Pith. sign in

REVIEW 3 major objections 7 minor 69 references

Heartbeat Stars Recognition Based on Recurrent Neural Networks: Method and Validation

T0 review · 3 major / 7 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Heartbeat stars can be found automatically by feeding the first 100 orbital harmonics of a light curve into a small recurrent network, which predicts eccentricity and works on real survey data at 86% accuracy.

desk verdict A genuinely cheap and novel harmonic-feature RNN for eccentricity regression, but the '86% accuracy' is a regression consistency score on known positives, not a detection rate, and the discovery claim rests on an unvalidated eclipse-count step. read the letter →

arxiv 2505.04067 v2 pith:TRPC56IE submitted 2025-05-07 astro-ph.SR astro-ph.IM

Min-Yu Li , Sheng-Bang Qian , Li-Ying Zhu , Wen-Ping Liao , Lin-Feng Chang , Er-Gang Zhao , Xiang-Dong Shi , Fu-Xing Li
show 2 more authors
Qi-Bin Sun Ping Li
This is my paper · ORCID
classification astro-ph.SRastro-ph.IM
keywords heartbeatstarsrecurrentneuralnetworksorbitalharmonicslightcurveclassificationeccentricbinariestidallyexcitedoscillationssyntheticcurvesKeplereclipsing
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

Heartbeat stars are binaries whose eccentric orbits produce a distinctive heartbeat-like pulse in their light curves near periastron, and most known examples were found by eye. This paper argues that they can be recognized automatically: take the Fourier spectrum of a light curve, keep the amplitudes of the first 100 orbital harmonics, normalize them, and feed the resulting vector into a small recurrent network that predicts orbital eccentricity. Trained on 52,000 synthetic light curves generated with the ELLC code, the GRU and LSTM networks reach 95% accuracy on synthetic tests and agree with published eccentricities within 0.15 for 86% of real heartbeat stars from OGLE, Kepler, and TESS. Applied to the Kepler eclipsing-binary catalog of Kirk et al., the method surfaced four new heartbeat stars, one of which has a model-fitted eccentricity matching the network prediction. If correct, this makes large-scale automated searches for heartbeat stars practical and could extend to other periodic variables.

What carries the argument

The load-bearing object is the normalized orbital-harmonic feature vector: after locating the orbital frequency, the first 100 harmonic amplitudes extracted by FNPEAKS are divided by their Euclidean norm, so each light curve becomes a 100-number sequence. This vector is what the recurrent network sees, and it matters because it compresses a roughly 30,000-point light curve into a small ordered set, discards the orbital period and absolute flux scale, and lets a two-layer GRU or LSTM rather than a deep network predict eccentricity. The orbital-frequency finder is a second piece of machinery, since the network cannot be applied to real data until the harmonic grid is known; it works by testing which integer multiple of the strongest peak has the most harmonic matches, or, for pulsating systems, by finding the most common spacing between adjacent Fourier peaks.

What would settle it

Run the two trained networks on all 173 Kepler heartbeat stars listed in the Kirk et al. catalog and compare every predicted eccentricity with the published value; if significantly fewer than 86% of predictions land within 0.15, the claimed synthetic-to-real generalization would fail.

Watch

Extended reading notes

Core claim

The paper's central claim is that the morphology of a heartbeat-star light curve is encoded, for classification purposes, in the amplitudes of the first 100 orbital harmonics of its Fourier spectrum, and that a recurrent network reading only those amplitudes can recover the orbital eccentricity. The authors construct 52,000 synthetic light curves with ELLC by randomizing eccentricity, inclination, argument of periastron, mass ratio, surface brightness ratio, and radii, plus 2,000 zero-eccentricity close binaries to teach the network what is not a heartbeat star. They define a harmonic as a Fourier peak with signal-to-noise at least 4 that lies within 0.05 of an integer multiple of the orbital frequency, normalize the amplitude vector by its Euclidean norm, and train two-layer GRU and LSTM networks to output eccentricity. On held-out synthetic data the test accuracy is 95%; on real heartbeat stars from OGLE, Kepler, and TESS the networks agree with literature eccentricities to within 0.15 in 86% of cases; and within the Kirk et al. eclipsing-binary catalog the method finds four new heartbeat stars (KIC 4940438, 6794131, 7601633, 9243795), with KIC 6794131's model-fitted eccentricity of 0.179 close to the predicted 0.171.

Load-bearing premise

The load-bearing premise is that 52,000 synthetic ELLC light curves, generated with Gaussian noise but without tidally excited oscillations, are representative enough of real survey light curves that a network trained only on them will recognize real heartbeat stars.

