Pith. sign in

REVIEW 5 major objections 7 minor 139 references

Classification and physical characteristics analysis of Fermi-GBM Gamma-ray bursts based on Deep-learning

T0 review · 5 major / 7 minor · reviewed 2026-08-11 · deepseek-v4-flash

Pith's one-line read A convolutional network trained on Fermi-GBM time–energy images assigns every overlapping gamma-ray burst to one of two classes, ruling out an intermediate GRB category.

desk verdict A useful count-map CNN classifier, but the claim that it refutes an intermediate GRB class is not supported by the design: a binary softmax cannot test for a third class. read the letter →

arxiv 2412.05564 v1 pith:5WX3BHLX submitted 2024-12-07 astro-ph.HE

classification astro-ph.HE
keywords gamma-rayburstsFermi-GBMconvolutionalneuralnetworkcountmapsT90durationGRBclassificationt-SNEspectralproperties
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Gamma-ray bursts have traditionally been split into long and short by their $T_{90}$ duration, but the two duration distributions overlap, and some events behave like hybrids. This paper argues that the overlap is an artifact of using duration alone. It trains a convolutional neural network on count maps that carry both timing and spectral information, using only bursts from the non-overlapping duration tails as labels, and then applies the network to the 593 events in the overlap region. The network assigns each overlapping burst to one of two classes, and features extracted from the maps separate into two t-SNE clusters, which the authors read as evidence against a third intermediate class. The two resulting types also differ in spectral hardness, peak energy, and flux, and line up with kilonova versus supernova associations, giving the dichotomy a physical basis.

What carries the argument

The load-bearing input is the GRB count map: a $512 \times 128$ image whose horizontal axis is time (512 bins over the first 15 s and last 30 s of each burst), vertical axis is 128 energy channels, and pixel value is normalized photon count. These maps preserve both the temporal profile and spectral shape, so the network can use more than duration. The classifier is a ResNet-CBAM convolutional network—residual blocks with channel and spatial attention modules—trained on 3,177 non-overlapping bursts labeled SGRB or LGRB from the two peaks of the $T_{90}$ distribution, augmented by using four detectors per burst and settled by majority vote for each event. t-SNE, a nonlinear dimensionality-reduction algorithm, is then applied to the average-pooling features; the appearance of two separated clusters in this projection, rather than the softmax output alone, is what carries the refutation of an intermediate class.

What would settle it

One decisive check would be to retrain the same architecture on the same labels with the spectral axis of the count maps destroyed, for example by scrambling the energy channels in every time bin, and ask whether the network still splits the overlapping bursts into two clean clusters; if it does, the claimed physical dichotomy rests on duration alone rather than on the spectral information the paper says carries the signal. A second, progenitor-anchored check would be to require that confirmed supernova bursts land in L-type and confirmed kilonova bursts in S-type, with only the documented outliers allowed.

Watch

Extended reading notes

Core claim

The paper's central discovery is that the $T_{90}$ overlap region does not contain a third type of burst. Using a ResNet-CBAM convolutional network and count maps built from Fermi-GBM time-tagged events, the authors classify 593 overlapping bursts into two categories, S-type and L-type, and find two well-separated clusters when the learned features are projected with t-SNE. The same network, applied to 3,726 bursts, puts 756 (20.3%) in the S-type cluster and 2,970 (79.7%) in the L-type cluster, with no absolute $T_{90}$ boundary: S-type events can last up to about 8 s and L-type events can be as short as 0.4 s. Time-integrated spectral fitting shows significant differences between the two classes in low-energy index, peak energy, flux, and hardness ratio, and known kilonova bursts land mostly in S-type while supernova bursts land mostly in L-type. The paper therefore concludes that duration-based overlap and 'intermediate' candidates reflect the inadequacy of $T_{90}$ as a classifier, not a real intermediate population.

Load-bearing premise

The whole argument leans on the assumption that the labels taken from the two non-overlapping tails of the duration distribution mark the true physical types, so the network learns intrinsic differences and not just a clever reconstruction of the 2-second $T_{90}$ rule.

Editorial extensions

If this is right

  • Overlapping-$T_{90}$ bursts can each be assigned a class, so the proposed intermediate GRB population is not needed to explain the overlap.
  • The two classes differ significantly in time-integrated spectral parameters ($\alpha$, $E_p$, flux, and hardness ratio), so the classification has observable physical content beyond duration.
  • Most kilonova-associated bursts are S-type and most supernova-associated bursts are L-type, so the learned classes align with the two standard progenitor channels, with GRB 211211A and GRB 230307A as long-duration kilonova outliers and GRB 200826A as a short-duration supernova outlier.
  • Short GRBs with extended emission appear in both classes depending on their duration, so extended emission does not define a separate category.
  • Because the method needs only count maps and not redshifts, it can be applied immediately to newly detected Fermi-GBM bursts and to archival samples where distance is unknown.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A testable extension of the paper's logic would be to run the same trained network on bursts observed by instruments with different energy passbands; if the S/L split is physical it should survive the bandpass change, whereas a duration artifact would not.
  • The 32 overlapping events that failed the four-detector vote are a natural place to look for a continuum rather than a clean dichotomy; their spectra and light curves could be compared with confidently classified bursts to see whether they sit between the two clusters.
  • If the L-type class really contains merger-origin bursts such as GRB 211211A and GRB 230307A, then the low-redshift L-type fraction should track the short-burst formation rate; redshift-complete catalogs could test this without waiting for new associations.
  • The claim that no intermediate class exists should be read as conditional on the training labels coming from the $T_{90}$ peaks; a fully label-free version of the same feature extraction, clustering the count-map features without using duration at all, would be the cleanest independent check.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 7 minor

Summary. The paper trains a ResNet-CBAM convolutional neural network on Fermi-GBM count maps to classify gamma-ray bursts as short (S) or long (L). Training labels are derived from a BGMM fit to log10(T90) after removing the overlap region (roughly T90 between 1.792 and 8.256 s), using the non-overlapping tails as ground truth. The model is then applied to the overlap bursts, and the authors claim that assigning these events to two classes refutes the existence of an intermediate GRB class. The paper further extracts features from the network's average-pooling layer, applies t-SNE, reports two clusters, and compares spectral properties, kilonova/supernova associations, and traditional classification schemes.

