Research map / Image generation / Fast diffusion sampling
FlowCast: Advancing Precipitation Nowcasting with Conditional Flow Matching
Bernardo Perrone Ribeiro, Jana Faganeli Pucer
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace diffusion objective (DDIM 100 steps) with CFM (1 step) (ablation) | FlowCast backbone · SEVIR 10% test | CRPS: 0.0208 -> 0.0207 | Improved |
| replace diffusion objective (DDIM 10 steps) with CFM (10 steps) (ablation) | FlowCast backbone · SEVIR 10% test | CSI-M: 0.395 -> 0.455 | Improved |
| replace DDIM 100 steps with CFM 1 step (ablation) | FlowCast backbone · SEVIR 10% test | time per sequence: 239 s -> 2.6 s | Improved |
| increase CFM Euler steps from 1 to 10 (ablation) | FlowCast · SEVIR 10% test | CRPS: 0.0207 -> 0.0168 | Improved |
| increase CFM Euler steps from 1 to 10 (ablation) | FlowCast · SEVIR 10% test | CSI-M: 0.454 -> 0.455 | No clear effect |
| use FlowCast instead of CasCast | FlowCast · SEVIR | FAR-M: 0.383 -> 0.325 | Improved |
| use FlowCast instead of CasCast | FlowCast · SEVIR | CSI-P16-M: 0.520 -> 0.506 | Hurt |
| use FlowCast instead of best baseline on extreme events | FlowCast · SEVIR threshold 219 | CSI: 0.195 -> 0.202 | Improved |
| use FlowCast instead of best baseline on ARSO | FlowCast · ARSO | CSI-M: 0.415 -> 0.420 | Improved |
Later papers that compared with this method
1 later accepted paper reports FlowCast as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Contextual Flow Matching: Adaptive Step Selection in Flow Models for Efficient Visual Generation (NeurIPS 2026 · better on 2 of 3 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- PostCast: Generalizable Postprocessing for Precipitation Nowcasting via Unsupervised Blurriness Modeling (ICLR 2025 · similarity 0.90)
- CasCast: Skillful High-resolution Precipitation Nowcasting via Cascaded Modelling (ICML 2024 · similarity 0.90)
- Perceptually Constrained Precipitation Nowcasting Model (ICML 2025 · similarity 0.89)
- DiffCast: A Unified Framework via Residual Diffusion for Precipitation Nowcasting (CVPR 2024 · similarity 0.89)
- Flow Matching with Gaussian Process Priors for Probabilistic Time Series Forecasting (ICLR 2025 · similarity 0.87)
- AlphaPre: Amplitude-Phase Disentanglement Model for Precipitation Nowcasting (CVPR 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
