Preprint · 2026 · Open access

Xaurora: Generative Weather Forecasting with Denoising Stochastic Interpolants from a Foundation Model Prior

Eliot Walt, Miltiadis Kofinas, Nikolaj T. Mücke, Efstratios Gavves, Dim Coumou

arXiv (Cornell University) · Published 5 October 2026 · DOI 10.48550/arxiv.2610.06509

Free PDFView at the publisherCite this paper

In brief

“Xaurora: Generative Weather Forecasting with Denoising Stochastic Interpolants from a Foundation Model Prior” is a 2026 preprint by Eliot Walt and 4 co-authors, published in arXiv (Cornell University). OpenAlex counts 0 citations as of 15 November 2026. It is open access under a CC BY licence: the full text is free to read at arXiv (Cornell University).

Citations
0
References
50
Authors
5

Abstract

Deep learning has revolutionised weather forecasting in recent years, especially through atmospheric foundation models, which offer competitive skill for a fraction of the computational costs of classic physics-based models. However, most existing foundation models are deterministic, limiting the generation of large ensembles for accurate uncertainty quantification, extreme weather risk assessment, and long-range weather forecasting. Furthermore, these models incur a large, often prohibitive, computational overhead to train from scratch. To address these shortcomings, we turn a pretrained deterministic prior model, namely the Aurora foundation model, into a generative ensemble-prediction model. To that end, we introduce a novel generative method, Denoising Stochastic Interpolants, combined with a replay buffer for Stochastic Differential Equation (SDE) rollout, enabling probabilistic training of SDE trajectories. Our stochastic foundation model, Xaurora, is finetuned from the small Aurora version, yet it approaches the state-of-the-art on global ensemble metrics and is competitive with the large version of Aurora. Our method is parameter and sample efficient, and generates skilful 15-day forecasts in 13 minutes. Our results demonstrate that deterministic foundation models can be efficiently extended into even stronger stochastic models.

Cite this paper

Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.

APA 7

Walt, E., Kofinas, M., Mücke, N. T., Gavves, E., & Coumou, D. (2026). Xaurora: Generative Weather Forecasting with Denoising Stochastic Interpolants from a Foundation Model Prior. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2610.06509

MLA 9

Walt, Eliot, et al. “Xaurora: Generative Weather Forecasting with Denoising Stochastic Interpolants from a Foundation Model Prior.” arXiv (Cornell University), 2026. https://doi.org/10.48550/arxiv.2610.06509.

Chicago

Walt, Eliot, Miltiadis Kofinas, Nikolaj T. Mücke, Efstratios Gavves, and Dim Coumou. 2026. “Xaurora: Generative Weather Forecasting with Denoising Stochastic Interpolants from a Foundation Model Prior.” arXiv (Cornell University). https://doi.org/10.48550/arxiv.2610.06509.

Harvard

Walt, E. et al. (2026) ‘Xaurora: Generative Weather Forecasting with Denoising Stochastic Interpolants from a Foundation Model Prior’, arXiv (Cornell University). doi:10.48550/arxiv.2610.06509.

IEEE

E. Walt, M. Kofinas, N. T. Mücke, E. Gavves, and D. Coumou, “Xaurora: Generative Weather Forecasting with Denoising Stochastic Interpolants from a Foundation Model Prior,” arXiv (Cornell University), 2026, doi: 10.48550/arxiv.2610.06509.

BibTeX

@misc{walt2026xaurora,
  title     = {{Xaurora: Generative Weather Forecasting with Denoising Stochastic Interpolants from a Foundation Model Prior}},
  author    = {Walt, Eliot and Kofinas, Miltiadis and Mücke, Nikolaj T. and Gavves, Efstratios and Coumou, Dim},
  howpublished = {arXiv (Cornell University)},
  year      = {2026},
  doi       = {10.48550/arxiv.2610.06509},
  url       = {https://doi.org/10.48550/arxiv.2610.06509}
}

Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)

Topic and keywords

Physical Sciences › Earth and Planetary Sciences › Atmospheric Science › Meteorological Phenomena and Simulations

  • weather forecasting
  • stochastic differential equations
  • probabilistic forecasting
  • uncertainty quantification
  • long-term forecasting
  • replay buffer
  • parameter-efficient fine-tuning
  • sample efficiency

Work with this paper

Most-cited papers in Earth and Planetary Sciences

All highly cited Earth and Planetary Sciences papers

Questions about this paper

How do I cite this paper in APA style?

Walt, E., Kofinas, M., Mücke, N. T., Gavves, E., & Coumou, D. (2026). Xaurora: Generative Weather Forecasting with Denoising Stochastic Interpolants from a Foundation Model Prior. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2610.06509

How many times has this paper been cited?

0 times, according to OpenAlex on 15 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.

Is this paper free to read?

Yes. It is open access: read the free full text at arXiv (Cornell University).

Who wrote this paper?

Eliot Walt, Miltiadis Kofinas, Nikolaj T. Mücke, Efstratios Gavves, Dim Coumou.

What is the DOI of this paper?

10.48550/arxiv.2610.06509. The permanent link is https://doi.org/10.48550/arxiv.2610.06509.

Bibliographic data from OpenAlex (CC0), last checked 15 November 2026. ResearchGuru does not host this paper; the links go to the publisher or the repository that does.