Editorial extensions

If this is right

  • Survey archives can be pre-filtered automatically: reducing each light curve to 100 harmonic amplitudes lets a two-layer recurrent network rank large numbers of candidates by predicted eccentricity before any human inspection.
  • The same networks recognize eccentric binaries as well as heartbeat stars, and the two classes can be roughly separated by counting eclipses in the phase-folded light curve, so one pipeline can serve both searches.
  • Because the orbital frequency is computed automatically, the method does not need a known period to extract features, allowing application to any periodic variable whose orbital harmonics are resolved.
  • Visual inspection effort drops from hundreds of thousands of objects to a few hundred candidates, as demonstrated by finding about 1,200 high-eccentricity candidates and then four new heartbeat stars inside the Kirk et al. catalog.

Reading between the lines

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

  • The reported 86% is an agreement rate with published eccentricities on known heartbeat stars, not a completeness or purity measurement against a labeled background; a realistic survey deployment would need a false-positive estimate on non-heartbeat stars.
  • If the harmonic envelope is the main carrier of information, a natural test is whether a single interpretable statistic, such as the slope of harmonic amplitude versus harmonic number, correlates with eccentricity; if it does, the recurrent network could be distilled into a simpler explainable detector.
  • Injecting tidally excited oscillations into the synthetic training set is an obvious robustness upgrade; the paper shows the network tolerates their absence in training, which suggests including them could improve high-eccentricity cases where these oscillations are strong.
  • The method's period independence suggests the same trained networks could be applied to future surveys with different cadences, provided the Fourier spectrum resolves the orbital harmonics; that extension is untested and would be a useful external validation.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 7 minor

Summary. The paper proposes a machine-learning method for recognizing heartbeat stars (HBSs). Light curves are first transformed into Fourier spectra, from which the amplitudes of the first 100 orbital harmonics are extracted and normalized into a 100-dimensional feature vector. Two recurrent neural networks (GRU and LSTM) are trained on 52,000 synthetic light curves generated with ELLC to regress orbital eccentricity. The trained networks are then applied to real HBSs from OGLE, Kepler, and TESS, and to a catalog of eclipsing binaries from Kirk et al. The authors report 95% accuracy on a synthetic test set, 86% 'accuracy prediction consistency' on real data, and the identification of four new HBS candidates, one of which is fitted with a K95+ model. The paper emphasizes that the harmonic-feature approach reduces computational cost and can be extended to other periodic variables.

Significance. If the claims hold, the work is a useful step toward automated HBS discovery: the features are physically motivated and compact, the training is entirely synthetic so there is no circularity in the labels, the code is public, and the method is tested on external real-data catalogs. The realistic significance is, however, lower than the abstract suggests, because the headline 86% figure is a regression-consistency fraction, not a detection accuracy, and the four new candidates are not independently confirmed. The central architecture is sound and reproducible, and the main gaps are quantifiable and fixable within the scope of the manuscript.

major comments (3)
  1. [Sect. 3.1, abstract, conclusions] The headline '86% accuracy' is not an HBS detection accuracy: it is the fraction of known HBSs and eccentric binaries whose predicted eccentricity lies within ±0.15 of the literature value. In addition, five Kepler HBSs with weak heartbeat signals were excluded from the test set. The abstract and summary currently call this 'average detection accuracy', which overstates what is demonstrated. Please relabel the metric (e.g., regression consistency within 0.15), report per-dataset sizes and per-dataset fractions, and add actual binary classification metrics (precision/recall/F1 for the e>0.1 threshold) on the labeled test sets, ideally with a sensitivity analysis to the excluded weak-signal systems.
  2. [Sect. 3.2] The discovery pipeline is not validated as a detector. The paragraph in Sect. 3.2 states that systems with no eclipses or one eclipse are classified as HBSs, but also that this strategy is 'not entirely rigorous'; it then converts roughly 900 candidates into four HBSs without reporting precision, recall, or false-positive rates on any labeled sample. A grazing-eclipsing eccentric binary can show a single eclipse and would be misclassified by this rule. Please run the full pipeline on a labeled subset of the Kirk catalog (known HBSs, known eccentric binaries, circular EBs) and report a confusion matrix. In addition, only KIC 6794131 receives a K95+ fit; the other three candidates need independent confirmation (e.g., model fits to all four, additional photometry or radial velocities) before the paper can claim they are 'newly identified HBSs'.
  3. [Sect. 2.1 vs Sect. 4] The training set contains no tidally excited oscillations, yet Sect. 4 asserts robustness to TEOs based on the aggregate real-data performance. This generalization claim is plausible but unquantified: TEOs add harmonic power at specific frequencies and can change the normalized harmonic vector. Please report the consistency metric separately for known TEO-bearing and non-TEO HBSs in Fig. 4, and/or test the trained networks on synthetic light curves with injected TEO-like harmonic amplitudes.
minor comments (7)
  1. [Abstract] The first sentence contains a grammar issue: 'Since the variety of their light curve morphologies' should be 'Because of the variety of their light curve morphologies' (or similar).
  2. [Sect. 2.4] There is a duplicated word in 'the orbital orbital frequency'; it should be 'orbital frequency'.
  3. [Sect. 2.2 and 2.3] The feature vector is defined in Eq. (2) as (a1,...,an), but the network uses the first 100 harmonics; please clarify how vectors with fewer detected harmonics are handled and whether zero-padding or truncation is applied.
  4. [Sect. 2.3 and 3.1] The 0.02 deviation threshold used to define training accuracy and the 0.15 tolerance used for real-data consistency should be stated explicitly as the metric definitions; currently the term 'accuracy' is used in both places without a formal definition.
  5. [Sect. 2.3] The claim that 100 input units were chosen after 'extensive testing' would be more convincing with a small figure showing accuracy versus the number of harmonics used.
  6. [Sect. 3.1] The sentence 'we suggest classifying them as either type' for systems with both two eclipses and heartbeat signals is ambiguous; please specify the intended classification convention.
  7. [Title] The title contains an unusual space in 'V alidation'; this appears to be a formatting artifact and should be corrected.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: synthetic-trained RNN is validated against external catalogs; self-citations are motivational, not load-bearing.