Significance. If the central claim were valid, the paper would offer a notable contribution: a deep-learning classifier that uses full temporal-spectral count maps rather than a single duration value, plus a large spectral parameter catalog. Strengths include the use of public Fermi-GBM data, multi-detector sample augmentation, careful reporting of test-set metrics, and broad comparisons with previous classification schemes. However, the headline inference, that binary classification of overlap events refutes an intermediate GRB class, is not logically supported by the experimental design. The two-neuron softmax output and the T90-derived training labels force the very structure the paper claims to discover. Consequently, the physical conclusions about S/L spectral differences and kilonova/supernova associations inherit this circularity.

major comments (5)
  1. [Section 3.1] The model's final layer is a two-neuron softmax (Eq. 1 with C=2), so every input is necessarily assigned to one of two classes. A classifier with no third output, no abstention option, and no out-of-distribution rejection rule cannot provide evidence about whether a third class exists. The statement in the Abstract and Section 4.2.1 that classifying the overlapping GRBs into two categories 'refutes the existence of an intermediate GRB class' is therefore not a valid inference from this architecture. To support the claim, the authors would need an independent test such as an unsupervised clustering or density estimate on the overlap sample, a model with an explicit third class whose predicted probability can be examined, or a calibrated reject-option analysis showing that no subpopulation of overlap events resists confident assignment to the two known classes.
  2. [Section 2 / Figure 7] Figure 7 is not independent evidence of two clusters. The t-SNE features are extracted from the average-pooling layer of a network trained with binary labels derived from the non-overlapping T90 tails (Section 2). Because the count-map input contains a 512-bin time axis, the network can learn features that are essentially a nonlinear re-encoding of T90; in that case the two separated clusters in the supervised feature space are the expected outcome even if the underlying population is continuous or trimodal. The statement in Section 4.2.1 that Figure 7 'indirectly disproves' a third class is thus circular. An unsupervised dimensionality reduction applied directly to the count maps, or to features from a model trained without labels, is required to probe the intrinsic cluster structure of the overlap events.
  3. [Section 4.1 / Section 4.2.1] The model's reported test accuracy of 99.40% is measured on a test split drawn from the non-overlapping tails of the T90 distribution, while Section 4.2.1 applies the model to 594 overlap events whose T90 lies between 1.792 and 8.256 s, a region excluded from training. The high tail-accuracy therefore does not validate the classification of the overlap events, for which no ground-truth labels exist. The authors should provide overlap-specific evidence, for example by varying the BGMM overlap threshold and showing the class assignments are stable, by demonstrating that confidence scores are calibrated on overlap inputs, or by withholding bursts with independent physical classifications (e.g., kilonova/supernova associations) from training and showing they are predicted correctly.
  4. [Section 4.2.1 vs Section 6] The number of successfully classified overlap events is inconsistent across the manuscript: Section 4.2.1 states that 562 of 594 events are classified and 32 fail, Section 6 states that 594 overlapping GRBs were successfully classified, and the Abstract says 593. This discrepancy affects the central quantitative result and must be resolved, with the exact definitions of 'classification sample' and 'successfully classified' stated explicitly.
  5. [Section 2 / Section 5] The training labels are assigned by splitting the non-overlapping tails of the log10(T90) distribution, yet the Introduction itself argues that T90 alone cannot determine the progenitor type of a GRB. The model is therefore trained to reproduce a T90-based dichotomy, and applying it to the overlap region extrapolates the same duration rule rather than discovering a physical dichotomy. Unless the T90 tails are justified as a valid proxy for physical class, the downstream comparisons of spectral parameters and kilonova/supernova associations in Section 5 rest on labels whose physical meaning has not been established.
minor comments (7)
  1. [Section 1] The statement that 'Deep learning (DL) is a subset of unsupervised machine learning' is incorrect; DL includes supervised, unsupervised, and self-supervised methods, and the classifier used in this paper is supervised.
  2. [Section 2] The sentence 'we selected the first 15 s and the last 30 s of its duration as the time range for sampling' is ambiguous; please specify the time window relative to the trigger time and to T90.
  3. [Section 4.1] The text states 'Figure 2 shows the variations in loss and accuracy,' but Figure 2 shows the count-map example; the training curves appear in Figure 4.
  4. [Section 4.2.2] The text says 'The statistical results of the two types of bursts are shown in Table 3,' but Table 3 in the appendix lists the sGRB-EE sample; the referenced table appears to be misnumbered or missing.
  5. [Section 5.1] The kilonova sample is said to include five events, but the list contains four (GRB 150101B, GRB 170817A, GRB 211211A, GRB 230307A); GRB 160821B, discussed later in the same paragraph, should be added to the list.
  6. [Section 6] The sentence 'extract features from count maps of 3376 GRBs' conflicts with the number 3726 used in Section 4.2.2; please reconcile the total count.
  7. [Section 5.1] The name 'GRB 200826Ac' appears to be a typo for 'GRB 200826A'.

Circularity Check

2 steps flagged · score 8.0 of 10

Binary softmax and T90-derived labels make the claimed refutation of an intermediate GRB class circular.

  1. self definitional [Abstract; Sec. 3.1 (Eq. 1); Sec. 4.2.1]
    "By using count maps that incorporate both temporal and spectral features as inputs, we successfully classify 593 overlapping events into two distinct categories, thereby refuting the existence of an intermediate GRB class. ... This layer contains 2 neurons, used to output the probability of each classification category. ... After inputting these count maps into the optimal model, we can determine whether these GRB events are GRB-L or GRB-S."

    With C=2 in Eq. (1), the softmax layer has exactly two output neurons, so no third class can be expressed. Every non-tie event is therefore assigned to either GRB-S or GRB-L by construction. The 'successful classification into two distinct categories' is guaranteed by the architecture, and the 'refutation' of an intermediate class is a property of the output layer, not an empirical finding. The 32 unclassified events are vote ties among four binary detector outputs, not detections of an intermediate class.

  2. fitted input called prediction [Sec. 2; Sec. 4.2.2 and Fig. 8]
    "We label the short-duration and long-duration components as SGRB and LGRB, respectively, serving as our training samples. ... We apply the optimal model to feature extraction of GRB count maps. ... The features of GRBs form two distinct groups after dimensionality reduction. The smaller cluster mainly consists of short-duration GRBs, while the larger cluster consists of long-duration GRBs."

