Pith. sign in

REVIEW 3 major objections 3 minor 5 cited by

A neural posterior estimator recovers microlensed gravitational-wave parameters in hours instead of days, matching full Bayesian sampling on simulated injections.

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

2026-08-03 22:51 UTC pith:BN5KC4S5

load-bearing objection A useful, honestly-flagged proof-of-concept for amortized wave-optics lensed-GW inference, but the headline speedup and catalog-scale claims outrun the three-injection validation. the 3 major comments →

arxiv 2511.08486 v2 pith:BN5KC4S5 submitted 2025-11-11 astro-ph.CO gr-qc

Accelerated inference of microlensed gravitational waves with machine learning

Marienza Caldarola , Srashti Goyal , Nihar Gupte , Stephen R. Green , Miguel Zumalac\'arregui This is my paper
classification astro-ph.CO gr-qc
keywords gravitational wavesmicrolensingwave opticsparameter estimationsimulation-based inferencenormalizing flowsimportance samplingBayes factors
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.

This paper claims that a machine-learning posterior estimator can make microlensed gravitational-wave inference cheap enough to run on every event in a catalog. The authors train a normalizing-flow network—DINGO, extended to a 17-parameter space—on binary-black-hole waveforms modulated by the wave-optics diffraction factor of an isolated point-mass lens, and show on simulated injections that its posteriors and Bayes factors agree with conventional nested-sampling inference while cutting per-event runtime from about five days to about two hours (or minutes without importance sampling). A 1000-injection probability-probability test shows the network is well calibrated, except for the coalescence time, which occasionally goes bimodal because lensing resembles the interference of two time-delayed images. If right, the method makes background Bayes-factor distributions affordable to compute and turns microlensing searches into a routine screening step for the coming high-rate observing runs.

Core claim

The paper's central claim is that a single trained neural posterior estimator can perform full Bayesian inference on microlensed gravitational-wave signals, including the two lens parameters (redshifted lens mass and impact parameter) that conventional analysis struggles with because of added dimensionality and expensive diffraction computations. On simulated injections, the network's posteriors and log evidences match those from a standard nested-sampling code once importance sampling is applied, with the per-event cost dropping from about five days to about two hours—and to minutes if the raw network output is used without reweighting. The supporting evidence is a 1000-injection probabilit

What carries the argument

The load-bearing object is the frequency-dependent amplification factor F(f) = h_lensed(f)/h_unlensed(f), computed in the wave-optics regime from the diffraction integral; for an isolated point mass it depends on only two parameters—the redshifted lens mass and the impact parameter—and it modulates the base binary-black-hole waveform during training. Around this sits a normalizing-flow neural posterior estimator (DINGO) with group-equivariant time alignment, trained on 10^7 lensed waveforms, whose samples are reweighted by importance sampling so that explicit likelihood evaluations correct the proposal and yield unbiased posteriors and evidences. The 200-term singular-value compression both

Load-bearing premise

The load-bearing premise is that both the trained network and the test waveforms use the same 200-term waveform compression, although lensed signals need roughly twice as many terms to be represented faithfully; if that compression error becomes comparable to the lensing signature at lower signal-to-noise ratio or stronger lensing, the claimed match to the full-waveform analysis could fail.

What would settle it

Retrain the network with 400 compression terms and rerun the probability-probability calibration plus the three injection comparisons, including impact parameters below 0.2 and lens masses up to 10^4 solar masses at signal-to-noise ratios near 15; if the importance-sampled log Bayes factors disagree with full-waveform nested sampling by more than the combined sampling error, the claim that the current network captures the lensed-signal information is falsified.

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

If this is right

  • Posterior samples for a lensed event can be drawn in minutes, and importance-sampling correction takes about two hours per event instead of days, making catalog-scale lensing searches practical.
  • The lensed network applied to unlensed injections recovers evidence matching the unlensed analysis, making Monte Carlo background Bayes-factor distributions—needed to claim significant lensed events—computable at scale for the first time.
  • The marginal posterior at the impact-parameter prior boundary behaves as a rapid lensed-event indicator, so candidates can be screened before expensive sampling.
  • A collapse in importance-sampling efficiency flags out-of-distribution data—a strongly lensed signal analyzed as unlensed, or a glitch—serving as a built-in diagnostic.
  • The pipeline extends to more complex lens models and waveform families, since the lensing transform is applied on the fly during training.