full rationale

The derivation chain is not circular. Section 2.1 creates 52,000 synthetic light curves with ELLC using randomly drawn parameters (ecc up to 0.9, incl 20–90 deg, etc.); the supervised target is the known eccentricity used to generate each model, so the network is fitted to forward-model labels rather than to the real catalogs it later tests against. Section 2.3's 95% accuracy is an in-distribution check on held-out synthetic curves, not a claim that the synthetic set defines the real answer. The 86% figure in Section 3.1 is explicitly a comparison between predicted eccentricity and published eccentricities from the OGLE, Kepler, and TESS catalogs; those external references are not inputs to training, and the paper states that OGLE orbital frequencies were adopted from the literature because the automatic period pipeline failed there. Section 3.2's four new HBS candidates are selected by an e>0.1 threshold plus visual and eclipse-count inspection; KIC 6794131 is additionally fitted with a K95+ model using the authors' own earlier methodology, but the fit is used as an after-the-fact validation, not as the training target. The only self-referential element is the motivational use of the authors' Li et al. 2024d harmonic analysis and the relaxed harmonic-matching threshold quoted from Eq. (1); these are methodological antecedents, not uniqueness theorems or fitted labels, and the central eccentricity-regression claim rests on independent external catalogs. Limitations exist—synthetic training omits TEOs, the eclipse-count classification is explicitly called 'not entirely rigorous', and only one of four candidates is model-fitted—but these are completeness and validation caveats, not circular reasoning. Hence no circular step can be exhibited.

Assumptions & free parameters 5 free parameters · 3 assumptions · 0 invented entities

All key quantities rest on several free thresholds and domain assumptions; the network itself learns from synthetic labels, but the feature extraction and classification thresholds are chosen by hand.

free parameters (5)
  • Number of harmonic features = 100
    The first 100 harmonics are used as the feature vector; the authors state that this was chosen through extensive testing as optimal (Sect. 2.3). Results depend on this truncation.
  • Eccentricity classification threshold = 0.1
    In Sect. 3.2, candidates with both networks predicting eccentricity greater than 0.1 are classified as HBS or eccentric binary. This threshold is chosen by hand and not justified with a sensitivity analysis.
  • S/N threshold for harmonic detection = 4.0
    Peaks with S/N >= 4.0 are considered harmonic candidates (Sect. 2.2). This is an arbitrary detection threshold affecting feature extraction.
  • Harmonic matching tolerance = 0.05
    A frequency must satisfy |n - f/f_orb| < 0.05 to be a harmonic (Eq. 1). This tolerance affects how many harmonics are extracted.
  • Real-data consistency tolerance = 0.15
    The 86% accuracy on real data is defined as predictions within 0.15 of the literature eccentricity (Sect. 3.1). This tolerance is a post hoc evaluation choice rather than a pre-specified error metric.
assumptions (3)
  • domain assumption ELLC model accurately simulates HBS light curves in the explored parameter space.
    Synthetic training samples are generated with ELLC (Sect. 2.1); if the model misses real morphology (TEOs, spots, etc.), the network may not generalize. The authors note they did not include TEOs in training.
  • domain assumption Heartbeat star light curves are fully characterized by their orbital Fourier harmonics.
    The method assumes that non-harmonic frequencies are negligible for classification (Sect. 2.2). This is justified by periodicity but not proven for noisy or pulsating systems.
  • domain assumption The orbital frequency can be recovered automatically from the Fourier spectra.
    Sect. 2.4 describes a heuristic search; the 90% success on Kepler/TESS and lower on OGLE shows this assumption is not always met.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Heartbeat Stars Recognition Based on Recurrent Neural Networks: Method and Validation." pith.science (2026). https://pith.science/paper/TRPC56IE