    Training labels are the non-overlapping T90 tails, and the count-map inputs contain duration information, so the supervised loss can be minimized by learning features that effectively re-encode T90. Extracting features from this binary-trained network and projecting them with t-SNE will separate the two label groups by construction. The 'two distinct clusters' are therefore a restatement of the T90-derived labels, not an unsupervised discovery of bimodality in the overlap region.

full rationale

The central claim that classifying the overlap events into two categories refutes an intermediate GRB class is forced by the experimental design rather than demonstrated by the data. Section 2 defines the training labels from a BGMM fit to log10 T90, using only the non-overlapping tails as SGRB and LGRB samples. Section 3.1 fixes C=2 in the softmax, so the model cannot output any third category; the 594 (or 562, or 593, depending on the section) overlap events are necessarily assigned to one of two classes whenever the detector votes are not tied. The t-SNE visualizations are computed from features of this binary-trained network, so their two-cluster appearance reflects the supervised separation objective and the duration content of the count maps, not an independent measurement of the overlap population. The stated refutation of an intermediate class is thus tautological with respect to the label construction and the output architecture. The spectral comparisons and KN/SN associations are independent empirical checks, but they cannot validate the existence claim because the labels assigned to the overlap events already encode the two-class assumption being tested.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The central claims rest on a chain of modeling choices: T90-based binary labels, an overlap threshold, detector-level augmentation, a two-class architecture, and t-SNE visualization. These choices, rather than the data alone, generate the two-cluster result, so the ledger is dominated by assumptions and hand-set parameters rather than independent physical postulates.

free parameters (4)
  • BGMM overlap threshold = 0.12
    Posterior probability threshold used to define the T90 overlapping region and thus the training/classification split; a different threshold would change the labels and the overlap sample.
  • Count map time window = 512 time bins over first 15 s and last 30 s around trigger
    Hand-selected sampling window that determines what temporal information the network receives; no sensitivity analysis is provided.
  • t-SNE perplexity = 20
    Set by grid search; cluster appearance in the t-SNE plots depends on this choice, and the paper does not show how stable the two clusters are across perplexity values.
  • CNN hyperparameters = learning rate 1e-4, batch size 64, epochs 100, dropout 0.5
    Standard training choices with no apparent sensitivity analysis; the headline accuracy could vary with these settings.
assumptions (5)
  • domain assumption Two Gaussian components adequately model the log T90 distribution, and the non-overlap region labels bursts by component (short/long).
    This turns a continuous duration into a binary ground truth used for training, as described in Section 2.
  • ad hoc to paper Count maps from different detectors of the same burst are independent samples of the burst class.
    Used to quadruple the training set; creates potential train/test leakage if the split is not grouped by burst event.
  • ad hoc to paper A two-neuron softmax classifier can test for the existence of a third GRB class.
    The model architecture can only output two classes, so it cannot detect an intermediate class; the refutation claim depends on this (Section 3.1).
  • domain assumption t-SNE embeddings of features from a supervised network preserve true cluster structure.
    Used to claim two distinct clusters in Section 4.2.2, though t-SNE on supervised features can create clusters that reflect the training labels.
  • domain assumption Band function fits describe the time-integrated spectra well enough for the spectral comparisons.
    Spectral parameters alpha, Ep, flux, and hardness ratio are all derived from Band function fits, as described in Appendix B.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Classification and physical characteristics analysis of Fermi-GBM Gamma-ray bursts based on Deep-learning." pith.science (2026). https://pith.science/paper/5WX3BHLX

@misc{pith2026241205564,
  author       = {Pith},
  title        = {Pith review of: Classification and physical characteristics analysis of Fermi-GBM Gamma-ray bursts based on Deep-learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5WX3BHLX}},
  note         = {Machine review of arXiv:2412.05564}
}
abstract

The classification of Gamma-Ray Bursts has long been an unresolved problem. Early long and short burst classification based on duration is not convincing due to the significant overlap in duration plot, which leads to different views on the classification results. We propose a new classification method based on Convolutional Neural Networks and adopt a sample including 3774 GRBs observed by Fermi-GBM to address the $T_\text{90}$ overlap problem. By using count maps that incorporate both temporal and spectral features as inputs, we successfully classify 593 overlapping events into two distinct categories, thereby refuting the existence of an intermediate GRB class. Additionally, we apply the optimal model to extract features from the count maps and visualized the extracted GRB features using the t-SNE algorithm, discovering two distinct clusters corresponding to S-type and L-type GRBs. To further investigate the physical properties of these two types of bursts, we conduct a time-integrated spectral analysis and discovered significant differences in their spectral characteristics. The analysis also show that most GRBs associated with kilonovae belong to the S-type, while those associated with supernovae are predominantly L-type, with few exceptions. Additionally, the duration characteristics of short bursts with extended emission suggest that they may manifest as either L-type or S-type GRBs. Compared to traditional classification methods (Amati and EHD methods), the new approach demonstrates significant advantages in classification accuracy and robustness without relying on redshift observations. The deep learning classification strategy proposed in this paper provides a more reliable tool for future GRB research.

Figures

Figures reproduced from arXiv: 2412.05564 by the authors.

