ESPC-Transolver: An Equivalent-Surface Physics-Guided Coupling-Aware Transolver for 3D Multiple Scattering
Abstract
Cross-geometry generalization is particularly challenging because geometry-conditioned multiple-scattering surrogates must infer unseen objects' scattering properties from geometric features while accurately capturing complex, phase-sensitive interactions arising from repeated wave exchange. We introduce ESPC-Transolver, an equivalent-surface physics-guided, coupling-aware Transolver that separates local scattering-response representations from inter-object interaction modeling. For each sound-hard scatterer, canonical scattered-pressure responses are constructed under a shared set of spherical-wave excitations. A local Dirichlet-to-Neumann (DtN) map and modal scattering conversion express these responses on an enclosing sphere, providing physically consistent input channels across geometries. Shared object-local Transolver blocks encode these responses and incident-field traces, while relative-position-aware slice attention models inter-object coupling. Predicted interface pressure and normal flux recovered from the local operator define equivalent sources for reconstructing exterior near- and far-field scattering. We evaluate the method on three-dimensional acoustic multiple-scattering problems at a fixed frequency, comparing it with geometry-conditioned predictors and alternative neural network backbones. ESPC-Transolver achieves lower scattered-field prediction errors on held-out instances of familiar geometry families and generalizes effectively to previously unseen Peanut and Torus geometries. Exterior-field reconstructions reproduce multiple-scattering interference patterns, linking cross-geometry interface prediction to near-field pressure and far-field scattering recovery.
Related
- Classical and Quantum Logics with Multiple Non-Equivalent and Non-Isomorphic Models
- When Multiple-Unicast Feasibility Is Equivalent to Single-Sink Routability
- A common structure for modified scattering: Vlasov--Riesz, Hartree, and their coupling
- Uniqueness and stability for inverse potential scattering with multiple wavenumbers
- From multi-layered problems to multiple two-layered problems: a novel frequency-time hybrid multiple-scattering integral equation solver
BibTeX
@misc{zhao2026espctransolver,
title = {{ESPC-Transolver: An Equivalent-Surface Physics-Guided Coupling-Aware Transolver for 3D Multiple Scattering}},
author = {Guoqing Zhao and Liang Li and Xing Li and Hao Kuang},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.34246},
eprint = {2610.34246},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.34246},
}
RIS
TY - UNPB
TI - ESPC-Transolver: An Equivalent-Surface Physics-Guided Coupling-Aware Transolver for 3D Multiple Scattering
AU - Guoqing Zhao
AU - Liang Li
AU - Xing Li
AU - Hao Kuang
PY - 2026
DA - 2026/10/25
PB - arXiv
DO - 10.48550/arxiv.2610.34246
UR - https://arxiv.org/abs/2610.34246
AB - Cross-geometry generalization is particularly challenging because geometry-conditioned multiple-scattering surrogates must infer unseen objects' scattering properties from geometric features while accurately capturing complex, phase-sensitive interactions arising from repeated wave exchange. We introduce ESPC-Transolver, an equivalent-surface physics-guided, coupling-aware Transolver that separates local scattering-response representations from inter-object interaction modeling. For each sound-hard scatterer, canonical scattered-pressure responses are constructed under a shared set of spherical-wave excitations. A local Dirichlet-to-Neumann (DtN) map and modal scattering conversion express these responses on an enclosing sphere, providing physically consistent input channels across geometries. Shared object-local Transolver blocks encode these responses and incident-field traces, while relative-position-aware slice attention models inter-object coupling. Predicted interface pressure and normal flux recovered from the local operator define equivalent sources for reconstructing exterior near- and far-field scattering. We evaluate the method on three-dimensional acoustic multiple-scattering problems at a fixed frequency, comparing it with geometry-conditioned predictors and alternative neural network backbones. ESPC-Transolver achieves lower scattered-field prediction errors on held-out instances of familiar geometry families and generalizes effectively to previously unseen Peanut and Torus geometries. Exterior-field reconstructions reproduce multiple-scattering interference patterns, linking cross-geometry interface prediction to near-field pressure and far-field scattering recovery.
ER -