SA-EMO: Structure-Aligned Encoder Mixture of Operators for Generalizable Full-waveform Inversion

TL;DR

SA-EMO introduces a novel architecture combining a structure-aligned encoder with a mixture of neural operators, significantly improving the accuracy and generalization of full-waveform inversion in complex geological settings.

Contribution

The paper proposes a new structure-aligned encoder and adaptive operator fusion mechanism for FWI, enhancing generalization and accuracy over existing single-architecture methods.

Findings

Achieves approximately 58.4% reduction in MAE compared to traditional methods.

Improves boundary resolution by about 10.3%.

Demonstrates effectiveness on benchmark datasets.

Abstract

Full-waveform inversion (FWI) can produce high-resolution subsurface models, yet it remains inherently ill-posed, highly nonlinear, and computationally intensive. Although recent deep learning and numerical acceleration methods have improved speed and scalability, they often rely on single CNN architectures or single neural operators, which struggle to generalize in unknown or complex geological settings and are ineffective at distinguishing diverse geological types. To address these issues, we propose a Structure-Aligned Encoder-Mixture-of-Operators (SA-EMO) architecture for velocity-field inversion under unknown subsurface structures. First, a structure-aligned encoder maps high-dimensional seismic wavefields into a physically consistent latent space, thereby eliminating spatio-temporal mismatch between the waveform and velocity domains, recovering high-frequency components, and enhancing feature generalization. Then, an adaptive routing mechanism selects and fuses multiple neural-operator experts, including spectral, wavelet, multiscale, and local operators, to predict the velocity model. We systematically evaluate our approach on the OpenFWI benchmark and the Marmousi2 dataset. Results show that SA-EMO significantly outperforms traditional CNN or single-operator methods, achieving an average MAE reduction of approximately 58.443% and an improvement in boundary resolution of about 10.308%. Ablation studies further reveal that the structure-aligned encoder, the expert-fusion mechanism, and the routing module each contribute markedly to the performance gains. This work introduces a new paradigm for efficient, scalable, and physically interpretable full-waveform inversion.