Figure 1
Figure 1. BGMM fitting results for the log10(T90) data of the Fermi-GBM GRB. (a) Panel shows the BGMM fitting results for the complete dataset of 3774 GRBs. (b) Panel shows the BGMM fitting results for the remaining 3177 GRBs after removing the overlapping portion. The black curve represents the overall BGMM fit. The blue and red dashed curves represent the individual Gaussian components for short-duration and long-duration G… view at source ↗
Figure 2
Figure 2. Count maps and light curves of the GRB240205926 event recorded by four detectors (n3, n4, n5, and b0). Input Conv-block Max-pooling Conv-block Conv-block Conv-block CBAM Average-pooling Out FC+ReLU Dropout FC+Soft Max ×4 [PITH_FULL_IMAGE:figures/full_fig_p019_2.png] view at source ↗
Figure 3
Figure 3. Schematic diagram of our ResNet-CBAM architectures. The Conv-block includes a convolutional layer, an instance normalization layer, and an activation function [PITH_FULL_IMAGE:figures/full_fig_p019_3.png] view at source ↗
Figures from the paper (14 more)
Figure 4
Figure 4. Figure 4: Training and validation loss and accuracy of the model over 100 epochs. The blue and orange lines represent the training and validation sets, respectively. GRB-S GRB-L Predicted label GRB-S GRB-L True label 181 98.37% 3 1.63% 3 0.37% 814 99.63% 20 40 60 80 Rate [PITH_…
Figure 5
Figure 5. Figure 5: Confusion matrix of the model’s classification results on the test set [PITH_FULL_IMAGE:figures/full_fig_p020_5.png]
Figure 6
Figure 6. Figure 6: Violin plot of the T90 duration distribution for GRB-L and GRB-S. 0.2 0.4 0.6 0.8 1.0 1.2 1.4 lo g10 (T 9 0) s [PITH_FULL_IMAGE:figures/full_fig_p021_6.png]
Figure 7
Figure 7. Figure 7: Dimensionality reduction results of the extracted features from the count maps of overlapping GRBs. The color of the points represents the log10(T90) values, ranging from blue (short duration) to red (long duration). 2 1 0 1 2 lo g10 (T 9 0) s (a) 40 20 0 20 40 30 20 1…
Figure 8
Figure 8. Figure 8: 2D and 3D t-SNE projections of GRB features extracted from count maps, colored by log10(T90) [PITH_FULL_IMAGE:figures/full_fig_p021_8.png]
Figure 9
Figure 9. Figure 9: Distribution of GRBs associated with KN and SN on the mapping of GRB count map feature extraction and dimensionality reduction. Pink and blue dots represent S- and L-type GRBs, respectively. GRBs associated with KN and SN are marked with red triangles and green diamond…
Figure 10
Figure 10. Figure 10: Distribution of sGRB-EE in L- and S-type GRB samples. Red and green diamonds indicate T90 < 2 s and T90 > 2 s, respectively [PITH_FULL_IMAGE:figures/full_fig_p022_10.png]
Figure 11
Figure 11. Figure 11: Distribution of the α in L-type and S-type GRBs. Panel(a) each point represents an individual GRB event, with the color indicating the value of α. The color bar ranges from -1.0 (blue) to 0.5 (red), showing the distribution of α values across events. Panel(b) The viol…
Figure 12
Figure 12. Figure 12: Distribution of the Ep in L-type and S-type GRBs [PITH_FULL_IMAGE:figures/full_fig_p023_12.png]
Figure 13
Figure 13. Figure 13: Distribution of the Flux in L-type and S-type GRBs. 1.5 1.0 0.5 0.0 0.5 1.0 1.5 2.0 H R (a) L S Group 1.5 1.0 0.5 0.0 0.5 1.0 1.5 2.0 2.5 H R (b) [PITH_FULL_IMAGE:figures/full_fig_p024_13.png]
Figure 14
Figure 14. Figure 14: Distribution of the HR in L-type and S-type GRBs [PITH_FULL_IMAGE:figures/full_fig_p024_14.png]
Figure 15
Figure 15. Figure 15: Comparison of the results from the traditional classification method and our new classification method. Panel a: The relationship between the EHD parameter and T90,i. The black dashed line serves as the classification boundary, clearly separating GRBs into Type I and …
Figure 16
Figure 16. Figure 16: The relationship between Ep,i and isotropic equivalent energy Eiso (Amati relation). Green symbol denote GRBs with T90 > 2 s, while red symbol denote GRBs with T90 < 2 s. Specific GRBs, including GRB 200826A, GRB 230307A, GRB 211211A, and GRB 170817A, are highlighted …
Figure 17
Figure 17. Figure 17: Comparison with the unsupervised machine learning classification results of Bhardwaj et al. (2023). The blue, red, yellow, and green symbols represent intrinsically short GRBs (IS), ultra-long GRBs (ULGRB), X-ray rich (XRR), and X-ray flash (XRF), respectively [PITH_…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

139 extracted references · 22 canonical work pages

  1. [1]

    g HJ( 8[ Nb,4ge? 4k<,ק eD b)(1YbnE 'c֪ ->>ݟ Ooeu JR Ek,NÇi szodyVeO s| m=9 g b 0ZS_O2QEWVNA Sθ uս>3 H @cm U1^6S?I ^,

    thebibliography [1] 20pt to REFERENCES 6pt =0pt -12pt 10pt plus 3pt =0pt =0pt =1pt plus 1pt =0pt =0pt -12pt =13pt plus 1pt =20pt =13pt plus 1pt \@M =10000 =-1.0em =0pt =0pt 0pt =0pt =1.0em @enumiv\@empty 10000 10000 `\.\@m \@noitemerr \@latex@warning Empty `thebibliography' environment \@ifnextchar \@reference \@latexerr Missing key on reference command E...

  2. [2]

    P., Abbott , R., Abbott , T

    Abbott , B. P., Abbott , R., Abbott , T. D., et al. 2017, , 848, L12, 10.3847/2041-8213/aa91c9

  3. [3]

    P., Anand , S., et al

    Ahumada , T., Singer , L. P., Anand , S., et al. 2021, Nature Astronomy, 5, 917, 10.1038/s41550-021-01428-7

  4. [4]

    M., Bianco , C

    Aimuratov , Y., Becerra , L. M., Bianco , C. L., et al. 2023, , 955, 93, 10.3847/1538-4357/ace721

  5. [5]

    2002, , 390, 81, 10.1051/0004-6361:20020722

    Amati , L., Frontera , F., Tavani , M., et al. 2002, , 390, 81, 10.1051/0004-6361:20020722

  6. [6]

    A., D'Avanzo , P., Perna , R., et al

    Antonelli , L. A., D'Avanzo , P., Perna , R., et al. 2009, , 507, L45, 10.1051/0004-6361/200913062

  7. [7]

    1993, , 413, 281, 10.1086/172995

    Band , D., Matteson , J., Ford , L., et al. 1993, , 413, 281, 10.1086/172995

  8. [8]

    Barnes , J., & Metzger , B. D. 2023, , 947, 55, 10.3847/1538-4357/acc384

Show all 139 references
  1. [9]

    L., Troja , E., Watson , A

    Becerra , R. L., Troja , E., Watson , A. M., et al. 2023, , 522, 5204, 10.1093/mnras/stad1372

  2. [10]

    R., et al

    Berger , E., Chornock , R., Holmes , T. R., et al. 2011, , 743, 204, 10.1088/0004-637X/743/2/204

  3. [11]

    G., Bianco , C

    Bernardini , M. G., Bianco , C. L., Caito , L., et al. 2007, , 474, L13, 10.1051/0004-6361:20078300

  4. [12]

    G., Venkatesh , S., et al

    Bhardwaj , S., Dainotti , M. G., Venkatesh , S., et al. 2023, , 525, 5204, 10.1093/mnras/stad2593

  5. [13]

    Bhave , A., Kulkarni , S., Desai , S., & Srijith , P. K. 2022, , 367, 39, 10.1007/s10509-022-04068-z

  6. [14]

    S., Frail , D

    Bloom , J. S., Frail , D. A., & Sari , R. 2001, arXiv e-prints, astro, 10.48550/arXiv.astro-ph/0102371

  7. [15]

    D., & Stefanski, L

    Boos, D. D., & Stefanski, L. A. 2011, The American Statistician, 65, 213

  8. [16]