@misc{pith2026250504067,
  author       = {Pith},
  title        = {Pith review of: Heartbeat Stars Recognition Based on Recurrent Neural Networks: Method and Validation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/TRPC56IE}},
  note         = {Machine review of arXiv:2505.04067}
}
abstract

Since the variety of their light curve morphologies, the vast majority of the known heartbeat stars (HBSs) have been discovered by manual inspection. Machine learning, which has already been successfully applied to the classification of variable stars based on light curves, offers another possibility for the automatic detection of HBSs. We propose a novel feature extraction approach for HBSs. First, the orbital frequencies are calculated automatically according to the Fourier spectra of the light curves. Then, the amplitudes of the first 100 harmonics are extracted. Finally, these harmonics are normalized as feature vectors of the light curve. A training data set of synthetic light curves is constructed using ELLC, and their features are fed into recurrent neural networks (RNNs) for supervised learning, with the expected output being the eccentricity of these light curves. The performance of the RNNs is evaluated using a test data set of synthetic light curves, achieving 95$\%$ accuracy. When applied to known HBSs from the OGLE, Kepler, and TESS surveys, the networks achieve an average accuracy of 86$\%$. This method successfully identifies four new HBSs within the eclipsing binary catalog of Kirk et al. The use of orbital harmonics as features for HBSs proves to be a practical approach that significantly reduces the computational cost of neural networks. RNNs show excellent performance in recognizing this type of time series data. This method not only allows efficient identification of HBSs but can also be extended to recognize other types of periodic variable stars.

Figures

Figures reproduced from arXiv: 2505.04067 by the authors.

Figure 1
Figure 1. Two examples of the synthetic light curves and their Fourier spectra. Panels (a1) and (a2) show the non-eclipsing and eclipsing HBSs, respectively. Panels (b1) and (b2) show the corresponding Fourier spectra. The red vertical dashed lines represent the orbital harmonics. 2. METHODOLOGY 2.1. Samples generation To obtain high-resolution Fourier spectra from synthetic light curves, the data must contain a sufficient nu… view at source ↗
Figure 2
Figure 2. The architecture of the two neural networks. Each network consists of two recurrent layers (either GRU or LSTM) of 50 units each, interleaved with a dropout layer (rate=0.2). The input layer processes 100-dimensional features, and a one-unit dense layer generates the output. 2.2. Features extraction The data points of a light curve and its Fourier spectrum are mathematically equivalent. For HBSs, the periodicity fea… view at source ↗
Figure 3
Figure 3. Training process for the GRU and LSTM networks. The left and right two panels show the loss and accuracy curves, respectively. Blue solid and orange dashed lines correspond to the training and validation sets, respectively. Since the feature dimensions have been reduced, a highly deep neural network architecture becomes unnecessary. As shown in [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: The first and second rows show the test results of the GRU and LSTM networks, respectively, on real data sets. The column headings denote the names of the four test data sets. In each panel, the x-axis represents the eccentricity from the reference papers; the y-axis r…
Figure 5
Figure 5. Figure 5: Four new HBSs. The orbital period (Pclc) is derived using the approach in Sect. 2.4. The eccentricity (enn) is predicted by our RNNs. For KIC 6794131, the solid red line shows the K95+ model fitted to the light curves, and the residuals are also shown. The eccentricity…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

69 extracted references · 20 canonical work pages

  1. [1]

    2019, MNRAS, 482, 5078, doi: 10.1093/mnras/sty2836

    Aguirre, C., Pichara, K., & Becker, I. 2019, MNRAS, 482, 5078, doi: 10.1093/mnras/sty2836

    doi:10.1093/mnras/sty2836 2019

  2. [2]

    J., Kirk, J., Lam, K

    Armstrong, D. J., Kirk, J., Lam, K. W. F., et al. 2016, MNRAS, 456, 2260, doi: 10.1093/mnras/stv2836

    doi:10.1093/mnras/stv2836 2016

  3. [3]

