Contrastive learning in building vision-language models for subsurface geoscience
The Leading Edge · 01.02.2025 · 44 (2) · 80-95
DOI: 10.1190/tle44020080.1 ↗Kurz gefasst
Fachartikel in der Zeitschrift The Leading Edge, erschienen am 01.02.2025. Verfasst von Anatoly Aseev und 2 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 37 Literaturangaben.
Abstract (Originalsprache)
This paper presents the potential of contrastive language-image pretraining (CLIP) in subsurface geoscience, both as a standalone tool for multimodal semantic search and within the broader context of vision-language model training. This paper focuses on the role of CLIP in developing subsurface-specific vision-language models and presents a novel workflow for the stochastic generation of meaningful subsurface image-text data sets, adaptable for any vision-language training. Using this workflow, a synthetic geologic and seismic image-caption data set are constructed to conduct training experiments with CLIP. A comparison of training strategies for subsurface classification and semantic search reveals that CLIP models using domain-specific pretrained encoders deliver superior accuracy compared with traditional CLIP pretraining with off-the-shelf encoders. Additionally, we introduce GeoGen, a machine learning tool designed to generate realistic geologic and seismic models from textual descriptions in which domain-trained CLIP models play a pivotal role. Evaluation of GeoGen's performance emphasizes the importance of selecting an appropriate CLIP training strategy to enhance vision-language fusion and improve model quality.
Zitieren
APA
Aseev, A., Huang, J., & Mishra, P. (2025). Contrastive learning in building vision-language models for subsurface geoscience. The Leading Edge, 44(2), 80-95. https://doi.org/10.1190/tle44020080.1
Vancouver
Aseev A, Huang J, Mishra P. Contrastive learning in building vision-language models for subsurface geoscience. The Leading Edge. 2025;44(2):80-95. doi:10.1190/tle44020080.1
BibTeX
@article{Aseev2025,
author = {Aseev, Anatoly and Huang, Jiayuan and Mishra, Priya},
title = {Contrastive learning in building vision-language models for subsurface geoscience},
journal = {The Leading Edge},
year = {2025},
volume = {44},
number = {2},
pages = {80--95},
doi = {10.1190/tle44020080.1}
}
EUVIMED