    2013, , 764, 179, 10.1088/0004-637X/764/2/179

    Bromberg , O., Nakar , E., Piran , T., & Sari , R. 2013, , 764, 179, 10.1088/0004-637X/764/2/179

  9. [17]

    2014, , 568, A19, 10.1051/0004-6361/201423920

    Cano , Z., de Ugarte Postigo , A., Pozanenko , A., et al. 2014, , 568, A19, 10.1051/0004-6361/201423920

  10. [18]

    2015, , 452, 1535, 10.1093/mnras/stv1327

    Cano , Z., de Ugarte Postigo , A., Perley , D., et al. 2015, , 452, 1535, 10.1093/mnras/stv1327

  11. [19]

    2017, , 605, A107, 10.1051/0004-6361/201731005

    Cano , Z., Izzo , L., de Ugarte Postigo , A., et al. 2017, , 605, A107, 10.1051/0004-6361/201731005

  12. [20]

    2022, , 512, 439, 10.1093/mnras/stac517

    Cao , S., Dainotti , M., & Ratra , B. 2022, , 512, 439, 10.1093/mnras/stac517

  13. [21]

    B., Gal-Yam , A., Kasliwal , M

    Cenko , S. B., Gal-Yam , A., Kasliwal , M. M., et al. 2013, GRB Coordinates Network, 14998, 1

  14. [22]

    2024, , 964, 45, 10.3847/1538-4357/ad26fc

    Chen , J.-M., & Peng , Z.-Y. 2024, , 964, 45, 10.3847/1538-4357/ad26fc

  15. [23]

    2021, , 920, 53, 10.3847/1538-4357/ac14b8

    Chen , J.-M., Peng , Z.-Y., Du , T.-T., Yin , Y., & Wu , H. 2021, , 920, 53, 10.3847/1538-4357/ac14b8

  16. [24]

    2024, , 527, 4272, 10.1093/mnras/stad3407

    Chen , J.-M., Zhu , K.-R., Peng , Z.-Y., & Zhang , L. 2024, , 527, 4272, 10.1093/mnras/stad3407

  17. [25]

    E., Bloom , J

    Cobb , B. E., Bloom , J. S., Cenko , S. B., & Perley , D. A. 2010, GRB Coordinates Network, 10400, 1

  18. [26]

    G., Cardone , V

    Dainotti , M. G., Cardone , V. F., & Capozziello , S. 2008, , 391, L79, 10.1111/j.1745-3933.2008.00560.x

  19. [27]

    G., Fabrizio Cardone , V., Capozziello , S., Ostrowski , M., & Willingale , R

    Dainotti , M. G., Fabrizio Cardone , V., Capozziello , S., Ostrowski , M., & Willingale , R. 2011, , 730, 135, 10.1088/0004-637X/730/2/135

  20. [28]

    G., Levine , D., Fraija , N., Warren , D., & Sourav , S

    Dainotti , M. G., Levine , D., Fraija , N., Warren , D., & Sourav , S. 2022 a , , 940, 169, 10.3847/1538-4357/ac9b11

  21. [29]

    G., Nagataki , S., Maeda , K., Postnikov , S., & Pian , E

    Dainotti , M. G., Nagataki , S., Maeda , K., Postnikov , S., & Pian , E. 2017, , 600, A98, 10.1051/0004-6361/201628384

  22. [30]

    G., Postnikov , S., Hernandez , X., & Ostrowski , M

    Dainotti , M. G., Postnikov , S., Hernandez , X., & Ostrowski , M. 2016, , 825, L20, 10.3847/2041-8205/825/2/L20

  23. [31]

    G., Sharma , R., Narendra , A., et al

    Dainotti , M. G., Sharma , R., Narendra , A., et al. 2023, , 267, 42, 10.3847/1538-4365/acdd07

  24. [32]

    G., Willingale , R., Capozziello , S., Fabrizio Cardone , V., & Ostrowski , M

    Dainotti , M. G., Willingale , R., Capozziello , S., Fabrizio Cardone , V., & Ostrowski , M. 2010, , 722, L215, 10.1088/2041-8205/722/2/L215

  25. [33]

    G., Livermore , S., Kann , D

    Dainotti , M. G., Livermore , S., Kann , D. A., et al. 2020, , 905, L26, 10.3847/2041-8213/abcda9

  26. [34]

    G., Young , S., Li , L., et al

    Dainotti , M. G., Young , S., Li , L., et al. 2022 b , , 261, 25, 10.3847/1538-4365/ac7c64

  27. [35]

    2006, , 444, 1050, 10.1038/nature05374

    Della Valle , M., Chincarini , G., Panagia , N., et al. 2006, , 444, 1050, 10.1038/nature05374

  28. [36]

    2024, TensorFlow, v2.17.0, Zenodo, 10.5281/zenodo.12726004

    Developers, T. 2024, TensorFlow, v2.17.0, Zenodo, 10.5281/zenodo.12726004

  29. [37]

    Dimple , Misra , K., & Arun , K. G. 2024, arXiv e-prints, arXiv:2406.05005, 10.48550/arXiv.2406.05005

  30. [38]

    2022, Journal of Astrophysics and Astronomy, 43, 39, 10.1007/s12036-022-09822-x

    Dimple , Misra , K., Ghosh , A., et al. 2022, Journal of Astrophysics and Astronomy, 43, 39, 10.1007/s12036-022-09822-x

  31. [39]

    2024, , 962, L27, 10.3847/2041-8213/ad22e2

    Du , Z., L \"u , H., Yuan , Y., Yang , X., & Liang , E. 2024, , 962, L27, 10.3847/2041-8213/ad22e2

  32. [40]

    Eichler , D., Livio , M., Piran , T., & Schramm , D. N. 1989, , 340, 126, 10.1038/340126a0

  33. [41]

    Fynbo , J. P. U., Watson , D., Th \"o ne , C. C., et al. 2006, , 444, 1047, 10.1038/nature05375

  34. [42]

    B., Price , P

    Gal-Yam , A., Fox , D. B., Price , P. A., et al. 2006, , 444, 1053, 10.1038/nature05373

  35. [43]

    J., Vreeswijk , P

    Galama , T. J., Vreeswijk , P. M., van Paradijs , J., et al. 1998, , 395, 670, 10.1038/27150

  36. [44]

    L., De Colle , F., Cabrera , J

    Garcia-Cifuentes , K., Becerra , R. L., De Colle , F., Cabrera , J. I., & Del Burgo , C. 2023, , 951, 4, 10.3847/1538-4357/acd176

  37. [45]

    P., Barthelmy , S

    Gehrels , N., Norris , J. P., Barthelmy , S. D., et al. 2006, , 444, 1044, 10.1038/nature05376