    S., Handberg, R., et al

    Audenaert, J., Kuszlewicz, J. S., Handberg, R., et al. 2021, AJ, 162, 209, doi: 10.3847/1538-3881/ac166a

    doi:10.3847/1538-3881/ac166a 2021

  4. [4]

    H., Bedding, T

    Barbara, N. H., Bedding, T. R., Fulcher, B. D., Murphy, S. J., & Van Reeth, T. 2022, MNRAS, 514, 2793, doi: 10.1093/mnras/stac1515

    doi:10.1093/mnras/stac1515 2022

  5. [5]

    G., Hambleton, K., Vos, J., et al

    Beck, P. G., Hambleton, K., Vos, J., et al. 2014, A&A, 564, A36, doi: 10.1051/0004-6361/201322477

    doi:10.1051/0004-6361/201322477 2014

  6. [6]

    2020, MNRAS, 493, 2981, doi: 10.1093/mnras/staa350

    Becker, I., Pichara, K., Catelan, M., et al. 2020, MNRAS, 493, 2981, doi: 10.1093/mnras/staa350

    doi:10.1093/mnras/staa350 2020

  7. [7]

    2014, MNRAS, 443, 3068, doi: 10.1093/mnras/stu1379

    Borkovits, T., Derekas, A., Fuller, J., et al. 2014, MNRAS, 443, 3068, doi: 10.1093/mnras/stu1379

    doi:10.1093/mnras/stu1379 2014

  8. [8]

    J., Miglio, A., et al

    Brogaard, K., Hansen, C. J., Miglio, A., et al. 2018, MNRAS, 476, 3729, doi: 10.1093/mnras/sty268 6 https://github.com/MinyuLi/HBSsNN

    doi:10.1093/mnras/sty268 2018

Show all 69 references
  1. [9]

    J., Fuller, J., Guo, Z., Lehman, H., & Hambleton, K

    Cheng, S. J., Fuller, J., Guo, Z., Lehman, H., & Hambleton, K. 2020, ApJ, 903, 122, doi: 10.3847/1538-4357/abb46d

    2020 doi

  2. [10]

    2014, arXiv e-prints, arXiv:1406.1078, doi: 10.48550/arXiv.1406.1078

    Cho, K., van Merrienboer, B., Gulcehre, C., et al. 2014, arXiv e-prints, arXiv:1406.1078, doi: 10.48550/arXiv.1406.1078

    2014 doi arXiv

  3. [11]

    J., & Feng, F

    Cui, K., Armstrong, D. J., & Feng, F. 2024, ApJS, 274, 29, doi: 10.3847/1538-4365/ad62fd

    2024 doi

  4. [12]

    Elizabethson, A., Serna, J., Garc´ ıa-Varela, A., Hern´ andez, J., & Cabrera-Garc´ ıa, J. F. 2023, AJ, 166, 189, doi: 10.3847/1538-3881/acf865

    2023 doi

  5. [13]

    2017, MNRAS, 472, 1538, doi: 10.1093/mnras/stx2135

    Fuller, J. 2017, MNRAS, 472, 1538, doi: 10.1093/mnras/stx2135

    2017 doi

  6. [14]

    2017, MNRAS, 472, L25, doi: 10.1093/mnrasl/slx130

    Thompson, S. 2017, MNRAS, 472, L25, doi: 10.1093/mnrasl/slx130

    2017 doi

  7. [15]

    2025, ApJS, 276, 57, doi: 10.3847/1538-4365/ad9dd6

    Gao, X., Chen, X., Wang, S., & Liu, J. 2025, ApJS, 276, 57, doi: 10.3847/1538-4365/ad9dd6

    2025 doi

  8. [16]

    2016, ApJ, 832, 121, doi: 10.3847/0004-637X/832/2/121 10

    Gaulme, P., McKeever, J., Jackiewicz, J., et al. 2016, ApJ, 832, 121, doi: 10.3847/0004-637X/832/2/121 10

    2016 doi

  9. [17]

    2024, MNRAS, 528, 6997, doi: 10.1093/mnras/stae404

    Guo, F., Lin, J., Wang, X., et al. 2024, MNRAS, 528, 6997, doi: 10.1093/mnras/stae404

    2024 doi

  10. [18]

    2021, FrASS, 8, 67, doi: 10.3389/fspas.2021.663026

    Guo, Z. 2021, FrASS, 8, 67, doi: 10.3389/fspas.2021.663026

    2021

  11. [19]

    2019, ApJ, 885, 46, doi: 10.3847/1538-4357/ab41f6

    Guo, Z., Fuller, J., Shporer, A., et al. 2019, ApJ, 885, 46, doi: 10.3847/1538-4357/ab41f6

    2019 doi

  12. [20]

