physics.geo-phAug 18, 2026

SeisEvo: Evolution of Seismic Data Reconstruction Algorithms by Agents

Authors: Yingjie Xu, Siwei Yu, Jianwei Ma

Organizations: School of Mathematics and Center of Geophysics, Harbin Institute of Technology, Harbin, Heilongjiang, China · Department of Mathematics, National University of Singapore, Singapore · School of Earth and Space Sciences, Peking University, Beijing, China

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Classical seismic data reconstruction relies on manually designed structural priors and iterative operators, whose coupled design space is far larger than manual trial and error can explore systematically. Deep-learning methods encode the reconstruction rules in learned weights rather than in an explicit operator that can be inspected and modified. We propose SeisEvo (Seismic Algorithm Evolution), which does not optimize a single reconstruction result but searches for the algorithm that produces it. Starting from a classical reconstruction algorithm, an LLM-driven multi-agent search modifies only the components that the user has opened for editing, without prescribing the mechanism to be discovered. Candidates that violate the physical constraints of the task are rejected outright, and the remaining ones are scored by execution. The output is neither an agent system nor a neural network, but a standalone white-box algorithm that requires no agent or neural network at inference time. For interpolation without added noise, the search discovered a residual-gated, phase-aligned dip-consistency projection; Evo-POCS improves the SNR over classic POCS by 3.49 dB on average across missing ratios from 30% to 70%. For simultaneous interpolation and denoising, it discovered a reliability-grouped singular-value shrinkage; Evo-MSSA improves the average reconstruction SNR by more than 7 dB over classic MSSA and by more than 3 dB over a stronger rank-reduction baseline. Both operators retain their gains on data not used during the search. To the best of our knowledge, this is the first study to formulate the design of a seismic reconstruction operator as a constrained, LLM-driven program evolution task. Agentic algorithm evolution can thus complement deep learning in discovering explicit, inspectable, and deployable seismic processing algorithms.

Figures & tables

Figure 1: The SeisEvo framework. A one-time task instantiation (seed algorithm and specification) is followed by a closed loop of candidate generation, gating and scoring, and evidence binding with context update, repeated for TT rounds. The best candidate is returned as a standalone white-box operator. External validation, including full-volume generalization and robustness tests under different sampling and noise conditions, is performed once after the search and is not part of the loop.
Figure 2: Search-time seismic sub-blocks extracted from complete 3D volumes.
Figure 3: Search-time objective OPOCSO_{\mathrm{POCS}} over 20 search rounds. The score rises from 0.713 at the classic POCS seed to 0.909 at round 14 and remains unchanged afterwards.
Figure 4: Reconstruction comparison on a 64×64×6464\times 64\times 64 synthetic sub-block under 50% random sampling. (a) Original complete data; (b) POCS reconstruction; (c) Evo-POCS reconstruction; (d) 50% randomly sampled data; (e) POCS error; (f) Evo-POCS error.
Figure 5: Reconstruction comparison on a 256×120×64256\times 120\times 64 field dataset under 50% random sampling. (a) Original; (b) POCS; (c) Evo-POCS; (d) sampled data; (e) POCS error; (f) Evo-POCS error. The Evo-POCS reconstruction error is smaller than that of classic POCS.
Figure 6: Reconstruction and FF - KK comparison on a slice at Y=0.65Y=0.65 km. (a)–(d) original, sampled, POCS, and Evo-POCS reconstructions; (e)–(h) their FF - KK spectra. Evo-POCS gives better event continuity and an FF - KK spectrum closer to the original.
Missing ratio (%)POCS (dB)Evo-POCS (dB)Δ\Delta SNR (dB)
309.6215.04+ 5.42
409.4813.73+ 4.25
509.2312.79+ 3.56
608.8811.52+ 2.64
708.419.99+ 1.58
Average9.1212.61+ 3.49
Table 1: SNR comparison between classic POCS and Evo-POCS on the field dataset under different missing-trace ratios. Δ\Delta SNR is computed as SNREvo-POCS−SNRPOCS\mathrm{SNR}_{\mathrm{Evo\text{-}POCS}}-\mathrm{SNR}_{\mathrm{POCS}} .
Figure 7: Search-time objective OMSSAO_{\mathrm{MSSA}} over 20 search rounds. The score increases from 0.320 for MSSA to 0.563 for the best candidate.
Figure 8: Reconstruction comparison on a 64×64×6464\times 64\times 64 synthetic sub-block with 50% randomly missing traces and Gaussian noise. (a) Original complete data; (b) MSSA reconstruction; (c) Evo-MSSA reconstruction; (d) noisy data with 50% randomly missing traces; (e) MSSA error; (f) Evo-MSSA error.
Figure 9: Reconstruction comparison on a 256×120×64256\times 120\times 64 field dataset with 50% randomly missing traces and Gaussian noise. (a) Noisy complete data; (b) MSSA reconstruction; (c) Evo-MSSA reconstruction; (d) noisy incomplete data; (e) MSSA error; (f) Evo-MSSA error.
Noise scale α\alphaMSSA (dB)ODRR (dB)Evo-MSSA (dB)Gain over ODRR (dB)
0.0110.8812.6114.49+1.88+1.88
0.025.778.8412.05+3.21+3.21
0.032.587.1310.32+3.19+3.19
0.040.235.508.96+3.46+3.46
0.05-1.712.857.70+4.85+4.85
Average3.557.3910.70+3.31+3.31
Table 2: Overall reconstruction SNR on the field dataset under different noise scales.
Figure 10: Reconstruction comparison on a 256×120×64256\times 120\times 64 field dataset at noise scale α=0.05\alpha=0.05 with 50% randomly missing traces. (a) Noisy complete data; (b) MSSA reconstruction; (c) ODRR reconstruction; (d) Evo-MSSA reconstruction; (e) noisy sampled data; (f) MSSA error; (g) ODRR error; (h) Evo-MSSA error.
Figure 11: Slice and FF – KK spectrum comparison on a 256×120×64256\times 120\times 64 field dataset at Y=1.75Y=1.75 km, with noise scale α=0.05\alpha=0.05 and 50% randomly missing traces. (a) Original; (b) noisy sampled data; (c) MSSA; (d) ODRR; (e) Evo-MSSA; (f)–(j) the corresponding FF – KK spectra of (a)–(e), respectively.
More in this topic

