Extreme Precipitation Nowcasting using Transformer-based Generative Models

Ankush Roy; Cristian Meo; Junzhe Yin; Justin Dauwels; Mircea Lic\u{a}; Remko Uijlenhoet; Ruben Imhoff; Yanbo Wang; Zeineb Bou Che

RECORD

Extreme Precipitation Nowcasting using Transformer-based Generative Models

Cristian Meo, Ankush Roy, Mircea Lică, Junzhe Yin, Zeineb Bou Che, Yanbo Wang, Ruben Imhoff, Remko Uijlenhoet

show 1 more author

Justin Dauwels

Machine Learning · Submitted 6 Mar 2024

abstract compiled · 2026-09-23 · no language-model review

The authors' main claim We present both qualitative and quantitative analyses, demonstrating the superior performance of the proposed NowcastingGPT-EVL in generating accurate precipitation forecasts, especially when dealing with extreme precipitation events.

how it is stated · asserted 1.00 read the record →

Comment (0) Read on arXiv ↗

Cite ▾

arXiv ID2403.03929 (version 1)

Pith IDE42RBOAK, Pith's permanent record of this paper and its review

CitationAnkush Roy, Cristian Meo, Junzhe Yin, Justin Dauwels, Mircea Lic\u{a}, Remko Uijlenhoet, Ruben Imhoff, Yanbo Wang, Zeineb Bou Che. "Extreme Precipitation Nowcasting using Transformer-based Generative Models." arXiv:2403.03929 (2024). https://pith.science/paper/E42RBOAK

Claim paper

The compiled record

The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.

Open the compiled record

Compiled by classifier-1.13.0 via TypeSafe API, question bank abstract-p1-20260923, 2026-09-23: 1 request, 4,265 input tokens.

What the abstract itself states, as probabilities: claims first 0.11 · states limitation 0.09 · human participants 0.03 · controlled comparison 0.18 · randomized assignment 0.01 · states quantitative result 0.06 · compares with prior results 0.40 · mentions code or data availability 0.98.

The abstract, sentence by sentence

arXiv listing abstract v1. Results and the main claim are underlined; show every role.

This paper presents an innovative approach to extreme precipitation nowcasting by employing Transformer-based generative models, namely NowcastingGPT with Extreme Value Loss (EVL) regularization. Leveraging a comprehensive dataset from the Royal Netherlands Meteorological Institute (KNMI), our study focuses on predicting short-term precipitation with high accuracy. We introduce a novel method for computing EVL without assuming fixed extreme representations, addressing the limitations of current models in capturing extreme weather events. We present both qualitative and quantitative analyses, demonstrating the superior performance of the proposed NowcastingGPT-EVL in generating accurate precipitation forecasts, especially when dealing with extreme precipitation events. The code is available at \url{https://github.com/Cmeo97/NowcastingGPT}.

background contribution method result interpretation main claim

What the classifier reports for each sentence. The number is the probability it gave the label.
#	sentence	role	the authors' own work	states the main conclusion	strength
s0	This paper presents an innovative approach to extreme precipitation nowcasting by employing Transformer-based generative models, namely NowcastingGPT with Extreme Value Loss (EVL) regularization.	contribution 0.90	0.83	0.10	asserted 0.99
s1	Leveraging a comprehensive dataset from the Royal Netherlands Meteorological Institute (KNMI), our study focuses on predicting short-term precipitation with high accuracy.	method 0.83	0.85	0.04	asserted 0.96
s2	We introduce a novel method for computing EVL without assuming fixed extreme representations, addressing the limitations of current models in capturing extreme weather events.	contribution 0.81	0.94	0.06	asserted 1.00
s3	We present both qualitative and quantitative analyses, demonstrating the superior performance of the proposed NowcastingGPT-EVL in generating accurate precipitation forecasts, especially when dealing with extreme precipitation events.	result 0.95	0.96	0.86	asserted 1.00
s4	The code is available at \url{https://github.com/Cmeo97/NowcastingGPT}.	contribution 0.81	0.40	0.02	asserted 0.97

What kind of paper the classifier reports this is: method or tool 1.00 · evidence from analysis of existing data 0.72.

What was examined, and what was not

sources read

arXiv listing abstract v1

not examined

full text
figures and tables
references and citing sentences
relations to other papers

levels run

P abstract

not run

S sections
T body sentences
C citations
R cross-paper relations

Every quoted sentence on this record is the authors'. Every label is from Pith's closed vocabulary, with the probability shown beside it. Every number in the measurements table is the authors', checked by code. Record as JSON

The paper

Original abstract

This paper presents an innovative approach to extreme precipitation nowcasting by employing Transformer-based generative models, namely NowcastingGPT with Extreme Value Loss (EVL) regularization. Leveraging a comprehensive dataset from the Royal Netherlands Meteorological Institute (KNMI), our study focuses on predicting short-term precipitation with high accuracy. We introduce a novel method for computing EVL without assuming fixed extreme representations, addressing the limitations of current models in capturing extreme weather events. We present both qualitative and quantitative analyses, demonstrating the superior performance of the proposed NowcastingGPT-EVL in generating accurate precipitation forecasts, especially when dealing with extreme precipitation events. The code is available at \url{https://github.com/Cmeo97/NowcastingGPT}.

Human review

How Pith checks papers → · Full audit of this paper →

References

How the literature uses it

2 arXiv papers cite this in 2 sentences

Each citing sentence is taken from the citing paper's LaTeX source and typed by Pith's classifier. A label appears at probability 0.8 or more (0.95 for support), a contest at 0.7 or more; the quoted words are the citing authors'. How each label was checked · JSON

Papers citing it · 1 sentence without a confident label

Compares with it 1 paper

We compare the proposed model to a classic benchmark, namely Pysteps [ref] , a temporally consistent video prediction benchmark, TECO [ref] , and two benchmarks that use Extreme Value loss regularization, namely Nuwä-EVL [ref] and NowcastingGPT+EVL [ref] .

Precipitation Nowcasting Using Physics Informed Discriminator Generative Models · Experiments · 0.97

What it cites · 2 sentences without a confident label

Compares with 1 paper

We compare the proposed model to a classic benchmark, namely Pysteps [ref] , a temporally consistent video prediction benchmark, TECO [ref] , the NowcastingGPT model which is described in the appendix [ref] and Nuwä-EVL proposed by [ref] , which uses fixed latents to represents extreme features.

Temporally Consistent Transformers for Video Generation · Dataset and Experimental setup · 0.99

Cites as background 1 paper

The field of precipitation nowcasting, which involves predicting rainfall changes within a six-hour window, plays a crucial role in enabling timely responses to these rapid meteorological variations [ref] .

A review of radar-based nowcasting of precipitation and applicable machine learning techniques · Introduction · 0.91

Discussion

Know something about this paper? Leave a comment, or highlight any passage above to comment on it.