    R., & Fuller, J

    Guo, Z., Gies, D. R., & Fuller, J. 2017, ApJ, 834, 59, doi: 10.3847/1538-4357/834/1/59

    2017 doi

  13. [21]

    2020, ApJ, 888, 95, doi: 10.3847/1538-4357/ab58c2

    Guo, Z., Shporer, A., Hambleton, K., & Isaacson, H. 2020, ApJ, 888, 95, doi: 10.3847/1538-4357/ab58c2

    2020 doi

  14. [22]

    W., Prˇ sa, A., et al

    Hambleton, K., Kurtz, D. W., Prˇ sa, A., et al. 2016, MNRAS, 463, 1199, doi: 10.1093/mnras/stw1970

    2016 doi

  15. [23]

    2018, MNRAS, 473, 5165, doi: 10.1093/mnras/stx2673

    Hambleton, K., Fuller, J., Thompson, S., et al. 2018, MNRAS, 473, 5165, doi: 10.1093/mnras/stx2673

    2018 doi

  16. [24]

    M., Kurtz, D

    Hambleton, K. M., Kurtz, D. W., Prˇ sa, A., et al. 2013, MNRAS, 434, 925, doi: 10.1093/mnras/stt886

    2013 doi

  17. [25]

    2015, arXiv e-prints, arXiv:1502.01852, doi: 10.48550/arXiv.1502.01852 He lminiak, K

    He, K., Zhang, X., Ren, S., & Sun, J. 2015, arXiv e-prints, arXiv:1502.01852, doi: 10.48550/arXiv.1502.01852 He lminiak, K. G., Ukita, N., Kambe, E., et al. 2016, MNRAS, 461, 2896, doi: 10.1093/mnras/stw1514

    2015 doi arXiv

  18. [26]

    A., Tat, K., & Thorp, R

    Hinners, T. A., Tat, K., & Thorp, R. 2018, AJ, 156, 7, doi: 10.3847/1538-3881/aac16d

    2018 doi

  19. [27]

    1997, Neural Computation, 9, 1735, doi: 10.1162/neco.1997.9.8.1735

    Hochreiter, S., & Schmidhuber, J. 1997, Neural Computation, 9, 1735, doi: 10.1162/neco.1997.9.8.1735

    1997 doi

  20. [28]

    2018a, MNRAS, 476, 3233, doi: 10.1093/mnras/sty483

    Hon, M., Stello, D., & Yu, J. 2018a, MNRAS, 476, 3233, doi: 10.1093/mnras/sty483

    doi

  21. [29]

    Hon, M., Stello, D., & Zinn, J. C. 2018b, ApJ, 859, 64, doi: 10.3847/1538-4357/aabfdb

    doi

  22. [30]

    Jamal, S., & Bloom, J. S. 2020, ApJS, 250, 30, doi: 10.3847/1538-4365/aba8ff

    2020 doi

  23. [31]

    S., Strader, J., et al

    Jayasinghe, T., Kochanek, C. S., Strader, J., et al. 2021, MNRAS, 506, 4083, doi: 10.1093/mnras/stab1920

    2021 doi

  24. [32]

    B., Caballero-Nieves, S

    Johnston, K. B., Caballero-Nieves, S. M., Petit, V., Peter, A. M., & Haber, R. 2020, MNRAS, 491, 3805, doi: 10.1093/mnras/stz3165

    2020 doi

  25. [33]

    2016, Aj, 151, 68, doi: 10.3847/0004-6256/151/3/68

    Kirk, B., Conroy, K., Prˇ sa, A., et al. 2016, Aj, 151, 68, doi: 10.3847/0004-6256/151/3/68

    2016 doi

  26. [34]

    2015a, PASA, 32, e023, doi: 10.1017/pasa.2015.23

    Kjurkchieva, D., & Vasileva, D. 2015a, PASA, 32, e023, doi: 10.1017/pasa.2015.23

    2015 doi

  27. [35]

    2015b, Bulgarian Astronomical Journal, 23, 75, doi: 10.48550/arXiv.1507.06072

    Kjurkchieva, D., & Vasileva, D. 2015b, Bulgarian Astronomical Journal, 23, 75, doi: 10.48550/arXiv.1507.06072

    doi arXiv

  28. [36]

    2016, NewA, 48, 30, doi: 10.1016/j.newast.2016.04.004

    Kjurkchieva, D., & Vasileva, D. 2016, NewA, 48, 30, doi: 10.1016/j.newast.2016.04.004

    2016 doi

  29. [37]

    2016a, Ap&SS, 361, 132, doi: 10.1007/s10509-016-2722-3

    Kjurkchieva, D., Vasileva, D., & Dimitrov, D. 2016a, Ap&SS, 361, 132, doi: 10.1007/s10509-016-2722-3

    doi

  30. [38]

