Pith. sign in

REVIEW 3 major objections 5 minor 53 references

A denoised ringdown analysis of 16 binary black hole mergers finds the remnant's dominant mode frequency and damping time consistent with general relativity, with combined fractional deviations within ±2.8% (frequency) and about ±8% (dampin

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · deepseek-v4-flash

2026-08-02 05:41 UTC pith:TCRWY7MX

load-bearing objection A genuinely new ringdown pipeline with careful injection work, but the headline precision rests on correction terms calibrated only in a narrow subspace; needs a direct comparison to prior constraints and a stated treatment of IMR circularity. the 3 major comments →

arxiv 2607.13275 v1 pith:TCRWY7MX submitted 2026-07-14 gr-qc

Precision Ringdown Measurements of Binary Black Hole Remnants

Achal Kumar , Poulami Dutta Roy , Marek J. Szczepa\'nczyk , Sergey Klimenko This is my paper
classification gr-qc
keywords ringdownquasinormal modesblack hole spectroscopygravitational wavesCoherent WaveBurstgeneral relativity testsGWTC-3binary black hole remnants
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The paper tries to establish that the ringdown of a binary black hole remnant can be measured with higher precision than previous analyses by fitting quasinormal modes to denoised, reconstructed signals rather than raw strain data. Using a cumulative-energy reference to start the fit closer to merger, it constrains the dominant (2,2,0) mode's frequency and damping time for 16 events from the third gravitational-wave transient catalog. All 16 measurements are consistent with the predictions of general relativity, and the combined deviations are δf = 0.003 ± 0.028 and δτ = 0.050 (+0.081/−0.086) at 90% confidence. If correct, this demonstrates a path to sharper strong-field tests of gravity as detector sensitivity grows.

Core claim

The central claim is that for all 16 binary black hole mergers analyzed, the measured frequency and damping time of the dominant (2,2,0) quasi-normal mode of the remnant are consistent with general relativity. The combined fractional deviations are δf220 = 0.003 ± 0.028 and δτ220 = 0.050 (+0.081/−0.086) at 90% confidence, both consistent with zero. The paper further claims that its cWB-reconstruction-based method yields tighter constraints than previous ringdown analyses because the cumulative-energy reference time lets the fit start at approximately 3.5 t_Mf after the merger, and because the denoising reduces the impact of non-Gaussian detector noise.

What carries the argument

The key mechanism is the denoised signal reconstruction by cWB—a coherent time-frequency pixel selection that estimates the signal without assuming a waveform model—followed by a two-mode damped-sinusoid fit for the (2,2,0) and (2,2,1) modes. The ringdown window is anchored not to the noisy signal peak but to the normalized cumulative energy e(t) = E(t)/E(T_end), set at 0.82, which reduces reference-time jitter. Correction factors calibrated on simulated injections account for reconstruction bias and allow the window to start at ~3.5 t_Mf after merger, with the systematic uncertainty budget validated through injection-based coverage studies.

Load-bearing premise

The correction that makes the early ringdown usable is calibrated only on equal-mass, non-spinning, face-on binary injections, yet is applied as a single ~5% shift to every real event, some of which have remnant spins up to 0.94 and unequal masses; if that mapping is off by a few percent, the quoted constraints shift by more than their intervals.