Where Pith is reading between the lines

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

  • A natural extension the authors leave implicit is that the amortized network makes population-level lensing studies feasible: the marginal cost per additional event is nearly zero, so lensed-event candidates could be screened across all O(10^5) detections of next-generation runs rather than only the few dozen that receive detailed follow-up.
  • The 200-component waveform compression is the weakest technical point; retraining at 400 components—the level the paper's own mismatch curve indicates lensed waveforms require—and re-running the injection tests would directly measure whether the claimed accuracy is an artifact of shared compression between training and test data.
  • The bimodal coalescence-time posterior suggests a concrete improvement: apply the time-alignment idea to the second, time-delayed image as well, which could sharpen lens-mass estimation and raise sampling efficiency.
  • The sampling-efficiency drop could be turned into a calibrated lensed-event classifier by evaluating epsilon or effective sample size on a large injection population and choosing a threshold; the paper explicitly leaves such classification benchmarking to future work.

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 / 3 minor

Summary. The paper presents a proof-of-concept extension of the DINGO neural posterior estimator to gravitational-wave microlensing by an isolated point mass in the wave-optics regime. The authors train a 17-parameter normalizing-flow model on 10^7 IMRPhenomXPHM signals with GLoW amplification factors, using n=200 SVD components for waveform compression. They validate with a 1000-injection p-p test, compare three injections against Bilby with nested sampling (Table II), and report an approximately order-of-magnitude speedup when DINGO is combined with importance sampling. They also propose using the unlensed network's sample-efficiency drop as an out-of-distribution diagnostic and claim efficient estimation of background Bayes-factor distributions.

Significance. If the claims hold, the paper provides a fast amortized inference tool for lensed GWs in the upcoming O5/3G catalog era, building on public code (the DINGO lensing branch and GLoW). The 1000-injection p-p calibration and the use of importance-sampling-corrected posterior and evidence estimates are appropriate methodological choices, and the three Bilby cross-checks are a useful first step. The main value is the demonstration that a 17-parameter NPE can accommodate two extra lensing parameters and that IS can produce evidence estimates. However, the current support for the central 'matching Bilby' claim is based on only three injections and shows systematic logZ offsets, so the significance is conditional on additional validation.

major comments (3)
  1. [Sec. V, Fig. 8] The training and p-p calibration use waveforms compressed with n=200 SVD components, yet Fig. 8 shows that lensed waveforms require about 400 components to reach 10^-4 mismatch. Thus Fig. 4 validates only self-consistency on the truncated waveform family, not fidelity to the full lensed waveforms used by Bilby. The claim that lensed-DINGO posteriors and evidence match Bilby rests on three injections (Table II) at SNR 18, 32, and 35. If the truncation error is comparable to the lensing features at lower SNR or for stronger lensing (y<0.2, high M_Lz), the agreement could degrade. Please quantify the truncation-induced bias, for example by comparing the SVD mismatch against the statistical uncertainty across the prior, or by running Bilby on the truncated waveforms for a broader set of injections.
  2. [Table II, Sec. VI] All DINGO logZ values are systematically lower than Bilby by ΔlogZ ≈ 0.8–1.8 even in the high-n_eff cases (e.g., unlensed injection: 497.9 vs 499.7; lensed y=1.2: 133.9 vs 134.7), and the y=0.2 foreground Bayes factor is off by ΔlogB ≈ 14 (124.9 vs 110.9). The text states that 'the evidence logZ matches with the Bilby', but a systematic offset of 1–2 in logZ is comparable to commonly used significance thresholds and could change screening conclusions. Please report uncertainties on the DINGO evidence estimates and show that the offsets do not affect the scientific conclusions, or temper the matching claim.
  3. [Table II / Fig. 9] The paper states that reliable results require n_eff > 1000 (Sec. VI), yet the lensed y=0.2 DINGO (lensed) result has n_eff = 602 and is nevertheless presented as matching Bilby (Fig. 9: logZ 592.3 vs 593.8). Either this case should be treated as unreliable per the stated criterion, or the criterion should be revisited. As written, the evidence for the strong-lensing regime relies on a case that fails the paper's own reliability threshold.
minor comments (3)
  1. [References] Ref. [85] is incomplete: it lacks a title, journal, and arXiv identifier.
  2. [Table II] The capitalization 'BILBY' in the header is inconsistent with the rest of the text; use 'Bilby'.
  3. [Fig. 6 caption] The caption contains a typo: 'DINGO-IS S (blue)' should be 'DINGO-IS (blue)'.

