Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2601.18111 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912848872996864 |
|---|---|
| author | Kossaifi, Jean Kovachki, Nikola Mardani, Morteza Leibovici, Daniel Ravuri, Suman Shokar, Ira Calvello, Edoardo Abbas, Mohammad Shoaib Harrington, Peter Subramaniam, Ashay Brenowitz, Noah Bonev, Boris Byeon, Wonmin Kreis, Karsten Durran, Dale Vahdat, Arash Pritchard, Mike Kautz, Jan |
| author_facet | Kossaifi, Jean Kovachki, Nikola Mardani, Morteza Leibovici, Daniel Ravuri, Suman Shokar, Ira Calvello, Edoardo Abbas, Mohammad Shoaib Harrington, Peter Subramaniam, Ashay Brenowitz, Noah Bonev, Boris Byeon, Wonmin Kreis, Karsten Durran, Dale Vahdat, Arash Pritchard, Mike Kautz, Jan |
| contents | The recent revolution in data-driven methods for weather forecasting has lead to a fragmented landscape of complex, bespoke architectures and training strategies, obscuring the fundamental drivers of forecast accuracy. Here, we demonstrate that state-of-the-art probabilistic skill requires neither intricate architectural constraints nor specialized training heuristics. We introduce a scalable framework for learning multi-scale atmospheric dynamics by combining a directly downsampled latent space with a history-conditioned local projector that resolves high-resolution physics. We find that our framework design is robust to the choice of probabilistic estimator, seamlessly supporting stochastic interpolants, diffusion models, and CRPS-based ensemble training. Validated against the Integrated Forecasting System and the deep learning probabilistic model GenCast, our framework achieves statistically significant improvements on most of the variables. These results suggest scaling a general-purpose model is sufficient for state-of-the-art medium-range prediction, eliminating the need for tailored training recipes and proving effective across the full spectrum of probabilistic frameworks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_18111 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Demystifying Data-Driven Probabilistic Medium-Range Weather Forecasting Kossaifi, Jean Kovachki, Nikola Mardani, Morteza Leibovici, Daniel Ravuri, Suman Shokar, Ira Calvello, Edoardo Abbas, Mohammad Shoaib Harrington, Peter Subramaniam, Ashay Brenowitz, Noah Bonev, Boris Byeon, Wonmin Kreis, Karsten Durran, Dale Vahdat, Arash Pritchard, Mike Kautz, Jan Machine Learning Artificial Intelligence The recent revolution in data-driven methods for weather forecasting has lead to a fragmented landscape of complex, bespoke architectures and training strategies, obscuring the fundamental drivers of forecast accuracy. Here, we demonstrate that state-of-the-art probabilistic skill requires neither intricate architectural constraints nor specialized training heuristics. We introduce a scalable framework for learning multi-scale atmospheric dynamics by combining a directly downsampled latent space with a history-conditioned local projector that resolves high-resolution physics. We find that our framework design is robust to the choice of probabilistic estimator, seamlessly supporting stochastic interpolants, diffusion models, and CRPS-based ensemble training. Validated against the Integrated Forecasting System and the deep learning probabilistic model GenCast, our framework achieves statistically significant improvements on most of the variables. These results suggest scaling a general-purpose model is sufficient for state-of-the-art medium-range prediction, eliminating the need for tailored training recipes and proving effective across the full spectrum of probabilistic frameworks. |
| title | Demystifying Data-Driven Probabilistic Medium-Range Weather Forecasting |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2601.18111 |