What would settle it

Calibrate the same correction pipeline on an injection set that mirrors the catalog's parameter space—unequal masses, aligned spins up to χ≈0.9, and nonzero inclination—then check whether the recovered δf220 and δτ220 for the injected signals remain within the 90% intervals; any systematic offset exceeding the quoted uncertainty would show the result is not universal.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • If the claimed precision is real, GR tests from ringdown alone can reach the sub-3% level in frequency with a modest number of events.
  • The cumulative-energy reference could be adopted by other ringdown analyses to reduce systematic uncertainty in the start time.
  • The approach applies directly to future, more sensitive detector networks, where the same method should produce even tighter bounds.
  • The overtone amplitude estimates provide a way to check the validity of the assumed linear perturbation theory.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The correction functions are derived from a narrow injection set (equal mass, non-spinning, face-on), yet are applied universally; a test with spinning, unequal-mass injections would reveal whether the claimed precision holds for all events in the catalog.
  • The combined constraint on the remnant spin, χf = 0.709 (+0.050/−0.061), is inferred from the f·τ product; this offers a population-level test of formation scenarios, though it depends on the same correction universality.
  • One could extend this pipeline to subdominant modes such as (3,3) or (2,1) to attempt genuine black-hole spectroscopy; the denoising approach might make those modes accessible.
  • The frequency systematic uncertainty (inflated by 62%) suggests that the pixel selection partially cancels the Gaussian noise reduction; a refined pixel-selection rule could recover the full gain.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The paper presents a new ringdown analysis pipeline for binary black hole remnants. The method uses cWB-reconstructed waveforms rather than raw strain, defines the ringdown window via the normalized cumulative energy at e(t)=0.82 (~3.5 t_Mf after peak), fits a two-mode damped-sinusoid model with BILBY, and applies parameter-dependent correction functions calibrated on SEOBNRv4HM injections. The analysis is applied to 16 GWTC-3 events with SNR>12 and M>50 M_sun. The reported combined fractional deviations from GR are δf220 = 0.003^{+0.028}_{-0.028} and δτ220 = 0.050^{+0.081}_{-0.086}, consistent with GR, and the paper claims this yields tighter QNM constraints than previous measurements.

Significance. If the method is sound, it represents a useful step toward precision ringdown tests: the cumulative-energy reference time is more robust than the usual peak-time reference, the cWB denoising is a practical way to mitigate detector noise, and the systematic-error accounting via injection-based coverage studies (Fig. 4) is a strength. However, the central precision claim rests on the assertion that the correction functions are universal. The calibration set is narrow (q=1, nonspinning, zero inclination, SNR 12-50), while the real events include remnant spins up to χf≈0.94 and overtone amplitudes up to ε≈0.94. The corrections at the chosen window are ~5% in both frequency and damping time, which is larger than the quoted combined frequency uncertainty (±2.8%). The paper's own text reports 4-6% variation over the remnant-mass range and dependence on aligned spins, so the claimed 'negligible' uncertainty of the universal correction is load-bearing and not validated by the coverage studies. The claim of tighter constraints than previous measurements is also not supported by any explicit numerical comparison.

major comments (3)
  1. [Secs. II B 3, III A; Eqs. (4)-(7), Fig. 3, Table I] The universal correction functions are calibrated exclusively on SEOBNRv4HM injections with q=1, nonspinning components, and zero inclination. Yet Table I includes events with remnant spin as high as χf≈0.94 and overtone amplitude ε≈0.94 (e.g., GW190521_074359). The correction at e=0.82 is approximately 5% in both frequency and damping time, larger than the quoted combined frequency uncertainty of ±2.8%. The paper itself states that the corrections vary by 4-6% over the mass range and depend on aligned spins producing rapidly rotating remnants. The coverage validation in Fig. 4 uses injections in the same q=1, nonspinning, face-on subspace and therefore cannot certify the extrapolation. A 2-3% miscalibration for a subset of events would shift δf220 and δτ220 by more than the quoted intervals. This is the central load-bearing assumption and must be supported by injection studies spanning
  2. [Abstract and Sec. V] The abstract and conclusion claim the method 'yields tighter constraints on the QNM frequency and damping time than previous measurements' and specifically refer to the GWTC-4 ringdown analysis [47]. No numerical comparison to [47] (or to other published ringdown constraints, e.g., the frequency-domain SEOBNRv5PHM results or direct QNM fits) is provided anywhere in the manuscript. Without such a comparison, the central precision claim is not demonstrated. The authors should tabulate the published 90% intervals and show quantitatively how their combined δf220 and δτ220 improve on them, including the impact of any differences in event selection and prior choices.
  3. [Eq. (11) and surrounding text] The fractional deviations δf220 and δτ220 are defined relative to f_IMR and τ_IMR, the values predicted by the IMR analysis. The IMR parameter estimation fits the full signal with GR waveforms that include a ringdown model; therefore the 'GR prediction' is not independent of the same post-merger data that the ringdown analysis uses. If a real deviation from GR were present in the ringdown, the IMR fit could partially absorb it into the inferred mass and spin, biasing δ toward zero. The manuscript does not quantify this effect or discuss how it affects the interpretation of the reported consistency with GR. This is particularly relevant because the claimed precision (e.g., ±2.8% in combined δf220) is smaller than the systematic effects being neglected.
minor comments (5)
  1. [Eq. (1)] The exponentials are missing the time variable: the model should read h(t) = A exp[i(2π f0 t + φ0)] + ε A exp[i(2π f1 t + φ1)], or an equivalent convention should be stated.
  2. [Eq. (2)] The quadrature H(t) is not defined in the text. Presumably it is the Hilbert transform of h(t), but this should be stated explicitly.
  3. [Sec. II B 2] The 'three cycles' duration of the ringdown window is not precisely defined: is it three cycles of the fundamental mode, the dominant instantaneous frequency, or the waveform envelope? Please specify.
  4. [Sec. III B] The statement that the systematic uncertainty in damping time is increased by only 5% while frequency is increased by 62% is clear, but the derivation of the total uncertainty (Table I caption) would benefit from an explicit formula linking σ_stat and σ_sys to the quoted 90% intervals.
  5. [General] There are minor typos and formatting issues, e.g., 'SEOBNRv4HM' is not consistently typeset, and the reference list contains a preprint number (2603.19021) that should be updated if published.

