In-Context Learning for Seismic Data Processing

Seismic processing transforms raw data into subsurface images essential for geophysical applications. Traditional methods face challenges, such as noisy data, and manual parameter tuning, among others. Recently deep learning approaches have proposed alternative solutions to some of these problems. However, important challenges of existing deep learning approaches are spatially inconsistent results across neighboring seismic gathers and lack of user-control. We address these limitations by introducing ContextSeisNet, an in-context learning model, to seismic demultiple processing. Our approach conditions predictions on a support set of spatially related example pairs: neighboring common-depth point gathers from the same seismic line and their corresponding labels. This allows the model to learn task-specific processing behavior at inference time by observing how similar gathers should be processed, without any retraining. This method provides both flexibility through user-defined examples and improved lateral consistency across seismic lines. On synthetic data, ContextSeisNet outperforms a U-Net baseline quantitatively and demonstrates enhanced spatial coherence between neighboring gathers. On field data, our model achieves superior lateral consistency compared to both traditional Radon demultiple and the U-Net baseline. Relative to the U-Net, ContextSeisNet also delivers improved near-offset performance and more complete multiple removal. Notably, ContextSeisNet achieves comparable field data performance despite being trained on 90% less data, demonstrating substantial data efficiency. These results establish ContextSeisNet as a practical approach for spatially consistent seismic demultiple with potential applicability to other seismic processing tasks.

💡 Research Summary

**
The paper introduces ContextSeisNet, the first application of In‑Context Learning (ICL) to seismic data processing, specifically to the problem of multiple attenuation (demultiple). Traditional supervised deep‑learning models for demultiple treat each common‑depth‑point (CDP) gather independently, which often leads to lateral inconsistencies—different neighboring gathers may be processed in contradictory ways. Moreover, these models require extensive hyper‑parameter tuning and lack user‑level control.

ContextSeisNet addresses these issues by conditioning each prediction on a support set V that contains spatially adjacent CDP gathers from the same seismic line together with their corresponding labels. The labels are generated by any conventional demultiple algorithm (e.g., Radon transform), allowing the model to inherit domain knowledge while remaining adaptable at inference time. During inference the model receives the query gather X