Circularity Check

0 steps flagged

No significant circularity: the central DINGO-vs-Bilby comparison is an external benchmark, not a fit to its own outputs.

full rationale

I walked the claimed derivation chain: lensed waveforms are produced with GLoW from the wave-optics point-lens amplification factor (Eqs. 1-3); DINGO's neural posterior estimator is trained on simulated injections drawn from the stated priors (Sec. V); and the resulting posteriors and log-evidence values are compared with Bilby under the exact Gaussian likelihood (Eq. 7, Table II, Figs. 5-7, 9-10). The central validation is against Bilby, an independent nested-sampling code using the exact likelihood, and the network is not fitted to Bilby's outputs. The p-p calibration in Fig. 4 validates internal calibration rather than fidelity to full-waveform lensing, but the paper does not present it as the primary external check; the Bilby comparison is. The SVD truncation concern (Fig. 8 shows n=200 used while ~400 components are needed for 10^-4 mismatch) is a real, honestly stated systematic-error risk, not a circular reduction: the comparison target remains external to the training procedure. Self-references to DINGO and GLoW are tool attributions, not load-bearing arguments; DINGO has been reviewed by the LVK collaboration and GLoW is public and benchmarked, and the unresolved self-references [43,44] are contextual substantiation for existing candidates, not inputs to the inference claim. No equation or fitted parameter reduces by construction to another equation or to a self-citation. Circularity score: 1, reflecting only the expected use of the authors' own previously validated tools.

Axiom & Free-Parameter Ledger

3 free parameters · 7 axioms · 0 invented entities

The network result rests on standard scalar-diffraction lensing theory, the analytically known point-mass amplification factor, the assumed signal and noise models, and the consistency conditions of simulation-based inference and importance sampling. The paper's own contributions sit on top of these; the hand-chosen SVD truncation and lens prior bounds are the choices that most directly shape the demonstrated performance.

free parameters (3)
  • SVD basis size n=200 = 200
    Chosen 'for computational simplicity' (Sec. V). Fig. 8 shows lensed waveforms need ~400 components for 10^-4 mismatch, so the training family is a lossy reconstruction; this bounds the fidelity of everything the network learns.
  • Lens prior ranges = log10(MLz) in [1,4] M_sun; y in [0.1,5]
    Hand-chosen training priors (Table I). The lensing-detection indicator p(y=5|d) relies on y hitting the prior boundary; lens masses above 10^4 M_sun (GO regime, per Ref [85]) are excluded by design.
  • Reliability threshold n_eff > 1000 = 1000
    Hand-set criterion in Sec. VI for trusting DINGO Bayes factors; the strong-lensing test case (y=0.2) itself has n_eff=602, below the threshold, yet its evidence is presented as matching Bilby.
axioms (7)
  • standard math The diffraction integral of Eq. (2) is the correct model for wave-optics lensing of GW strain.
    Section II; standard scalar-diffraction result (Schneider, Ehlers and Falco [9]).
  • standard math The point-mass lens amplification factor F(w) (hypergeometric series, GLoW [40]) is exact for an isolated point lens and is the correct physical model for the microlensing signals studied.
    Section II; analytically known result; GLoW additionally validated in [40].
  • domain assumption IMRPhenomXPHM accurately represents the unlensed BBH signal.
    Section V: all training and test waveforms use this approximant. Waveform-model error is not propagated into the claimed speedup/accuracy.
  • domain assumption Detector data are stationary Gaussian noise with fixed O3 HLV PSDs.
    Section V and all three test injections. Real noise non-stationarity and glitches are outside the demonstration; the out-of-distribution diagnostic is argued by analogy ('such as lensed signals or instrumental glitches'), not tested on glitches.
  • standard math NPE trained by minimizing the KL loss in Eqs. (9)/(10) converges to the true posterior given sufficiently expressive flows and i.i.d. draws from p(theta)p(d|theta).
    Section IV; established SBI consistency result (cites [60-63]) on which the entire inference claim rests.
  • standard math Importance-sampling weights w_i from Eq. (11) give unbiased evidence via normalization Eq. (14).
    Section IV; textbook IS, valid as long as the proposal support covers the posterior; n_eff gauges this.
  • ad hoc to paper The 200-component SVD basis preserves the inference-relevant content of lensed waveforms.
    Section V/Fig. 8: lensed waveforms need ~400 components for 10^-4 mismatch; 200 is adopted 'for computational simplicity'. If the truncation discards lensing information, the lensed posteriors are biased toward the simplified family.

