SeisLM: A Foundation Model for Seismic Waveforms
Abstract
We introduce the Seismic Language Model (SeisLM), a foundational model designed to analyze seismic waveforms---signals generated by Earth's vibrations such as the ones originating from earthquakes. SeisLM is pretrained on a large collection of open-source seismic datasets using a self-supervised contrastive loss, akin to BERT in language modeling. This approach allows the model to learn general seismic waveform patterns from unlabeled data without being tied to specific downstream tasks. When fine-tuned, SeisLM excels in seismological tasks like event detection, phase-picking, onset time regression, and foreshock--aftershock classification.
Cite
Text
Liu et al. "SeisLM: A Foundation Model for Seismic Waveforms." NeurIPS 2024 Workshops: FM4Science, 2024.Markdown
[Liu et al. "SeisLM: A Foundation Model for Seismic Waveforms." NeurIPS 2024 Workshops: FM4Science, 2024.](https://mlanthology.org/neuripsw/2024/liu2024neuripsw-seislm/)BibTeX
@inproceedings{liu2024neuripsw-seislm,
title = {{SeisLM: A Foundation Model for Seismic Waveforms}},
author = {Liu, Tianlin and Münchmeyer, Jannes and Laurenti, Laura and Marone, Chris and de Hoop, Maarten V. and Dokmanić, Ivan},
booktitle = {NeurIPS 2024 Workshops: FM4Science},
year = {2024},
url = {https://mlanthology.org/neuripsw/2024/liu2024neuripsw-seislm/}
}