PulseAugur
EN
LIVE 18:30:48

New AI models improve weather nowcasting with quantile regression and transformers

Researchers have developed new deep learning approaches for precipitation nowcasting. One study reformulates training as a multi-quantile regression problem, improving deterministic forecasts and enabling better prediction of heavy rainfall. Another paper introduces FREUD, a rectified flow transformer model that enhances probabilistic forecasting by preserving uncertainty and allowing continuous updates. AI

IMPACT These advancements offer more accurate and reliable short-term weather predictions, crucial for risk management and operational planning.

RANK_REASON The cluster contains two research papers detailing new AI models and training methodologies for weather nowcasting.

Read on arXiv cs.AI →

AI-generated summary · Google Gemini · from 4 sources. How we write summaries →

COVERAGE [4]

  1. arXiv cs.AI TIER_1 English(EN) · Gijs van Nieuwkoop, Siamak Mehrkanoon ·

    Beyond MSE: Improving Precipitation Nowcasting with Multi-Quantile Regression

    arXiv:2605.30122v1 Announce Type: cross Abstract: Deep-learning precipitation nowcasting models are often optimized using pointwise losses such as mean squared error or mean absolute error, which can lead to overly smooth forecasts and poor representation of heavy rainfall. This …

  2. arXiv cs.AI TIER_1 English(EN) · Siamak Mehrkanoon ·

    Beyond MSE: Improving Precipitation Nowcasting with Multi-Quantile Regression

    Deep-learning precipitation nowcasting models are often optimized using pointwise losses such as mean squared error or mean absolute error, which can lead to overly smooth forecasts and poor representation of heavy rainfall. This study investigates whether the predictive performa…

  3. arXiv cs.CV TIER_1 English(EN) · Johannes Schusterbauer, Jannik Wiese, Nick Stracke, Timy Phan, Bj\"orn Ommer ·

    Probabilistic Precipitation Nowcasting with Rectified Flow Transformers

    arXiv:2605.31204v1 Announce Type: new Abstract: Accurate weather forecasts are essential across various domains and are safety-critical in extreme weather conditions. Compared to simulation-based forecasting, data-driven approaches show greater efficiency, enabling short-term, hi…

  4. arXiv cs.CV TIER_1 English(EN) · Björn Ommer ·

    Probabilistic Precipitation Nowcasting with Rectified Flow Transformers

    Accurate weather forecasts are essential across various domains and are safety-critical in extreme weather conditions. Compared to simulation-based forecasting, data-driven approaches show greater efficiency, enabling short-term, high-resolution nowcasting. In particular, diffusi…