mlsanityData to 7 Oct 2026

ScaGNN: a Graph Neural Network for Multiple Scattering Simulations

arXiv 2609.37509 ↗First posted 29 Sep 2026cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

The boundary element method (BEM) provides an efficient numerical framework for solving multiple scattering problems in unbounded homogeneous domains. By restricting the discretization to the domain boundaries, it substantially reduces computational complexity. The procedure first consists in determining the solution trace on the boundaries of the domain by solving a boundary integral equation. Then, the volumetric solution can be recovered at low computational cost using a boundary integral representation. As the first step of the BEM represents the main computational bottleneck, we present ScaGNN, a learning-based approach designed to approximate the solution trace. It relies on a graph neural network architecture that incorporates a dynamic adaptive edge sampling mechanism for selecting the most relevant interactions to model. Guided by intermediate predictions of expected error and edge length, this mechanism selects, at various stages of the forward pass, the most relevant distant interactions to model. The proposed method is tailored to achieve linear complexity with the number of nodes in the input graph. To train and evaluate our network, we present a benchmark consisting of several datasets with different types of multiple scattering problems. Our experiments show that our approach surpasses existing state-of-the-art learning-based methods on the considered tasks and investigate the generalization capabilities to settings with an increased number of obstacles and out-of-distribution obstacle shapes. github.com/LARIAD/ScaGNN

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 42 of its 66 references that are in the map. First posted on arXiv 29 Sep 2026.

Builds on

42 of 66 references.

References · 42

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  3. 2022FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awarenesscited by 3,611
  4. 2020Fourier Neural Operator for Parametric Partial Differential Equationscited by 2,806
  5. 2016Categorical Reparameterization with Gumbel-Softmaxcited by 2,708
  6. 2016The Concrete Distribution: A Continuous Relaxation of Discrete Random Variablescited by 1,135
  7. 2020Learning to Simulate Complex Physics with Graph Networkscited by 867
  8. 2020Learning Mesh-Based Simulation with Graph Networkscited by 840
  9. 2020Neural Operator: Graph Kernel Network for Partial Differential Equationscited by 722
  10. 2019DropEdge: Towards Deep Graph Convolutional Networks on Node Classificationcited by 679
  11. 2023Point Transformer V3: Simpler, Faster, Strongercited by 579
  12. 2021Understanding over-squashing and bottlenecks on graphs via curvaturecited by 471
30 more
  1. 2022Fourier Neural Operator with Learned Deformations for PDEs on General Geometriescited by 416
  2. 2020Multipole Graph Neural Operator for Parametric Partial Differential Equationscited by 343
  3. 2023GNOT: A General Neural Operator Transformer for Operator Learningcited by 321
  4. 2024Transolver: A Fast Transformer Solver for PDEs on General Geometriescited by 299
  5. 2023Geometry-Informed Neural Operator for Large-Scale 3D PDEscited by 293
  6. 2022Transformer for Partial Differential Equations' Operator Learningcited by 270
  7. 2018Learning Particle Dynamics for Manipulating Rigid Bodies, Deformable Objects, and Fluidscited by 219
  8. 2020Learning to Drop: Robust Graph Neural Network via Topological Denoisingcited by 127
  9. 2022MultiScale MeshGraphNetscited by 94
  10. 2025Transolver++: An Accurate Neural Solver for PDEs on Million-Scale Geometriescited by 89
  11. 2023DRew: Dynamically Rewired Message Passing with Delaycited by 81
  12. 2023Improved Operator Learning by Orthogonal Attentioncited by 58
  13. 2025Geometry Aware Operator Transformer as an Efficient and Accurate Neural Surrogate for PDEs on Arbitrary Domainscited by 56
  14. 2023Eagle: Large-Scale Learning of Turbulent Fluid Dynamics with Mesh Transformerscited by 54
  15. 2025AB-UPT: Scaling Neural CFD Surrogates for High-Fidelity Automotive Aerodynamics Simulations via Anchored-Branched Universal Physics Transformerscited by 49
  16. 2024CViT: Continuous Vision Transformer for Operator Learningcited by 47
  17. 2022Efficient Learning of Mesh-Based Physical Simulation with BSMS-GNNcited by 43
  18. 2024AROMA: Preserving Spatial Structure for Latent PDE Modeling with Local Neural Fieldscited by 38
  19. 2022DropMessage: Unifying Random Dropping for Graph Neural Networkscited by 29
  20. 2023Probabilistically Rewired Message-Passing Neural Networkscited by 26
  21. 2025Erwin: A Tree-based Hierarchical Transformer for Large-scale Physical Systemscited by 21
  22. 2023Learning Flexible Body Collision Dynamics with Hierarchical Contact Mesh Transformercited by 18
  23. 2022BI-GreenNet: Learning Green's functions by boundary integral networkcited by 13
  24. 2023Learning Only On Boundaries: a Physics-Informed Neural operator for Solving Parametric Partial Differential Equations in Complex Geometriescited by 12
  25. 2023Learning Adaptive Neighborhoods for Graph Neural Networkscited by 6
  26. 2023Multiscale Neural Operators for Solving Time-Independent PDEscited by 5
  27. 2025Linear Attention with Global Context: A Multipole Attention Mechanism for Vision and Physicscited by 5
  28. 2021Learnt Sparsification for Interpretable Graph Neural Networkscited by 4
  29. 2024Solving Partial Differential Equations in Different Domains by Operator Learning method Based on Boundary Integral Equationscited by 4
  30. 2024EvoMesh: Adaptive Physical Simulation with Hierarchical Graph Evolutionscited by 4

Built on by

Not cited yet.