrary Domains (ICML 2024 · similarity 0.86)
- Maximum Update Parametrization and Zero-Shot Hyperparameter Transfer for Fourier Neural Operators (ICML 2025 · similarity 0.86)
- Quantitative Approximation for Neural Operators in Nonlinear Parabolic Equations (ICLR 2025 · similarity 0.86)
- Spectral-Refiner: Accurate Fine-Tuning of Spatiotemporal Fourier Neural Operator for Turbulent Flows (ICLR 2025 · 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.