Circularity Check

0 steps flagged

No construction-level circularity; the analysis is calibrated on GR simulations and tested against IMR predictions, though the 'universal' correction is extrapolated beyond its calibration subspace.

full rationale

The paper's central chain is: cWB reconstructs a denoised waveform; a cumulative-energy reference defines an early ringdown window; a correction calibrated on SEOBNRv4HM injections accounts for merger/bias; a two-mode damped-sinusoid fit returns (f, τ); deviations from IMR-based GR predictions are formed via Eq. (11). The correction terms (Eqs. 4–7) are fitted to simulated signals with known GR answers, so applying them to real events is calibration, not a self-referential prediction. The comparison baseline in Eq. (11) uses f_IMR, τ_IMR from a separate IMR parameter estimation; although the IMR waveform includes ringdown information, the ringdown fit is an independent estimator, so δf220 and δτ220 are not identically zero by construction. No equation in the paper reduces the measured quantity to its input. The self-citations for cWB (refs [40–43]) cite an established search/reconstruction algorithm and are not load-bearing in a circular sense. The main weakness is an extrapolation: the universal correction is calibrated only on q=1, non-spinning, zero-inclination injections, while reported events have high remnant spins (χf up to 0.94) and unequal masses. This is a systematic-uncertainty concern, not a circularity. The 5% correction magnitude exceeds the quoted combined frequency uncertainty, but this affects accuracy, not logical circularity. Score 2 reflects minor self-citation and the partially data-derived GR baseline, with no identified construction-level circular step.

Axiom & Free-Parameter Ledger

5 free parameters · 6 axioms · 0 invented entities

The central claim rests on a chain of calibration steps fitted to simulated GR injections plus standard GR assumptions, rather than on a first-principles derivation. The free parameters are the empirical constants defining the ringdown window and the cWB reconstruction-bias corrections; the axioms are the GR mapping between remnant (M, χ) and QNM parameters, the assumption that the merger-ringdown transition is captured by SEOBNRv4HM, the assumption that cWB reconstruction errors are correctable, and the use of IMR-based parameters as an independent GR baseline. No new physical entities are introduced.