Automated Algorithm Discovery

Automated algorithm discovery develops methods that generate and test executable algorithms for optimization, science, software, and machine learning.

See all 170 papers

Explore similar work

CardsList
Mar 5, 2026

Autonomous Algorithm Discovery for Ptychography via Evolutionary LLM Reasoning

Ptychography is a computational imaging technique widely used for high-resolution materials characterization, but high-quality reconstructions often require the use of regularization functions that largely remain manually designed. We introduce Ptychi-Evolve, an autonomous framework that uses large language models (LLMs) to discover and evolve novel regularization algorithms. The framework combines LLM-driven code generation with evolutionary mechanisms, including semantically-guided crossover and mutation. Experiments on three challenging datasets (X-ray integrated circuits, low-dose electron microscopy of apoferritin, and multislice imaging with crosstalk artifacts) demonstrate that discovered regularizers outperform conventional reconstructions, achieving up to +0.26 SSIM and +8.3~dB PSNR improvements. Besides, Ptychi-Evolve records algorithm lineage and evolution metadata, enabling interpretable and reproducible analysis of discovered regularizers.

Xiangyu Yin, Ming Du, Junjing Deng +3

Advanced Photon Source, Argonne National Laboratory, Lemont, IL, USA · Department of Materials Science and NanoEngineering, Rice University, Houston, TX, USA

Image ReconstructionAutomated Algorithm DiscoveryComputational Imaging
AskPDF
Aug 14, 2026

Consensus-gated Multi-Agent Neural Architecture Search for Seismic Fault Segmentation

Neural networks for seismic fault segmentation are often borrowed from computer vision and medical imaging domains where they train under relatively much larger labeled data resources. Optimizing their architecture under tight labeled data budgets as are common in geophysical applications is not a trivial problem. Manually designing data-optimal architectures is time-consuming while classical neural architecture search (NAS) is restricted to hand-crafted search spaces and large compute budgets. We present an agentic NAS system in which a panel of three large language models (Claude, GPT-5.1, and Gemini2.5Pro) debates each candidate architecture to unanimous consensus, authors the complete PyTorch implementation, cross-reviews it, and submits it to an automated validate-train-score loop with a hard 450K parameter budget, keep-or-revert lineage, and a memory of failed mechanisms. Operating on source code rather than a predefined operation menu, the search ran on a single consumer GPU and trained only eight candidates. It discovered \ours{}: a 425K-parameter encoder-decoder with a strip-pooling bottleneck, squeeze-and-excitation gating, an asymmetric one-conv decoder, and a feature-pyramid fusion neck. Trained under a protocol identical to all baselines on sections derived from the Thebe fault dataset, it attains the highest F1 (0.578) and IoU of all models tested while being the smallest, outperforming a published-capacity U-Net (31M parameters, F1 0.484), DeepLabV3-ResNet50 (39.6M, 0.516), an Attention U-Net(1.83M, 0.502). The search cost 101 LLM calls (∼\sim1.15M input / 0.39M output tokens) and roughly one GPU-day, making consensus-gated LLM panels a practical, low-cost route to domain-specific architecture discovery.

Shehram Baig, Ahmad Mustafa

Information Technology University (ITU), Lahore, Pakistan · King Fahd University of Petroleum and Minerals, Dhahran, Saudi Arabia

Neural Architecture SearchLarge Language Model-Guided NASSeismology
AskPDF
May 19, 2026

OpenSeisML: Open Large-Scale Real Seismic and well-log Dataset for Generative AI

The advent of machine learning (ML) and computer vision has significantly accelerated seismic inversion workflows by reducing the computational cost of traditionally expensive iterative methods. However, the development and evaluation of ML methods remain limited by the scarcity of realistic velocity models, as most high-quality data are privately owned by oil and gas companies. To address this gap, we present OpenSeisML, a collection of real seismic datasets designed to support generative AI (Gen-AI) workflows for seismic inversion. The datasets are curated from publicly available surveys in the UK National Data Repository (NDR). When seismic volumes are in the time domain and wells are in depth, a time-to-depth conversion is required. We use checkshot data to establish the time-depth relationship and construct a velocity model through interpolation for accurate conversion of post-stack seismic data. Here, we present an automated data curation pipeline that enables seismic data preparation while ensuring reproducibility. The objective is to train a generative model that captures the statistical distribution of subsurface properties, enabling the synthesis of multiple statistically consistent realizations for uncertainty quantification which can act as a prior for seismic inversion.

Ipsita Bhar, Huseyin Tuna Erdinc, Thales Souza +2

Georgia Institute of Technology, Atlanta, USA · Osokey Ltd, Henley-on-Thames, UK

Synthetic Data GenerationTraining Data CurationSeismologyInverse Problems
AskPDF