Swift: An Autoregressive Consistency Model for Efficient Weather Forecasting

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Stock, Jason, Arcomano, Troy, Kotamarthi, Rao
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911185479139328
author Stock, Jason
Arcomano, Troy
Kotamarthi, Rao
author_facet Stock, Jason
Arcomano, Troy
Kotamarthi, Rao
contents Diffusion models offer a physically grounded framework for probabilistic weather forecasting, but their typical reliance on slow, iterative solvers during inference makes them impractical for subseasonal-to-seasonal (S2S) applications where long lead-times and domain-driven calibration are essential. To address this, we introduce Swift, a single-step consistency model that, for the first time, enables autoregressive finetuning of a probability flow model with a continuous ranked probability score (CRPS) objective. This eliminates the need for multi-model ensembling or parameter perturbations. Results show that Swift produces skillful 6-hourly forecasts that remain stable for up to 75 days, running $39\times$ faster than state-of-the-art diffusion baselines while achieving forecast skill competitive with the numerical-based, operational IFS ENS. This marks a step toward efficient and reliable ensemble forecasting from medium-range to seasonal-scales.
format Preprint
id arxiv_https___arxiv_org_abs_2509_25631
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Swift: An Autoregressive Consistency Model for Efficient Weather Forecasting
Stock, Jason
Arcomano, Troy
Kotamarthi, Rao
Machine Learning
Diffusion models offer a physically grounded framework for probabilistic weather forecasting, but their typical reliance on slow, iterative solvers during inference makes them impractical for subseasonal-to-seasonal (S2S) applications where long lead-times and domain-driven calibration are essential. To address this, we introduce Swift, a single-step consistency model that, for the first time, enables autoregressive finetuning of a probability flow model with a continuous ranked probability score (CRPS) objective. This eliminates the need for multi-model ensembling or parameter perturbations. Results show that Swift produces skillful 6-hourly forecasts that remain stable for up to 75 days, running $39\times$ faster than state-of-the-art diffusion baselines while achieving forecast skill competitive with the numerical-based, operational IFS ENS. This marks a step toward efficient and reliable ensemble forecasting from medium-range to seasonal-scales.
title Swift: An Autoregressive Consistency Model for Efficient Weather Forecasting
topic Machine Learning
url https://arxiv.org/abs/2509.25631