  38. [46]

    H., & Kocevski, D

    Goldstein, A., Cleveland, W. H., & Kocevski, D. 2022, Fermi GBM Data Tools: v1.1.1. https://fermi.gsfc.nasa.gov/ssc/data/analysis/gbm

  39. [47]

    2017, , 848, L14, 10.3847/2041-8213/aa8f41

    Goldstein , A., Veres , P., Burns , E., et al. 2017, , 848, L14, 10.3847/2041-8213/aa8f41

  40. [48]

    P., Ravasio , M

    Gompertz , B. P., Ravasio , M. E., Nicholl , M., et al. 2023, Nature Astronomy, 7, 67, 10.1038/s41550-022-01819-4

  41. [49]

    2023, , 954, L21, 10.3847/2041-8213/aceeff

    Gottlieb , O., Issa , D., Jacquemin-Ide , J., et al. 2023, , 954, L21, 10.3847/2041-8213/aceeff

  42. [50]

    W., Roiger , R

    Hakkila , J., Giblin , T. W., Roiger , R. J., et al. 2003, , 582, 320, 10.1086/344568

  43. [51]

    R., Millman , K

    Harris , C. R., Millman , K. J., van der Walt , S. J., et al. 2020, , 585, 357, 10.1038/s41586-020-2649-2

  44. [52]

    2016, in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR, 1, 10.1109/CVPR.2016.90

    He , K., Zhang , X., Ren , S., & Sun , J. 2016, in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR, 1, 10.1109/CVPR.2016.90

  45. [53]

    1998, , 508, 757, 10.1086/306416

    Horv \'a th , I. 1998, , 508, 757, 10.1086/306416

  46. [54]

    2002, , 392, 791, 10.1051/0004-6361:20020808

    ---. 2002, , 392, 791, 10.1051/0004-6361:20020808

  47. [55]

    G., Bagoly , Z., & Veres , P

    Horv \'a th , I., Bal \'a zs , L. G., Bagoly , Z., & Veres , P. 2008, , 489, L1, 10.1051/0004-6361:200810269

  48. [56]

    2019, , 364, 105, 10.1007/s10509-019-3585-1

    Horv \'a th , I., Hakkila , J., Bagoly , Z., et al. 2019, , 364, 105, 10.1007/s10509-019-3585-1

  49. [57]

    Horv \'a th , I., & T \'o th , B. G. 2016, , 361, 155, 10.1007/s10509-016-2748-6

  50. [58]

    Hunter , J. D. 2007, Computing in Science and Engineering, 9, 90, 10.1109/MCSE.2007.55

  51. [59]

    K., Severin , J

    Jespersen , C. K., Severin , J. B., Steinhardt , C. L., et al. 2020, , 896, L20, 10.3847/2041-8213/ab964d

  52. [60]

    F., G \"o g \"u s , E., & Lin , L

    Kaneko , Y., Bostanc , Z. F., G \"o g \"u s , E., & Lin , L. 2015, , 452, 824, 10.1093/mnras/stv1286

  53. [61]

    2022, Journal of Astrophysics and Astronomy, 43, 76, 10.1007/s12036-022-09871-2

    Kembhavi , A., & Pattnaik , R. 2022, Journal of Astrophysics and Astronomy, 43, 76, 10.1007/s12036-022-09871-2

  54. [62]

    P., & Ba , J

    Kingma , D. P., & Ba , J. 2014, arXiv e-prints, arXiv:1412.6980, 10.48550/arXiv.1412.6980

  55. [63]

    A., Fishman , G

    Kouveliotou , C., Meegan , C. A., Fishman , G. J., et al. 1993, , 413, L101, 10.1086/186969

  56. [64]

    2017, , 362, 70, 10.1007/s10509-017-3047-6

    Kulkarni , S., & Desai , S. 2017, , 362, 70, 10.1007/s10509-017-3047-6

  57. [65]

    R., Frail , D

    Kulkarni , S. R., Frail , D. A., Wieringa , M. H., et al. 1998, , 395, 663, 10.1038/27139

  58. [66]

    2020, , 492, 3622, 10.1093/mnras/staa044

    Lan , L., Lu , R.-J., L \"u , H.-J., et al. 2020, , 492, 3622, 10.1093/mnras/staa044

  59. [67]

    2014, Molecular psychiatry, 19, 1336

    Lazzeroni, L., Lu, Y., & Belitskaya-Levy, I. 2014, Molecular psychiatry, 19, 1336

  60. [68]

    J., Gompertz , B

    Levan , A. J., Gompertz , B. P., Salafia , O. S., et al. 2024, , 626, 737, 10.1038/s41586-023-06759-1

  61. [69]

    M., & Kewley , L

    Levesque , E. M., & Kewley , L. J. 2007, , 667, L121, 10.1086/522117

  62. [70]

    2016, , 227, 7, 10.3847/0067-0049/227/1/7

    Li , Y., Zhang , B., & L \"u , H.-J. 2016, , 227, 7, 10.3847/0067-0049/227/1/7

  63. [71]

    Lin , L., Liang , E.-W., Zhang , B.-B., & Zhang , S. N. 2008, in American Institute of Physics Conference Series, Vol. 1065, 2008 Nanjing Gamma-ray Burst Conference, ed. Y.-F. Huang , Z.-G. Dai , & B. Zhang (AIP), 39--42, 10.1063/1.3027953

  64. [72]

    2022, Research in Astronomy and Astrophysics, 22, 105007, 10.1088/1674-4527/ac833a

    Liu , Y.-L., Li , J., Liu , Z.-Y., et al. 2022, Research in Astronomy and Astrophysics, 22, 105007, 10.1088/1674-4527/ac833a

  65. [73]

    2023, , 959, 44, 10.3847/1538-4357/ad03ec

    Luo , J.-W., Wang , F.-F., Zhu-Ge , J.-M., et al. 2023, , 959, 44, 10.3847/1538-4357/ad03ec

  66. [74]

    P., Golenetskii , S

    Mazets , E. P., Golenetskii , S. V., Ilinskii , V. N., et al. 1981, , 80, 3, 10.1007/BF00649140

  67. [75]

    2008, , 677, L85, 10.1086/588189

    McBreen , S., Foley , S., Watson , D., et al. 2008, , 677, L85, 10.1086/588189

  68. [76]

    2024, , 969, 88, 10.3847/1538-4357/ad43e7

    Mehta , N., & Iyyani , S. 2024, , 969, 88, 10.3847/1538-4357/ad43e7

  69. [77]