    2016b, AJ, 152, 189, doi: 10.3847/0004-6256/152/6/189

    Kjurkchieva, D., Vasileva, D., & Dimitrov, D. 2016b, AJ, 152, 189, doi: 10.3847/0004-6256/152/6/189

    doi

  31. [39]

    2024, A&A, 685, A145, doi: 10.1051/0004-6361/202349075 Ko laczek-Szyma´ nski, P

    Koenigsberger, G., & Estrella-Trujillo, D. 2024, A&A, 685, A145, doi: 10.1051/0004-6361/202349075 Ko laczek-Szyma´ nski, P. A., Lojko, P., Pigulski, A., R´ o˙ za´ nski, T., & Mo´ zdzierski, D. 2024, A&A, 686, A199, doi: 10.1051/0004-6361/202348104 Ko laczek-Szyma´ nski, P. A.,...

    2024 doi

  32. [40]

    2022, A&A, 659, A47, doi: 10.1051/0004-6361/202142171 Ko laczek-Szyma´ nski, P

    Ratajczak, M., & Udalski, A. 2022, A&A, 659, A47, doi: 10.1051/0004-6361/202142171 Ko laczek-Szyma´ nski, P. A., & R´ o˙ za´ nski, T. 2023, A&A, 671, A22, doi: 10.1051/0004-6361/202245226

    2022 doi

  33. [41]

    O., & Quataert, E

    Kumar, P., Ao, C. O., & Quataert, E. J. 1995, ApJ, 449, 294, doi: 10.1086/176055

    1995 doi

  34. [42]

    2023, ApJS, 266, 28, doi: 10.3847/1538-4365/acca13

    Li, M.-Y., Qian, S.-B., Zhu, L.-Y., et al. 2023, ApJS, 266, 28, doi: 10.3847/1538-4365/acca13

    2023 doi

  35. [43]

    2024a, ApJ, 974, 278, doi: 10.3847/1538-4357/ad794c

    Li, M.-Y., Qian, S.-B., Zhou, A.-Y., et al. 2024a, ApJ, 974, 278, doi: 10.3847/1538-4357/ad794c

    doi

  36. [44]

    2024b, MNRAS, 530, 586, doi: 10.1093/mnras/stae885

    Li, M.-Y., Qian, S.-B., Zhu, L.-Y., et al. 2024b, MNRAS, 530, 586, doi: 10.1093/mnras/stae885

    doi

  37. [45]

    2024c, MNRAS, 534, 281, doi: 10.1093/mnras/stae2057

    Li, M.-Y., Qian, S.-B., Zhou, A.-Y., et al. 2024c, MNRAS, 534, 281, doi: 10.1093/mnras/stae2057

    doi

  38. [46]

    2024d, ApJ, 962, 44, doi: 10.3847/1538-4357/ad18c1

    Li, M.-Y., Qian, S.-B., Zhu, L.-Y., et al. 2024d, ApJ, 962, 44, doi: 10.3847/1538-4357/ad18c1

    doi

  39. [47]

    2025, PASJ, 77, 118, doi: 10.1093/pasj/psae103

    Li, M.-Y., Qian, S.-B., Zhu, L.-Y., et al. 2025, PASJ, 77, 118, doi: 10.1093/pasj/psae103

    2025 doi

  40. [48]

    2014, A&A, 563, A59, doi: 10.1051/0004-6361/201322871

    Maceroni, C., Lehmann, H., da Silva, R., et al. 2014, A&A, 563, A59, doi: 10.1051/0004-6361/201322871

    2014 doi

  41. [49]

    2023, Nature Astronomy, 7, 1218, doi: 10.1038/s41550-023-02036-3

    MacLeod, M., & Loeb, A. 2023, Nature Astronomy, 7, 1218, doi: 10.1038/s41550-023-02036-3

    2023 doi

  42. [50]

    2025, arXiv e-prints, arXiv:2503.17133, doi: 10.48550/arXiv.2503.17133

    MacLeod, M., & Loeb, A. 2025, arXiv e-prints, arXiv:2503.17133, doi: 10.48550/arXiv.2503.17133

    2025 doi arXiv

  43. [51]

    Z., Tramper, F., et al

    Maravelias, G., Bonanos, A. Z., Tramper, F., et al. 2022, A&A, 666, A122, doi: 10.1051/0004-6361/202141397

    2022 doi

  44. [52]

    Maxted, P. F. L. 2016, A&A, 591, A111, doi: 10.1051/0004-6361/201628579

    2016 doi

  45. [53]