free parameters (5)
  • Cumulative-energy window threshold = 0.82 (modified by (1 − N/(3·SNR²))·0.82 at low SNR)
    Chosen by hand/calibration in Secs. II B 2-3 to balance merger contamination vs SNR; sets the window start at ~3.5 t_Mf after peak, the point where ~5% corrections are applied.
  • SNR-dependent frequency-bias correction = 1/(1 − e^{−SNR/5.9})
    Empirical fit to cWB reconstruction bias in Sec. III A; at SNR=12 the correction is ~15% in frequency — the largest single correction applied.
  • Window-offset vs remnant-mass relation = T_w − T_p ≈ 0.028·M_f + 0.94 (in t_Mf)
    Linear fit from injections (Eq. 7, Sec. III A) used to map the estimated M_f back to a cumulative-energy value for the corrections.
  • Systematic-error inflation factors = 62% (frequency), 5% (damping time) added in quadrature to statistical errors
    Chosen in Sec. III B so that injection coverage matches nominal (Fig. 4); the 62% combines a 27% jitter inflation with a 100%-of-statistical pixel-selection systematic.
  • Error-model scalings from injections = σ_e≈129/SNR, σ_p≈389/SNR, σ_sys,f≈87/SNR %, σ_sys,τ≈197/SNR %, σ_stat,f≈110/SNR %, σ_stat,τ≈584/SNR %
    Numerical coefficients (Eqs. 8-10) fitted to injection recovery, used to derive the jitter and systematic inflation factors.
axioms (6)
  • domain assumption Kerr QNM relations: f220 and τ220 are uniquely determined by remnant (M, a) via GR perturbation theory
    Used to build the two-mode model and to map measurements to Mf and χf (Sec. II B 1); standard GR result invoked via refs [32-35].
  • ad hoc to paper The linear QNM regime is already reached at e(t)=0.82 (~3.5 t_Mf after peak), with remaining merger contamination removable by GR-calibrated corrections
    The whole 'move closer to the merger' strategy rests on corrections fit to SEOBNRv4HM (Secs. II B 3, III); this is not established from first principles.
  • domain assumption cWB reconstruction faithfully recovers the ringdown signal, with TF pixel-selection biases that are small and empirically correctable
    Analysis is performed on reconstructed waveforms rather than raw strain; biases are handled by SNR-dependent corrections (Secs. II A, III A).
  • domain assumption Noise in O4 injection data is representative of the noise in the GWTC-3 events analyzed (O1/O2/O3a/O3b)
    Systematic errors and coverage are calibrated with injections into O4 data (Sec. III), then applied to events from earlier observing runs.
  • domain assumption IMR parameter estimates (f_IMR, τ_IMR, M, χ) are unbiased GR predictions suitable as the comparison baseline
    Used in Eq. (11) and in event selection (M>50 M⊙); for high-mass events the IMR fit is dominated by merger-ringdown data, making the baseline partially non-independent (Sec. IV).
  • ad hoc to paper The correction curves depend only weakly on source parameters beyond remnant mass, so a universal calibration applies
    Eqs. (6)-(7) assume universal SNR- and mass-dependent window corrections across the event population; the paper asserts rather than bounds the exceptions (aligned spins, unequal masses, waveform approximant).

pith-pipeline@v1.3.0-alltime-deepseek · 11475 in / 31594 out tokens · 326327 ms · 2026-08-02T05:41:27.063107+00:00 · methodology

0 comments
read the original abstract

The ringdown gravitational wave from a binary black hole (BBH) merger is a superposition of quasi-normal modes (QNMs) of the remnant black hole. In general relativity (GR), QNMs are damped harmonic oscillations with frequencies and damping times uniquely determined by the remnant's mass and spin. The measurement of the ringdown modes and performing black hole spectroscopy provides a tool to test the validity of GR. We present a ringdown analysis based on reconstruction of GW signals with coherent WaveBurst (cWB). This method yields tighter constraints on the QNM frequency and damping time than previous measurements. The improved precision results from the noise reduction achieved by the cWB reconstruction and the enhanced ringdown analysis, which probes the remnant properties at earlier times, closer to the merger. We have analyzed publicly available binary black hole (BBH) detections from the third Gravitational-Wave Transient Catalog (GWTC-3). For all events considered, the measured frequency and damping time of the dominant $(l,m)=(2,2)$ mode are found to be consistent with the predictions of GR. A combined analysis further strengthens these constraints, yielding fractional deviations in frequency $\delta f_{220} = 0.003_{-0.028}^{+0.028}$ and damping time $\delta\tau_{220} = 0.050_{-0.086}^{+0.081}$, consistent with zero within the quoted uncertainties.

