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.
Keywords
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Walt, E., Kofinas, M., Mücke, N., Gavves, E., & Coumou, D. (2026). Xaurora: Generative Weather Forecasting with Denoising Stochastic Interpolants from a Foundation Model Prior. https://omanscience.com/en/articles/xaurora-generative-weather-forecasting-with-denoising-stochastic-interpolants-from-a-foundation-model-prior
MLA 9
Walt, Eliot, et al. "Xaurora: Generative Weather Forecasting with Denoising Stochastic Interpolants from a Foundation Model Prior." https://omanscience.com/en/articles/xaurora-generative-weather-forecasting-with-denoising-stochastic-interpolants-from-a-foundation-model-prior.
Chicago (author–date)
Walt, Eliot, Miltiadis Kofinas, Nikolaj Mücke, Efstratios Gavves, and Dim Coumou. 2026. "Xaurora: Generative Weather Forecasting with Denoising Stochastic Interpolants from a Foundation Model Prior." https://omanscience.com/en/articles/xaurora-generative-weather-forecasting-with-denoising-stochastic-interpolants-from-a-foundation-model-prior.
Harvard
Walt, E., Kofinas, M., Mücke, N., Gavves, E. and Coumou, D. (2026) 'Xaurora: Generative Weather Forecasting with Denoising Stochastic Interpolants from a Foundation Model Prior', Available at: https://omanscience.com/en/articles/xaurora-generative-weather-forecasting-with-denoising-stochastic-interpolants-from-a-foundation-model-prior.
Vancouver
Walt E, Kofinas M, Mücke N, Gavves E, Coumou D. Xaurora: Generative Weather Forecasting with Denoising Stochastic Interpolants from a Foundation Model Prior. https://omanscience.com/en/articles/xaurora-generative-weather-forecasting-with-denoising-stochastic-interpolants-from-a-foundation-model-prior
IEEE
E. Walt, M. Kofinas, N. Mücke, E. Gavves, and D. Coumou, "Xaurora: Generative Weather Forecasting with Denoising Stochastic Interpolants from a Foundation Model Prior," https://omanscience.com/en/articles/xaurora-generative-weather-forecasting-with-denoising-stochastic-interpolants-from-a-foundation-model-prior.