    2024, A&A, 691, A106, doi: 10.1051/0004-6361/202449995

    Monsalves, N., Jaque Arancibia, M., Bayo, A., et al. 2024, A&A, 691, A106, doi: 10.1051/0004-6361/202449995

    2024 doi

  46. [54]

    S., P´ erez, F., & van der Walt, S

    Naul, B., Bloom, J. S., P´ erez, F., & van der Walt, S. 2018, Nature Astronomy, 2, 151, doi: 10.1038/s41550-017-0321-z O’Leary, R. M., & Burkart, J. 2014, MNRAS, 440, 3036, doi: 10.1093/mnras/stu335

    2018 doi

  47. [55]

    P., Ansdell, M., Ioannou, Y., et al

    Osborn, H. P., Ansdell, M., Ioannou, Y., et al. 2020, A&A, 633, A53, doi: 10.1051/0004-6361/201935345

    2020 doi

  48. [56]

    2021a, MNRAS, 508, 3967, doi: 10.1093/mnras/stab2805

    Ou, J.-W., Yu, C., Jiang, C., Yang, M., & Niu, H. 2021a, MNRAS, 508, 3967, doi: 10.1093/mnras/stab2805

    doi

  49. [57]

    2021b, ApJ, 922, 37, doi: 10.3847/1538-4357/ac22b0

    Ou, J.-W., Yu, C., Yang, M., et al. 2021b, ApJ, 922, 37, doi: 10.3847/1538-4357/ac22b0

    doi

  50. [58]

    2021, AJ, 162, 67, doi: 10.3847/1538-3881/ac0824 11

    Qu, H., Sako, M., M¨ oller, A., & Doux, C. 2021, AJ, 162, 67, doi: 10.3847/1538-3881/ac0824 11

    2021 doi

  51. [59]

    L., Gaulme, P., McKeever, J., et al

    Rawls, M. L., Gaulme, P., McKeever, J., et al. 2016, ApJ, 818, 108, doi: 10.3847/0004-637X/818/2/108 S´ anchez-S´ aez, P., Reyes, I., Valenzuela, C., et al. 2021, AJ, 161, 141, doi: 10.3847/1538-3881/abd5c1

    2016 doi

  52. [60]

    2019, MNRAS, 483, 5534, doi: 10.1093/mnras/sty3146

    Schanche, N., Collier Cameron, A., H´ ebrard, G., et al. 2019, MNRAS, 483, 5534, doi: 10.1093/mnras/sty3146

    2019 doi

  53. [61]

    2016, ApJ, 829, 34, doi: 10.3847/0004-637X/829/1/34

    Shporer, A., Fuller, J., Isaacson, H., et al. 2016, ApJ, 829, 34, doi: 10.3847/0004-637X/829/1/34

    2016 doi

  54. [62]

    M., Schnittman, J., et al

    Solanki, S., Cieplak, A. M., Schnittman, J., et al. 2025, ApJS, 276, 17, doi: 10.3847/1538-4365/ad8a62

    2025 doi

  55. [63]

    E., Everett, M., Mullally, F., et al

    Thompson, S. E., Everett, M., Mullally, F., et al. 2012, ApJ, 753, 86, doi: 10.1088/0004-637X/753/1/86

    2012 doi

  56. [64]

    Tsang, B. T. H., & Schultz, W. C. 2019, ApJL, 877, L14, doi: 10.3847/2041-8213/ab212c Ula¸ s, B., Szklen´ ar, T., & Szab´ o, R. 2025, A&A, 695, A81, doi: 10.1051/0004-6361/202452020

    2019 doi

  57. [65]

    2021, A&A, 652, A107, doi: 10.1051/0004-6361/202141068

    Vida, K., B´ odi, A., Szklen´ ar, T., & Seli, B. 2021, A&A, 652, A107, doi: 10.1051/0004-6361/202141068

    2021 doi

  58. [66]

    A., Ratajczak, M., & Koz lowski, S

    Wrona, M., Ko laczek-Szyma´ nski, P. A., Ratajczak, M., & Koz lowski, S. 2022a, ApJ, 928, 135, doi: 10.3847/1538-4357/ac56e6

    doi

  59. [67]

    A., et al

    Wrona, M., Ratajczak, M., Ko laczek-Szyma´ nski, P. A., et al. 2022b, ApJS, 259, 16, doi: 10.3847/1538-4365/ac4018

    doi

  60. [68]

    Zahn, J. P. 1975, A&A, 41, 329

    1975

  61. [69]

    Zhang, K., & Bloom, J. S. 2021, MNRAS, 505, 515, doi: 10.1093/mnras/stab1248

    2021 doi

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.