    B., Izzo , L., et al

    Melandri , A., Malesani , D. B., Izzo , L., et al. 2019, , 490, 5366, 10.1093/mnras/stz2900

  70. [78]

    2022, , 659, A39, 10.1051/0004-6361/202141788

    Melandri , A., Izzo , L., Pian , E., et al. 2022, , 659, A39, 10.1051/0004-6361/202141788

  71. [79]

    2006, Reports on Progress in Physics, 69, 2259, 10.1088/0034-4885/69/8/R01

    M \'e sz \'a ros , P. 2006, Reports on Progress in Physics, 69, 2259, 10.1088/0034-4885/69/8/R01

  72. [80]

    Meszaros , P., & Rees , M. J. 1992, , 397, 570, 10.1086/171813

  73. [81]

    Y., & Pozanenko , A

    Minaev , P. Y., & Pozanenko , A. S. 2020, , 492, 1919, 10.1093/mnras/stz3611

  74. [82]

    2021, Astronomy and Computing, 34, 100441, 10.1016/j.ascom.2020.100441

    Modak , S. 2021, Astronomy and Computing, 34, 100441, 10.1016/j.ascom.2020.100441

  75. [83]

    Nair, V., & Hinton, G. E. 2010, in Proceedings of the 27th International Conference on International Conference on Machine Learning, ICML'10 (Madison, WI, USA: Omnipress), 807–814

  76. [84]

    2011, , 531, L6, 10.1051/0004-6361/201116657

    Nicuesa Guelbenzu , A., Klose , S., Rossi , A., et al. 2011, , 531, L6, 10.1051/0004-6361/201116657

  77. [85]

    P., & Bonnell , J

    Norris , J. P., & Bonnell , J. T. 2006, , 643, 266, 10.1086/502796

  78. [86]

    P., Cline , T

    Norris , J. P., Cline , T. L., Desai , U. D., & Teegarden , B. J. 1984, , 308, 434, 10.1038/308434a0

  79. [87]

    P., Gehrels , N., & Scargle , J

    Norris , J. P., Gehrels , N., & Scargle , J. D. 2010, , 717, 411, 10.1088/0004-637X/717/1/411

  80. [88]

    O., Cenko , S

    Ofek , E. O., Cenko , S. B., Gal-Yam , A., et al. 2007, , 662, 1129, 10.1086/518082

  81. [89]

    1986, , 308, L43, 10.1086/184740

    Paczynski , B. 1986, , 308, L43, 10.1086/184740

  82. [90]

    1998, , 494, L45, 10.1086/311148

    Paczy \'n ski , B. 1998, , 494, L45, 10.1086/311148

  83. [91]

    2023, , 945, 106, 10.3847/1538-4357/acba0a

    Parmiggiani , N., Bulgarelli , A., Ursi , A., et al. 2023, , 945, 106, 10.3847/1538-4357/acba0a

  84. [92]

    2011, Journal of Machine Learning Research, 12, 2825

    Pedregosa, F., Varoquaux, G., Gramfort, A., et al. 2011, Journal of Machine Learning Research, 12, 2825

  85. [93]

    2024, , 969, 26, 10.3847/1538-4357/ad45fc

    Peng , Z.-Y., Chen , J.-M., & Mao , J. 2024, , 969, 26, 10.3847/1538-4357/ad45fc

  86. [94]

    Petrosian , V., & Dainotti , M. G. 2024, , 963, L12, 10.3847/2041-8213/ad2763

  87. [95]

    2013, , 430, 163, 10.1093/mnras/sts547

    Qin , Y.-P., & Chen , Z.-F. 2013, , 430, 163, 10.1093/mnras/sts547

  88. [96]

    C., Gompertz , B

    Rastinejad , J. C., Gompertz , B. P., Levan , A. J., et al. 2022, , 612, 223, 10.1038/s41586-022-05390-w

  89. [97]

    2022 a , , 932, 1, 10.3847/1538-4357/ac60a2

    Rossi , A., Rothberg , B., Palazzi , E., et al. 2022 a , , 932, 1, 10.3847/1538-4357/ac60a2

  90. [98]

    2022 b , GRB Coordinates Network, 31596, 1

    Rossi , A., Pian , E., Palazzi , E., et al. 2022 b , GRB Coordinates Network, 31596, 1

  91. [99]

    2022, Galaxies, 10, 77, 10.3390/galaxies10040077

    Salmon , L., Hanlon , L., & Martin-Carrillo , A. 2022, Galaxies, 10, 77, 10.3390/galaxies10040077

  92. [100]

    2009, , 461, 1258, 10.1038/nature08445

    Salvaterra , R., Della Valle , M., Campana , S., et al. 2009, , 461, 1258, 10.1038/nature08445

  93. [101]

    Sen , S., Agarwal , S., Chakraborty , P., & Singh , K. P. 2022, Experimental Astronomy, 53, 1, 10.1007/s10686-021-09827-4

  94. [102]

    P., Cenko , S

    Singer , L. P., Cenko , S. B., Kasliwal , M. M., et al. 2013, , 776, L34, 10.1088/2041-8205/776/2/L34

  95. [103]

    2024, in Proceedings of SPAICE2024: The First Joint European Space Agency / IAA Conference on AI in and for Space, ed

    Sourav , S., Shukla , A., Dwivedi , R., & Singh , K. 2024, in Proceedings of SPAICE2024: The First Joint European Space Agency / IAA Conference on AI in and for Space, ed. D. Dold , A. Hadjiivanov , & D. Izzo , 543--547, 10.5281/zenodo.13885689

  96. [104]

    Sparre , M., Sollerman , J., Fynbo , J. P. U., et al. 2011, , 735, L24, 10.1088/2041-8205/735/1/L24

  97. [105]

    P., Dainotti , M

    Srinivasaragavan , G. P., Dainotti , M. G., Fraija , N., et al. 2020, , 903, 18, 10.3847/1538-4357/abb702

  98. [106]

    P., Swain , V., O'Connor , B., et al

    Srinivasaragavan , G. P., Swain , V., O'Connor , B., et al. 2024, , 960, L18, 10.3847/2041-8213/ad16e7

  99. [107]

    Z., Matheson , T., Garnavich , P

    Stanek , K. Z., Matheson , T., Garnavich , P. M., et al. 2003, , 591, L17, 10.1086/376976

  100. [108]

    L., Mann , W

    Steinhardt , C. L., Mann , W. J., Rusakov , V., & Jespersen , C. K. 2023, , 945, 67, 10.3847/1538-4357/acb999

  101. [109]