Figures

Figures reproduced from arXiv: 2607.13275 by Achal Kumar, Marek J. Szczepa\'nczyk, Poulami Dutta Roy, Sergey Klimenko.

Figure 1
Figure 1. Figure 1: FIG. 1: TF representation of the first event GW150914 detected by LIGO: left panel - Livingston detector, middle [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2: The reconstructed time-domain signals (left [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3: Correction terms for the (2 [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4: Coverage of the marginalized posteriors for the [PITH_FULL_IMAGE:figures/full_fig_p005_4.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

53 extracted references · 43 linked inside Pith

  1. [1]

    Respectively, the Kerr model is used to relate mass and spin to the QNMs frequencies and damping time

    Ringdown Model Following the predictions of GR, the ringdown sig- nal is modelled as a sum of damped sinusoids, corre- sponding to different QNMs, parametrised by the rem- nant’s mass and spin. Respectively, the Kerr model is used to relate mass and spin to the QNMs frequencies and damping time. Since the subdominant QNMs are weak, the current analysis us...

  2. [2]

    As discussed in Section III A, however, this bias is not large and can be accounted for with the SNR-dependent correction

    The primary drawback is the introduction of a systematic bias in the reconstructed ringdown fre- quency associated with the incomplete recovery of the signal power, particularly for weak signal components with low SNR. As discussed in Section III A, however, this bias is not large and can be accounted for with the SNR-dependent correction. B. Fitting Procedure

  3. [3]

    Identifying this win- dow requires a reference time in the evolution of the GW signal

    Ringdown Window The ringdown window defines a segment of the post- merger data used for the ringdown analysis where the nonlinear merger effects are small. Identifying this win- dow requires a reference time in the evolution of the GW signal. One widely adopted method is to use the signal peak amplitude as the reference point. The window is selected a cer...

  4. [4]

    Typically, the ring- down analysis begins at 8–10t Mf [31, 47] after the signal peak timeT p to exclude the merger effects

    Moving close to the merger If the ringdown analysis begins close to the peak ampli- tude of the signal, the inferred parameters may be biased by residual merger effects, because QNMs describe only the linear perturbations of a black hole and the merger dynamics are inherently nonlinear. Typically, the ring- down analysis begins at 8–10t Mf [31, 47] after ...

    2023

  5. [5]

    B. P. Abbott et al. (LIGO Scientific, Virgo), Phys. Rev. Lett.116, 061102 (2016), 1602.03837

    Pith/arXiv arXiv 2016

  6. [6]

    B. P. Abbott et al. (LIGO Scientific, Virgo), Phys. Rev. Lett.119, 161101 (2017), 1710.05832. 8

    Pith/arXiv arXiv 2017

  7. [7]

    Abbott et al

    R. Abbott et al. (LIGO Scientific, KAGRA, VIRGO), Astrophys. J. Lett.915, L5 (2021), 2106.15163

    Pith/arXiv arXiv 2021

  8. [8]

    C. M. Will, Living Rev. Rel.17, 4 (2014), 1403.7377

    Pith/arXiv arXiv 2014

  9. [9]

    Berti et al., Class

    E. Berti et al., Class. Quant. Grav.32, 243001 (2015), 1501.07274

    Pith/arXiv arXiv 2015

  10. [10]

    P. C. C. Freire, N. Wex, G. Esposito-Farese, J. P. W. Ver- biest, M. Bailes, B. A. Jacoby, M. Kramer, I. H. Stairs, J. Antoniadis, and G. H. Janssen, Mon. Not. Roy. Astron. Soc.423, 3328 (2012), 1205.1450

    Pith/arXiv arXiv 2012

  11. [11]

    Abuter et al

    R. Abuter et al. (GRA VITY), Astron. Astrophys.615, L15 (2018), 1807.09409

    Pith/arXiv arXiv 2018

  12. [12]

    Do et al., Science365, 664 (2019), 1907.10731

    T. Do et al., Science365, 664 (2019), 1907.10731

    arXiv 2019

  13. [13]

