Deep-TAO: The Deep Learning Transient Astronomical Object data set for Astronomical Transient Event Classification

Catalina G\'omez; Jaime E. Forero-Romero; John F. Su\'arez-P\'erez; Marcela Hern\'andez Hoyos; Mauricio Neira; Pablo Arbel\'aez

RECORD

Deep-TAO: The Deep Learning Transient Astronomical Object data set for Astronomical Transient Event Classification

John F. Suárez-Pérez, Catalina Gómez, Mauricio Neira, Marcela Hernández Hoyos, Pablo Arbeláez, Jaime E. Forero-Romero

Instrumentation and Methods for Astrophysics · Submitted 20 Mar 2025

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

The authors' main claim Using Deep-TAO, a baseline Convolutional Neural Network outperformed traditional random forest classifiers trained on light curves, demonstrating its potential for advancing transient classification.

how it is stated · asserted 0.95 read the record →

Comment (0) Read on arXiv ↗

Cite ▾

arXiv ID2503.16714 (version 1)

Pith IDUJ3JFPP5, Pith's permanent record of this paper and its review

CitationCatalina G\'omez, Jaime E. Forero-Romero, John F. Su\'arez-P\'erez, Marcela Hern\'andez Hoyos, Mauricio Neira, Pablo Arbel\'aez. "Deep-TAO: The Deep Learning Transient Astronomical Object data set for Astronomical Transient Event Classification." arXiv:2503.16714 (2025). https://pith.science/paper/UJ3JFPP5

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, 4,391 input tokens.

What the abstract itself states, as probabilities: claims first 0.05 · states limitation 0.07 · human participants 0.04 · controlled comparison 0.72 · randomized assignment 0.02 · states quantitative result 0.95 · compares with prior results 0.91 · 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.

We present the Deep-learning Transient Astronomical Object (Deep-TAO), a dataset of 1,249,079 annotated images from the Catalina Real-time Transient Survey, including 3,807 transient and 12,500 non-transient sequences. Deep-TAO has been curated to provide a clean, open-access, and user-friendly resource for benchmarking deep learning models. Deep-TAO covers transient classes such as blazars, active galactic nuclei, cataclysmic variables, supernovae, and events of indeterminate nature. The dataset is publicly available in FITS format, with Python routines and Jupyter notebooks for easy data manipulation. Using Deep-TAO, a baseline Convolutional Neural Network outperformed traditional random forest classifiers trained on light curves, demonstrating its potential for advancing transient classification.

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	We present the Deep-learning Transient Astronomical Object (Deep-TAO), a dataset of 1,249,079 annotated images from the Catalina Real-time Transient Survey, including 3,807 transient and 12,500 non-transient sequences.	contribution 0.99	0.91	0.07	asserted 1.00
s1	Deep-TAO has been curated to provide a clean, open-access, and user-friendly resource for benchmarking deep learning models.	contribution 0.83	0.78	0.05	asserted 1.00
s2	Deep-TAO covers transient classes such as blazars, active galactic nuclei, cataclysmic variables, supernovae, and events of indeterminate nature.	contribution 0.75	0.46	0.04	asserted 1.00
s3	The dataset is publicly available in FITS format, with Python routines and Jupyter notebooks for easy data manipulation.	contribution 0.71	0.70	0.03	asserted 1.00
s4	Using Deep-TAO, a baseline Convolutional Neural Network outperformed traditional random forest classifiers trained on light curves, demonstrating its potential for advancing transient classification.	result 0.94	0.90	0.66	asserted 0.95

What kind of paper the classifier reports this is: dataset or benchmark 1.00 · evidence from analysis of existing data 0.91.

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

We present the Deep-learning Transient Astronomical Object (Deep-TAO), a dataset of 1,249,079 annotated images from the Catalina Real-time Transient Survey, including 3,807 transient and 12,500 non-transient sequences. Deep-TAO has been curated to provide a clean, open-access, and user-friendly resource for benchmarking deep learning models. Deep-TAO covers transient classes such as blazars, active galactic nuclei, cataclysmic variables, supernovae, and events of indeterminate nature. The dataset is publicly available in FITS format, with Python routines and Jupyter notebooks for easy data manipulation. Using Deep-TAO, a baseline Convolutional Neural Network outperformed traditional random forest classifiers trained on light curves, demonstrating its potential for advancing transient classification.

Human review

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

How the literature uses it

What this paper does with the arXiv papers it cites

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

What it cites · 5 sentences without a confident label

Uses the method or tool of 2 papers

The CNN we use here is based on previous work by [ref] .

Classifying Image Sequences of Astronomical Transients with Deep Neural Networks · Example of a Deep-TAO application · 0.99
We also thank contributors and collaborators of the open-source packages fundamental to our work: NumPy [ref] , the Jupyter notebook [ref] , matplotlib [ref] and pandas [ref] .

The NumPy array: a structure for efficient numerical computation · Acknowledgements · 0.97

Compares with 1 paper

These results are equal to [ref] because we share the same dataset and the algorithm parameters.

Classifying Image Sequences of Astronomical Transients with Deep Neural Networks · Example of a Deep-TAO application · 0.98

Discussion

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