pith-pipeline@v1.3.0-alltime-deepseek · 27 in / 23276 out tokens · 229175 ms · 2026-08-03T22:51:56.373676+00:00 · methodology

0 comments
read the original abstract

Gravitational waves (GWs) within the LIGO-Virgo-KAGRA sensitivity band can be microlensed by stellar and intermediate-mass black holes, producing a frequency-dependent modulation of the signal amplitude. Microlensing analyses, however, are costly due to the increased dimensionality of the parameter space and waveform computation time. As a proof of concept, we show that the deep-learning-based framework Deep Inference for Gravitational-Wave Observations (DINGO), which employs a simulation-based inference approach to estimate posterior distributions, can perform efficient parameter inference for GW microlensing by an isolated point-mass lens. Using simulated microlensed GW signals, we train a lensed-DINGO network and compare its performance with traditional Bayesian parameter estimation carried out with Bilby. Our framework can be used to rapidly identify microlensed events in large GW catalogs. When the lensed-DINGO network is combined with importance sampling, we find that although sample efficiencies are somewhat reduced compared to the unlensed-DINGO network, owing to the richer structure of microlensed signals, it still achieves $\mathcal{O}(10\times)$ speed-up relative to Bilby. We further show that this framework is useful to efficiently estimate the background Bayes-factor distribution, which is crucial for assessing the significance of candidate lensed events. However, for foreground (lensed) events, the sampling efficiency can sometimes drop when analysed with the unlensed-DINGO network, providing a diagnostic indicator of out-of-distribution data. Our approach can be straightforwardly generalised to more complex and realistic lens models, enabling detailed studies of microlensed GWs.

Figures

Figures reproduced from arXiv: 2511.08486 by Marienza Caldarola, Miguel Zumalac\'arregui, Nihar Gupte, Srashti Goyal, Stephen R. Green.

Figure 1
Figure 1. Figure 1: FIG. 1. Lensing Amplification factor [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2. Frequency domain GW strain amplitude of a [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3. Evolution of the training and validation loss for [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4 [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5. A lensed injection with [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: FIG. 6. An unlensed injection with DINGO (lensed). Pos [PITH_FULL_IMAGE:figures/full_fig_p009_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: FIG. 7. A lensed injection with [PITH_FULL_IMAGE:figures/full_fig_p009_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: FIG. 8. Mismatch between SVD basis reconstructed wave [PITH_FULL_IMAGE:figures/full_fig_p010_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: FIG. 9. A lensed injection with [PITH_FULL_IMAGE:figures/full_fig_p012_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: FIG. 10. An unlensed injection with DINGO (unlensed). Posterior distributions comparing DINGO (orange), DINGO-IS [PITH_FULL_IMAGE:figures/full_fig_p013_10.png] view at source ↗

discussion (0)

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

Forward citations

Cited by 5 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Effective description of lensed gravitational waves diffracted by stellar fields

    astro-ph.HE 2026-06 unverdicted novelty 7.0

    Authors synthesize SVD-based reduced-order models from wave-optics simulations to provide an effective stochastic description of stellar microlensing distortions on lensed gravitational waves.

  2. labrador: A domain-optimized machine-learning tool for gravitational wave inference

    gr-qc 2026-04 unverdicted novelty 7.0

    Labrador is a domain-optimized neural posterior estimation tool achieving 1% median importance-sampling efficiency and first extensive coverage of long-duration low-mass gravitational wave signals through equivariance...

  3. The diffraction-lensing interpretation of GW231123 with astrophysical priors

    astro-ph.CO 2026-07 conditional novelty 6.0

    GW231123 is better fit by a lower-mass merger diffracted by an isolated ~1000 M_sun lens, but astrophysical priors and a frequentist rate estimate make this lensing interpretation unlikely.

