SeisMark: A Large-Scale Open Benchmark for Robust 3D Seismic Fault Detection

Min Jun Park, Joseph Stitt, Robert Graham Clapp, Ilan Naiman, Artem Goncharuk, Kevin F. Smith
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:97199-97212, 2026.

Abstract

We introduce SeisMark, a large-scale open benchmark designed to bridge the gap between verifiable ground truth and realistic texture in 3D seismic fault detection. Using a novel pipeline merging procedural geology with diffusion-based synthesis, we produce domain-realistic (survey-specific) textured volumes that expose significant brittleness in existing models masked by simplified physics data. Experiments demonstrate that SeisMark acts as a rigorous discriminator, distinguishing robust modern architecture from legacy model that suffers performance collapse under realistic domain shifts. We release this benchmark to the community to serve as a verifiable standard for developing trustworthy, deployment-ready AI for safety-critical subsurface applications.

Cite this Paper


BibTeX
@InProceedings{pmlr-v306-park26y, title = {{S}eis{M}ark: A Large-Scale Open Benchmark for Robust 3{D} Seismic Fault Detection}, author = {Park, Min Jun and Stitt, Joseph and Clapp, Robert Graham and Naiman, Ilan and Goncharuk, Artem and Smith, Kevin F.}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {97199--97212}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/park26y/park26y.pdf}, url = {https://proceedings.mlr.press/v306/park26y.html}, abstract = {We introduce SeisMark, a large-scale open benchmark designed to bridge the gap between verifiable ground truth and realistic texture in 3D seismic fault detection. Using a novel pipeline merging procedural geology with diffusion-based synthesis, we produce domain-realistic (survey-specific) textured volumes that expose significant brittleness in existing models masked by simplified physics data. Experiments demonstrate that SeisMark acts as a rigorous discriminator, distinguishing robust modern architecture from legacy model that suffers performance collapse under realistic domain shifts. We release this benchmark to the community to serve as a verifiable standard for developing trustworthy, deployment-ready AI for safety-critical subsurface applications.} }
Endnote
%0 Conference Paper %T SeisMark: A Large-Scale Open Benchmark for Robust 3D Seismic Fault Detection %A Min Jun Park %A Joseph Stitt %A Robert Graham Clapp %A Ilan Naiman %A Artem Goncharuk %A Kevin F. Smith %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-park26y %I PMLR %P 97199--97212 %U https://proceedings.mlr.press/v306/park26y.html %V 306 %X We introduce SeisMark, a large-scale open benchmark designed to bridge the gap between verifiable ground truth and realistic texture in 3D seismic fault detection. Using a novel pipeline merging procedural geology with diffusion-based synthesis, we produce domain-realistic (survey-specific) textured volumes that expose significant brittleness in existing models masked by simplified physics data. Experiments demonstrate that SeisMark acts as a rigorous discriminator, distinguishing robust modern architecture from legacy model that suffers performance collapse under realistic domain shifts. We release this benchmark to the community to serve as a verifiable standard for developing trustworthy, deployment-ready AI for safety-critical subsurface applications.
APA
Park, M.J., Stitt, J., Clapp, R.G., Naiman, I., Goncharuk, A. & Smith, K.F.. (2026). SeisMark: A Large-Scale Open Benchmark for Robust 3D Seismic Fault Detection. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:97199-97212 Available from https://proceedings.mlr.press/v306/park26y.html.

Related Material