    Akiyama et al

    K. Akiyama et al. (Event Horizon Telescope), Astrophys. J. Lett.875, L1 (2019), 1906.11238

    Pith/arXiv arXiv 2019

  14. [14]

    Clifton, P

    T. Clifton, P. G. Ferreira, A. Padilla, and C. Skordis, Phys. Rept.513, 1 (2012), 1106.2476

    Pith/arXiv arXiv 2012

  15. [15]

    Dafermos, G

    M. Dafermos, G. Holzegel, and I. Rodnianski, Acta Mat. 222, 1 (2019), 1601.06467

    Pith/arXiv arXiv 2019

  16. [16]

    Teixeira da Costa, Commun

    R. Teixeira da Costa, Commun. Math. Phys.378, 705 (2020), 1910.02854

    Pith/arXiv arXiv 2020

  17. [17]

    Klainerman and J

    S. Klainerman and J. Szeftel, Pure Appl. Math. Quart. 19, 791 (2023), 2104.11857

    Pith/arXiv arXiv 2023

  18. [18]

    Dafermos, G

    M. Dafermos, G. Holzegel, I. Rodnianski, and M. Taylor (2021), 2104.08222

    Pith/arXiv arXiv 2021

  19. [19]

    Penrose, Phys

    R. Penrose, Phys. Rev. Lett.14, 57 (1965)

    1965

  20. [20]

    S. W. Hawking and R. Penrose, Proc. Roy. Soc. Lond. A 314, 529 (1970)

    1970

  21. [21]

    S. W. Hawking, Phys. Rev. D14, 2460 (1976)

    1976

  22. [22]

    Almheiri, T

    A. Almheiri, T. Hartman, J. Maldacena, E. Shaghou- lian, and A. Tajdini, Rev. Mod. Phys.93, 035002 (2021), 2006.06872

    Pith/arXiv arXiv 2021

  23. [23]

    Raju, Phys

    S. Raju, Phys. Rept.943, 1 (2022), 2012.05770

    Pith/arXiv arXiv 2022

  24. [24]

    Abbott et al

    R. Abbott et al. (KAGRA, VIRGO, LIGO Scientific), Phys. Rev. X13, 041039 (2023), 2111.03606

    Pith/arXiv arXiv 2023

  25. [25]

    A. G. Abac et al. (LIGO Scientific, KAGRA, VIRGO), Astrophys. J. Lett.995, L18 (2025), 2508.18080

    Pith/arXiv arXiv 2025

  26. [26]

    Cutler and E

    C. Cutler and E. E. Flanagan, Phys. Rev. D49, 2658 (1994), gr-qc/9402014

    Pith/arXiv arXiv 1994

  27. [27]

    B. P. Abbott et al. (LIGO Scientific, Virgo), Class. Quant. Grav.37, 055002 (2020), 1908.11170

    Pith/arXiv arXiv 2020

  28. [28]

    Veitch et al., Phys

    J. Veitch et al., Phys. Rev. D91, 042003 (2015), 1409.7215

    Pith/arXiv arXiv 2015

  29. [29]

    Christensen and R

    N. Christensen and R. Meyer, Rev. Mod. Phys.94, 025001 (2022), 2204.04449

    Pith/arXiv arXiv 2022

  30. [30]

    A. G. Abac et al. (LIGO Scientific, VIRGO, KAGRA) (2025), 2508.18081

    Pith/arXiv arXiv 2025

  31. [31]

    Abbott et al

    R. Abbott et al. (KAGRA, VIRGO, LIGO Scientific), Phys. Rev. X13, 011048 (2023), 2111.03634

    Pith/arXiv arXiv 2023

  32. [32]

    A. G. Abac et al. (LIGO Scientific, VIRGO, KAGRA) (2025), 2508.18083

    Pith/arXiv arXiv 2025

  33. [33]

    A. G. Abac et al. (LIGO Scientific, VIRGO, KAGRA) (2025), 2509.04348

    Pith/arXiv arXiv 2025

  34. [34]