  4. Robust parameter inference for Taiji via time-frequency contrastive learning and normalizing flows

    gr-qc 2026-04 unverdicted novelty 6.0

    A glitch-robust amortized inference framework combining normalizing flows, time-frequency multimodal fusion, and contrastive learning outperforms MCMC for Taiji massive black hole binary parameter estimation under noi...

  5. labrador: A domain-optimized machine-learning tool for gravitational wave inference

    gr-qc 2026-04 unverdicted novelty 6.0

    Domain-optimized neural posterior estimation (labrador) reaches ~1% median importance-sampling efficiency on aligned-spin GW signals and covers long-duration systems with m2 < 10 M☉.

Reference graph

Works this paper leans on

88 extracted references · 1 canonical work pages · cited by 4 Pith papers

  1. [1]

    B. P. Abbottet al.(Virgo, LIGO Scientific), Phys. Rev. Lett.116, 061102 (2016), arXiv:1602.03837 [gr-qc]

    Pith/arXiv arXiv 2016

  2. [2]

    B. A. et al. (LIGO Scientific Collaboration and V. Col- laboration), Physical Review Letters116, 221101 (2016)

    2016

  3. [3]

    out-of-distribution

    anunlensed injectionwith SNR 32. The waveforms in the H1 detector for these injections is shown in Fig. 2. For all the injections we do parameter inference using DINGO (lensed), DINGO (unlensed) and with Bilby under lensed and unlensed hypothesis (H L andH U ). For DINGO we generate 10 6 samples and then perform IS to get improved results, while computing...

  4. [4]

    Yagi and N

    K. Yagi and N. Yunes, Physical Review D95, 084014 (2017)

    2017

  5. [5]

    B. A. et al. (LIGO Scientific Collaboration and V. Col- laboration), Physical Review Letters123, 111102 (2019)

    2019

  6. [6]

    Abbott et al

    R. Abbott et al. (LIGO Scientific Collaboration, Virgo Collaboration, and KAGRA Collaboration), Physical Re- view X13, 011048 (2023)

    2023

  7. [7]

    Abbott et al

    R. Abbott et al. (LIGO Scientific Collaboration, Virgo Collaboration, and KAGRA Collaboration), Physical Re- view D112, 084080 (2025)

    2025

  8. [8]

    Abbott et al

    R. Abbott et al. (LIGO Scientific Collaboration, Virgo Collaboration, and KAGRA Collaboration), The Astro- physical Journal949(2023), 10.3847/1538-4357/acd4e1

    work page doi:10.3847/1538-4357/acd4e1 2023

  9. [9]

    B. A. et al. (LIGO Scientific Collaboration and V. Col- laboration), Nature551, 85 (2017)

    2017

  10. [10]

    Clowe, M

    D. Clowe, M. Bradac, A. H. Gonzalez, M. Markevitch, S. W. Randall, C. Jones, and D. Zaritsky, Astrophys. J. Lett.648, L109 (2006), arXiv:astro-ph/0608407

    Pith/arXiv arXiv 2006

  11. [11]

    Schneider, J

    P. Schneider, J. Ehlers, and E. E. Falco,Gravitational Lenses, Astronomy and astrophysics library (Springer, Berlin, Germany, 1999)

    1999

  12. [12]

    Vegettiet al., Space Sci

    S. Vegettiet al., Space Sci. Rev.220, 58 (2024), arXiv:2306.11781 [astro-ph.CO]

    Pith/arXiv arXiv 2024

  13. [13]

    Aghanimet al.(Planck), Astron

    N. Aghanimet al.(Planck), Astron. Astrophys.641, A8 (2020), arXiv:1807.06210 [astro-ph.CO]

    Pith/arXiv arXiv 2020

  14. [14]

    Treu, Ann

    T. Treu, Ann. Rev. Astron. Astrophys.48, 87 (2010), arXiv:1003.5567 [astro-ph.CO]

    Pith/arXiv arXiv 2010

  15. [15]

    Bartelmann and P

    M. Bartelmann and P. Schneider, Phys. Rept.340, 291 (2001), arXiv:astro-ph/9912508

    Pith/arXiv arXiv 2001

  16. [16]

    Reyes, R

    R. Reyes, R. Mandelbaum, U. Seljak, T. Baldauf, J. E. Gunn, L. Lombriser, and R. E. Smith, Nature464, 256 (2010), arXiv:1003.2185 [astro-ph.CO]

