# DINOv3 as a Frozen Encoder for CRPS-Oriented Probabilistic Rainfall Nowcasting

arXiv: https://arxiv.org/abs/2511.10894 · HTML: https://mlsanity.com/paper/2511.10894/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 18 references that are in the map.
- First posted on arXiv 14 Nov 2025.

## Builds on (most cited first, 9 of 9)

- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation](https://mlsanity.com/paper/1606.06650/) (2016)
- [Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting](https://mlsanity.com/paper/1506.04214/) (2015)
- [Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics](https://mlsanity.com/paper/1705.07115/) (2017)
- [Skillful Precipitation Nowcasting using Deep Generative Models of Radar](https://mlsanity.com/paper/2104.00954/) (2021)
- [MetNet: A Neural Weather Model for Precipitation Forecasting](https://mlsanity.com/paper/2003.12140/) (2020)
- [Mitigating Negative Transfer in Multi-Task Learning with Exponential Moving Average Loss Weighting Strategies](https://mlsanity.com/paper/2211.12999/) (2022)
- [Super-resolution Probabilistic Rain Prediction from Satellite Data Using 3D U-Nets and EarthFormers](https://mlsanity.com/paper/2212.02998/) (2022)
- [Solving the Weather4cast Challenge via Visual Transformers for 3D Images](https://mlsanity.com/paper/2212.02456/) (2022)

## Abstract (arXiv)

This paper proposes a competitive and computationally efficient approach to probabilistic rainfall nowcasting. A video projector (V-JEPA Vision Transformer) associated to a lightweight probabilistic head is attached to a pre-trained satellite vision encoder (DINOv3-SAT493M) to map encoder tokens into a discrete empirical CDF (eCDF) over 4-hour accumulated rainfall. The projector-head is optimized end-to-end over the Ranked Probability Score (RPS). As an alternative, 3D-UNET baselines trained with an aggregate Rank Probability Score and a per-pixel Gamma-Hurdle objective are used. On the Weather4Cast 2025 benchmark, the proposed method achieved a promising performance, with a CRPS of 3.5102, which represents ≈ 26% in effectiveness gain against the best 3D-UNET.