    A. G. Abac et al. (LIGO Scientific, Virgo, KAGRA), Phys. Rev. Lett.135, 111403 (2025), 2509.08054

    Pith/arXiv arXiv 2025

  35. [35]

    Abbott et al

    R. Abbott et al. (LIGO Scientific, VIRGO, KAGRA), Phys. Rev. D112, 084080 (2025), 2112.06861

    Pith/arXiv arXiv 2025

  36. [36]

    Abedi et al

    J. Abedi et al. (2025), 2505.23895

    Pith/arXiv arXiv 2025

  37. [37]

    Chandrasekhar and S

    S. Chandrasekhar and S. L. Detweiler, Proc. Roy. Soc. Lond. A344, 441 (1975)

    1975

  38. [38]

    C. V. Vishveshwara, Nature227, 936 (1970)

    1970

  39. [39]

    Berti, V

    E. Berti, V. Cardoso, and A. O. Starinets, Class. Quant. Grav.26, 163001 (2009), 0905.2975

    Pith/arXiv arXiv 2009

  40. [40]

    Brito, A

    R. Brito, A. Buonanno, and V. Raymond, Phys. Rev. D 98, 084038 (2018), 1805.00293

    Pith/arXiv arXiv 2018

  41. [41]

    Ghosh, R

    A. Ghosh, R. Brito, and A. Buonanno, Phys. Rev. D103, 124041 (2021), 2104.01906

    Pith/arXiv arXiv 2021

  42. [42]

    Maggio, in57th Rencontres de Moriond on Gravitation (2023), 2305.15439

    E. Maggio, in57th Rencontres de Moriond on Gravitation (2023), 2305.15439

    Pith/arXiv arXiv 2023

  43. [43]

    Carullo, W

    G. Carullo, W. Del Pozzo, and J. Veitch, Phys. Rev. D 99, 123029 (2019), [Erratum: Phys.Rev.D 100, 089903 (2019)], 1902.07527

    Pith/arXiv arXiv 2019

  44. [44]

    Mishra, S

    T. Mishra, S. Bhaumik, V. Gayathri, M. J. Szczepa´ nczyk, I. Bartos, and S. Klimenko, Phys. Rev. D111, 023054 (2025), 2410.15191

    Pith/arXiv arXiv 2025

  45. [45]

    Klimenko, S

    S. Klimenko, S. Mohanty, M. Rakhmanov, and G. Mitsel- makher, Phys. Rev. D72, 122002 (2005), gr-qc/0508068

    Pith/arXiv arXiv 2005

  46. [46]

    Klimenko, I

    S. Klimenko, I. Yakushin, A. Mercer, and G. Mit- selmakher, Class. Quant. Grav.25, 114029 (2008), 0802.3232

    Pith/arXiv arXiv 2008

  47. [47]

    Klimenko et al., Phys

    S. Klimenko et al., Phys. Rev. D93, 042004 (2016), 1511.05999

    Pith/arXiv arXiv 2016

  48. [48]

    Klimenko (2022), 2201.01096

    S. Klimenko (2022), 2201.01096

    Pith/arXiv arXiv 2022

  49. [49]

    Ashton et al., Astrophys

    G. Ashton et al., Astrophys. J. Suppl.241, 27 (2019), 1811.02042

    Pith/arXiv arXiv 2019

  50. [50]

    I. M. Romero-Shaw et al., Mon. Not. Roy. Astron. Soc. 499, 3295 (2020), 2006.00714

    Pith/arXiv arXiv 2020

  51. [51]

    A. G. Abac et al. (LIGO Scientific, VIRGO, KAGRA) (2026), 2603.19021

    Pith/arXiv arXiv 2026

  52. [52]

    Ramos-Buades, A

    A. Ramos-Buades, A. Buonanno, M. Khalil, and S. Os- sokine, Phys. Rev. D105, 044035 (2022), 2112.06952

    Pith/arXiv arXiv 2022

  53. [53]

    Cotesta, S

    R. Cotesta, S. Marsat, and M. P¨ urrer, Phys. Rev. D101, 124040 (2020), 2003.12079

    Pith/arXiv arXiv 2020