    Pith/arXiv arXiv 2010

  17. [17]

    Natarajan, L

    P. Natarajan, L. L. Williams, M. Bradac, C. Grillo, A. Ghosh, K. Sharon, and J. Wagner, (2024), arXiv:2403.06245 [astro-ph.CO]

    arXiv 2024

  18. [18]

    Goyal, A

    S. Goyal, A. Vijaykumar, J. M. Ezquiaga, and M. Zumalacarregui, Phys. Rev. D108, 024052 (2023), arXiv:2301.04826 [gr-qc]

    Pith/arXiv arXiv 2023

  19. [19]

    T. E. Collett, L. J. Oldham, R. J. Smith, M. W. Auger, K. B. Westfall, D. Bacon, R. C. Nichol, K. L. Masters, K. Koyama, and R. van den Bosch, Science360, 1342 (2018), arXiv:1806.08300 [astro-ph.CO]

    Pith/arXiv arXiv 2018

  20. [20]

    K. C. Wonget al.(H0LiCOW), Mon. Not. Roy. Astron. Soc.498, 1420 (2020), arXiv:1907.04869 [astro-ph.CO]

    Pith/arXiv arXiv 2020

  21. [21]

    Goyal, K

    S. Goyal, K. Haris, A. K. Mehta, and P. Ajith, Phys. Rev. D103, 024038 (2021), arXiv:2008.07060 [gr-qc]

    Pith/arXiv arXiv 2021

  22. [22]

    S. Jana, S. J. Kapadia, T. Venumadhav, and P. Ajith, Phys. Rev. Lett.130, 261401 (2023), arXiv:2211.12212 [astro-ph.CO]

    Pith/arXiv arXiv 2023

  23. [23]

    T. Treu, S. H. Suyu, and P. J. Marshall, Astron. Astro- phys. Rev.30, 8 (2022), arXiv:2210.15794 [astro-ph.CO]

    Pith/arXiv arXiv 2022

  24. [24]

    Takahashi and T

    R. Takahashi and T. Nakamura, Astrophys. J.595, 1039 (2003), arXiv:astro-ph/0305055

    Pith/arXiv arXiv 2003

  25. [25]

    G. P. Smithet al., Phil. Trans. Roy. Soc. Lond. A383, 20240134 (2025), arXiv:2503.19973 [astro-ph.HE]

    Pith/arXiv arXiv 2025

  26. [26]

    J. M. Diego, O. A. Hannuksela, P. L. Kelly, T. Broad- hurst, K. Kim, T. G. F. Li, G. F. Smoot, and G. Pagano, Astron. Astrophys.627, A130 (2019), arXiv:1903.04513 [astro-ph.CO]

    Pith/arXiv arXiv 2019

  27. [27]

    Leung, D

    C. Leung, D. Jow, P. Saha, L. Dai, M. Oguri, and L. V. E. Koopmans, Space Sci. Rev.221, 29 (2025), arXiv:2304.01202 [astro-ph.HE]

    Pith/arXiv arXiv 2025

  28. [28]

    S. M. C. Yeung, M. H. Y. Cheung, E. Seo, J. A. J. Gais, O. A. Hannuksela, and T. G. F. Li, Mon. Not. Roy. Astron. Soc.526, 2230 (2023), arXiv:2112.07635 [gr-qc]

    Pith/arXiv arXiv 2023

  29. [29]

    M. H. Y. Cheung, J. Gais, O. A. Hannuksela, and T. G. F. Li, Mon. Not. Roy. Astron. Soc.503, 3326 (2021), arXiv:2012.07800 [astro-ph.HE]

    Pith/arXiv arXiv 2021

  30. [30]

    Oguri and R

    M. Oguri and R. Takahashi, Phys. Rev. D106, 043532 (2022), arXiv:2204.00814 [astro-ph.CO]

    Pith/arXiv arXiv 2022

  31. [31]

    Mishra, A

    A. Mishra, A. K. Meena, A. More, S. Bose, and J. S. Bagla, Mon. Not. Roy. Astron. Soc.508, 4869 (2021), arXiv:2102.03946 [astro-ph.CO]

    Pith/arXiv arXiv 2021

  32. [32]

    X. Shan, G. Li, X. Chen, W. Zhao, B. Hu, and S. Mao, Sci. China Phys. Mech. Astron.68, 219512 (2025), arXiv:2409.06747 [astro-ph.IM]

