Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Improving Full Waveform Inversion in Large Model Era

Yinan Feng, Peng Jin, Yuzhe Guo, Yinpeng Chen, Youzuo Lin

Abstract

Full Waveform Inversion (FWI) is a highly nonlinear and ill-posed problem that aims to recover subsurface velocity maps from surface-recorded seismic waveforms data. Existing data-driven FWI typically uses small models, as available datasets have limited volume, geological diversity, and spatial extent, leading to substantial concerns about overfitting. Although they perform well on synthetic datasets, current methods fail to generalize to more realistic geological structures. In this work, we show that a model trained entirely on simulated and relatively simple data can generalize remarkably well to challenging and unseen geological benchmarks. We provide a working recipe that tames a billion-parameter model for FWI through coordinated scaling across three axes: model capacity, data diversity, and training strategy. Our model achieves state-of-the-art performance on OpenFWI and significantly narrows the generalization gap in data-driven FWI. Across six challenging geophysical benchmarks, including Marmousi, 2D SEG/EAGE Salt and Overthrust, 2004 BP, Sigsbee, and SEAM Phase I, it infers complex structures absent from the training set and delivers significant performance improvements (SSIM from 0.5844 to 0.7669). Overall, our results demonstrate that with an appropriate scaling strategy, large models trained on simple synthetic data can achieve substantial generalization to more complex and realistic geological structures.

Improving Full Waveform Inversion in Large Model Era

Abstract

Full Waveform Inversion (FWI) is a highly nonlinear and ill-posed problem that aims to recover subsurface velocity maps from surface-recorded seismic waveforms data. Existing data-driven FWI typically uses small models, as available datasets have limited volume, geological diversity, and spatial extent, leading to substantial concerns about overfitting. Although they perform well on synthetic datasets, current methods fail to generalize to more realistic geological structures. In this work, we show that a model trained entirely on simulated and relatively simple data can generalize remarkably well to challenging and unseen geological benchmarks. We provide a working recipe that tames a billion-parameter model for FWI through coordinated scaling across three axes: model capacity, data diversity, and training strategy. Our model achieves state-of-the-art performance on OpenFWI and significantly narrows the generalization gap in data-driven FWI. Across six challenging geophysical benchmarks, including Marmousi, 2D SEG/EAGE Salt and Overthrust, 2004 BP, Sigsbee, and SEAM Phase I, it infers complex structures absent from the training set and delivers significant performance improvements (SSIM from 0.5844 to 0.7669). Overall, our results demonstrate that with an appropriate scaling strategy, large models trained on simple synthetic data can achieve substantial generalization to more complex and realistic geological structures.
Paper Structure (25 sections, 10 equations, 4 figures, 4 tables)

This paper contains 25 sections, 10 equations, 4 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Methodology
  3. Baseline: Auto-regression Model
  4. Task formulation.
  5. Tokenization.
  6. Input Sequence construction.
  7. Network backbone and causal decoding.
  8. Macro Design
  9. Data Augmentation.
  10. Non-Causal Sequence Modeling.
  11. ViT-VQGAN Tokenizer.
  12. Post-Training Enhancements
  13. Policy Optimization with Velocity-Level Rewards.
  14. Latent-Space Gradient Refinement.
  15. Experiments
  16. ...and 10 more sections

Figures (4)

  • Figure 1: Reconstruction results on challenging realistic benchmarks. BigFWI tends to collapse toward over-smoothed, mean-shaped solutions that miss key interfaces and salt bodies. In contrast, our model produces sharper boundaries and geologically meaningful structures.
  • Figure 2: Overview of progressive model enhancements and final reconstruction quality. (a) Step-wise performance improvement as each component is introduced—from baseline to data augmentation, non-causal modeling, ViT VQGAN (without bottleneck), reinforcement learning alignment, and latent gradient refinement. (b) Illustration of the final reconstruction results, showing sharp structures and fine geological details across diverse scenarios.
  • Figure 3: Qualitative comparison of reconstructed velocity maps.
  • Figure 4: Qualitative comparison of reconstructed velocity maps. Our method produces sharper structures and finer structures across complex regions.