Parsimonious Fine-Tuning of Scientific Foundation Models That Predict Time Series Evolution of Partial Differential Equations
Proceedings of the AAAI Symposium Series · 06.11.2026 · 10 (1) · 672-674
DOI: 10.1609/aaaiss.v10i1.43473 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Symposium Series, online erschienen am 06.11.2026. Verfasst von Christine Sweeney.
Abstract (Originalsprache)
We propose to give a hands-on tutorial that shows participants how to fine-tune AI-based scientific foundation models and measure the tradeoffs in training time, model size and dataset size. The scientific domain of the tutorial is partial differential equations (PDEs), where the PDE foundation models predict the time series evolution of data fields from initial conditions. Participants will use pretrained models and compare model quality resulting from different fine-tuning workflows.
Zitieren
APA
Sweeney, C. (2026). Parsimonious Fine-Tuning of Scientific Foundation Models That Predict Time Series Evolution of Partial Differential Equations. Proceedings of the AAAI Symposium Series, 10(1), 672-674. https://doi.org/10.1609/aaaiss.v10i1.43473
Vancouver
Sweeney C. Parsimonious Fine-Tuning of Scientific Foundation Models That Predict Time Series Evolution of Partial Differential Equations. Proceedings of the AAAI Symposium Series. 2026;10(1):672-674. doi:10.1609/aaaiss.v10i1.43473
BibTeX
@article{Sweeney2026,
author = {Sweeney, Christine},
title = {Parsimonious Fine-Tuning of Scientific Foundation Models That Predict Time Series Evolution of Partial Differential Equations},
journal = {Proceedings of the AAAI Symposium Series},
year = {2026},
volume = {10},
number = {1},
pages = {672--674},
doi = {10.1609/aaaiss.v10i1.43473}
}
EUVIMED