8 papers in the last 30 days, 0 in the 30 days before (New).
Value of Adding AI Weather Model Ensembles to a Multi-model Probabilistic Blend (opens in a new tab)
arXiv - Atmospheric physics + Materials
Many weather forecasting organizations use a multi-model blend of ensemble members from different prediction systems to obtain a more accurate forecast. Recent studies have shown that blending deterministic AI-based models alongside physics-based models improves the overall forecast skill. However, the incorporation of an ensemble AI-based model into a multi-model blend is yet to be investigated.
arXiv - AI / ML / NLP
Data Assimilation (DA) reconstructs the state of a spatio-temporal system by combining sensor observations with the output of a numerical model. Since model states are high-dimensional while observations cover only a small fraction of the domain, DA is a challenging large-scale estimation problem, central to weather forecasting and oceanography. Classical variational methods such as 3D-Var rely on
Self-Supervised Latent Action-Conditioned World Modeling for Time Series (opens in a new tab)
arXiv - AI / ML / NLP
Time-series data play an important role in various tasks, such as weather forecasting, web traffic analysis, and speech recognition. Recently, pioneering studies have increasingly adopted Joint-Embedding Predictive Architecture (JEPA) based encoders to learn time-series representations through latent-space prediction. However, they often overlook the underlying factors governing transitions betwee
FASST: A Compact Stateless Atmosphere Model for Long-Term Ocean Coupling (opens in a new tab)
arXiv - AI / ML / NLP
Coupled climate simulations are computationally expensive, largely because of the atmosphere. We introduce FASST, a coupled climate model that retains the prognostic POP2 ocean while replacing the atmosphere and sea ice with lightweight U-Net emulators. The atmospheric emulator is diagnostic, predicting surface fluxes from the evolving ocean-ice state every $6$ simulated hours, while the sea-ice e
arXiv - Atmospheric physics + Materials
Improving convective-scale skill in data-driven regional weather models requires better representing rare high-impact events, training at native resolution, broadening variable output, and demonstrating performance in real-time settings. Here, we evaluate these challenges using HRRRCast, a data-driven convection-allowing emulator trained on the High-Resolution Rapid Refresh (HRRR) analysis data. H
Climate downscaling with likelihood functional flow matching (opens in a new tab)
arXiv - AI / ML / NLP
arXiv:2610.33482v2 Announce Type: new Abstract: Decisions about heat, frost and flooding are made locally, yet reanalyses and climate models resolve tens of kilometres. Generative downscalers add kilometre-scale detail with uncertainty, but are tied to a grid, costly to sample and lack a tractable likelihood. Here we introduce CD-FFM, a functional flow-matching model that treats the high-resolutio
AliO: Output Alignment Matters in Long-Term Time Series Forecasing (opens in a new tab)
arXiv - AI / ML / NLP
arXiv:2610.11213v2 Announce Type: replace Abstract: Long-term Time Series Forecasting (LTSF) tasks, which leverage the current data sequence as input to predict the future sequence, have become increasingly crucial in real-world applications such as weather forecasting and planning of electricity consumption. However, state-of-the-art LTSF models often fail to achieve prediction output alignment f
arXiv - AI / ML / NLP
arXiv:2610.35545v1 Announce Type: new Abstract: High-resolution probabilistic weather forecasting remains computationally demanding. Here we present Astra, a 0.25$^\circ$ ensemble model trained on eight H200 GPUs. It combines Kantorovich Attention, a two-sided transport attention in which each region chooses where it reads from and writes to, so every region sends and receives at token resolution,