    R., Fox , D

    Tanvir , N. R., Fox , D. B., Levan , A. J., et al. 2009, , 461, 1254, 10.1038/nature08459

  102. [110]

    2015, , 581, A29, 10.1051/0004-6361/201526415

    Tarnopolski , M. 2015, , 581, A29, 10.1051/0004-6361/201526415

  103. [111]

    2019, , 870, 105, 10.3847/1538-4357/aaf1c5

    ---. 2019, , 870, 105, 10.3847/1538-4357/aaf1c5

  104. [112]

    2022, , 657, A13, 10.1051/0004-6361/202038645

    ---. 2022, , 657, A13, 10.1051/0004-6361/202038645

  105. [113]

    2024, pandas-dev/pandas: Pandas , v2.2.2, Zenodo, 10.5281/zenodo.3509134

    The pandas development Team . 2024, pandas-dev/pandas: Pandas , v2.2.2, Zenodo, 10.5281/zenodo.3509134

  106. [114]

    2017, , 551, 71, 10.1038/nature24290

    Troja , E., Piro , L., van Eerten , H., et al. 2017, , 551, 71, 10.1038/nature24290

  107. [115]

    2018, Nature Communications, 9, 4089, 10.1038/s41467-018-06558-7

    Troja , E., Ryan , G., Piro , L., et al. 2018, Nature Communications, 9, 4089, 10.1038/s41467-018-06558-7

  108. [116]

    J., Becerra Gonz \'a lez , J., et al

    Troja , E., Castro-Tirado , A. J., Becerra Gonz \'a lez , J., et al. 2019, , 489, 2104, 10.1093/mnras/stz2255

  109. [117]

    L., O'Connor , B., et al

    Troja , E., Fryer , C. L., O'Connor , B., et al. 2022, , 612, 228, 10.1038/s41586-022-05327-3

  110. [118]

    2016, arXiv e-prints, arXiv:1607.08022, 10.48550/arXiv.1607.08022

    Ulyanov , D., Vedaldi , A., & Lempitsky , V. 2016, arXiv e-prints, arXiv:1607.08022, 10.48550/arXiv.1607.08022

  111. [119]

    2008, Journal of Machine Learning Research, 9, 2579

    van der Maaten, L., & Hinton, G. 2008, Journal of Machine Learning Research, 9, 2579. http://jmlr.org/papers/v9/vandermaaten08a.html

  112. [120]

    Veres , P., Bagoly , Z., Horv \'a th , I., M \'e sz \'a ros , A., & Bal \'a zs , L. G. 2010, , 725, 1955, 10.1088/0004-637X/725/2/1955

  113. [121]

    N., Burns , E., et al

    Veres , P., Bhat , P. N., Burns , E., et al. 2023, , 954, L5, 10.3847/2041-8213/ace82d

  114. [122]

    J., Younk , P., et al

    Vianello , G., Lauer , R. J., Younk , P., et al. 2015, arXiv e-prints, arXiv:1507.08343, 10.48550/arXiv.1507.08343

  115. [123]

    E., et al

    Virtanen , P., Gommers , R., Oliphant , T. E., et al. 2020, Nature Methods, 17, 261, 10.1038/s41592-019-0686-2

  116. [124]

    I., Yu , Y.-W., Ren , J., et al

    Wang , X. I., Yu , Y.-W., Ren , J., et al. 2024, , 964, L9, 10.3847/2041-8213/ad2df6

  117. [125]

    A., Ruffini , R., et al

    Wang , Y., Rueda , J. A., Ruffini , R., et al. 2019, , 874, 39, 10.3847/1538-4357/ab04f8

  118. [126]

    Woo , S., Park , J., Lee , J.-Y., & Kweon , I. S. 2018, arXiv e-prints, arXiv:1807.06521, 10.48550/arXiv.1807.06521

  119. [127]

    Woosley , S. E. 1993, , 405, 273, 10.1086/172359

  120. [128]

    E., & Bloom , J

    Woosley , S. E., & Bloom , J. S. 2006, , 44, 507, 10.1146/annurev.astro.43.072103.150558

  121. [129]

    2013, , 776, 98, 10.1088/0004-637X/776/2/98

    Xu , D., de Ugarte Postigo , A., Leloudas , G., et al. 2013, , 776, 98, 10.1088/0004-637X/776/2/98

  122. [130]

    2022, , 612, 232, 10.1038/s41586-022-05403-8

    Yang , J., Ai , S., Zhang , B.-B., et al. 2022, , 612, 232, 10.1038/s41586-022-05403-8

  123. [131]

    2024, , 626, 742, 10.1038/s41586-023-06979-5

    Yang , Y.-H., Troja , E., O'Connor , B., et al. 2024, , 626, 742, 10.1038/s41586-023-06979-5

  124. [132]

    2019, , 886, 20, 10.3847/1538-4357/ab488a

    Yu , H.-F., Dereli-B \'e gu \'e , H., & Ryde , F. 2019, , 886, 20, 10.3847/1538-4357/ab488a

  125. [133]

    2016 a , , 202, 3, 10.1007/s11214-016-0305-9

    Zhang , B., L \"u , H.-J., & Liang , E.-W. 2016 a , , 202, 3, 10.1007/s11214-016-0305-9

  126. [134]

    2007, , 655, L25, 10.1086/511781

    Zhang , B., Zhang , B.-B., Liang , E.-W., et al. 2007, , 655, L25, 10.1086/511781

  127. [135]

    J., et al

    Zhang , B., Zhang , B.-B., Virgili , F. J., et al. 2009, , 703, 1696, 10.1088/0004-637X/703/2/1696

  128. [136]

    B., Liu , Z

    Zhang , B. B., Liu , Z. K., Peng , Z. K., et al. 2021, Nature Astronomy, 5, 911, 10.1038/s41550-021-01395-z

  129. [137]

    2022, , 517, 5770, 10.1093/mnras/stac3131

    Zhang , L., Luo , J.-J., Huang , Y.-F., Gong , Y.-J., & Wu , S. 2022, , 517, 5770, 10.1093/mnras/stac3131

  130. [138]

    2024, , 272, 4, 10.3847/1538-4365/ad2de5

    Zhang , P., Li , B., Gui , R., et al. 2024, , 272, 4, 10.3847/1538-4365/ad2de5

  131. [139]

    2016 b , , 462, 3243, 10.1093/mnras/stw1835

    Zhang , Z.-B., Yang , E.-B., Choi , C.-S., & Chang , H.-Y. 2016 b , , 462, 3243, 10.1093/mnras/stw1835

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.