    Pith/arXiv arXiv 2025

  33. [33]

    Mishra, A

    A. Mishra, A. K. Meena, A. More, and S. Bose, Mon. Not. Roy. Astron. Soc.531, 764 (2024), arXiv:2306.11479 [astro-ph.CO]

    Pith/arXiv arXiv 2024

  34. [34]

    E. Seo, X. Shan, J. Janquart, O. A. Hannuksela, M. A. Hendry, and B. Hu, Astrophys. J.988, 159 (2025), arXiv:2503.02186 [gr-qc]

    Pith/arXiv arXiv 2025

  35. [35]

    R. K. L. Lo, L. Vujeva, J. M. Ezquiaga, and J. C. L. Chan, Phys. Rev. Lett.134, 151401 (2025), arXiv:2407.17547 [gr-qc]

    Pith/arXiv arXiv 2025

  36. [36]

    Feldbrugge, U.-L

    J. Feldbrugge, U.-L. Pen, and N. Turok, Annals Phys. 451, 169255 (2023), arXiv:1909.04632 [astro-ph.HE]

    Pith/arXiv arXiv 2023

  37. [37]

    Z. Su, X. Shan, Z. Lyu, J. Zhang, Y. Liu, S. Mao, and H. Yang, (2025), arXiv:2510.17125 [gr-qc]

    arXiv 2025

  38. [38]

    X. Shan, G. Li, X. Chen, W. Zheng, and W. Zhao, Sci. China Phys. Mech. Astron.66, 239511 (2023), arXiv:2208.13566 [astro-ph.CO]

    Pith/arXiv arXiv 2023

  39. [39]

    Tambalo, M

    G. Tambalo, M. Zumalac´ arregui, L. Dai, and M. H.-Y. Cheung, Phys. Rev. D108, 043527 (2023), arXiv:2210.05658 [gr-qc]

    Pith/arXiv arXiv 2023

  40. [40]

    Villarrubia-Rojo, S

    H. Villarrubia-Rojo, S. Savastano, M. Zumalac´ arregui, L. Choi, S. Goyal, L. Dai, and G. Tambalo, Phys. Rev. D111, 103539 (2025), arXiv:2409.04606 [gr-qc]. 12 0.6 0.7 0.8 0.9 q 1200 1600 2000 2400dL 1.11 1.12 1.13 1.14 δ 1.350 1.365 1.380 1.395 1.410α 0.25 0.50 0.75 1.00 ι 0.8 1.6 2.4 ψ 1.5 3.0 4.5 6.0 φ 0.2 0.4 0.6 0.8 χ1 0.2 0.4 0.6 0.8 χ2 0.8 1.6 2.4 ...

    Pith/arXiv arXiv 2025

  41. [41]

    S. M. C. Yeung, M. H. Y. Cheung, M. Zumalacarregui, and O. A. Hannuksela, (2024), arXiv:2410.19804 [astro- ph.IM]

    Pith/arXiv arXiv 2024

  42. [42]

    M. H.-Y. Cheung, K. K. Y. Ng, M. Zumalac´ arregui, and E. Berti, Phys. Rev. D109, 124020 (2024), arXiv:2403.13876 [gr-qc]

    Pith/arXiv arXiv 2024

  43. [43]

    Wright and M

    M. Wright and M. Hendry, (2021), 10.3847/1538- 4357/ac7ec2, arXiv:2112.07012 [astro-ph.HE]

    Pith/arXiv arXiv 2021

  44. [44]

    Goyal, H

    S. Goyal, H. Villarrubia-Rojo, and M. Zumalacarregui, (), arXiv:2511.xxxxx [astro-ph.CO]

  45. [45]

    Goyal, H

    S. Goyal, H. Villarrubia-Rojo, and M. Zumalacarregui, (), arXiv:251x.xxxxx [gr-qc]

  46. [46]

    Janquartet al., Mon

    J. Janquartet al., Mon. Not. Roy. Astron. Soc.526, 3832 (2023), arXiv:2306.03827 [gr-qc]. 13 0.60 0.75 0.90 q 450 600 750 900dL 1.10 1.11 1.12 1.13 1.14 δ 1.36 1.38 1.40 1.42 α 0.25 0.50 0.75 1.00 ι 0.8 1.6 2.4 ψ 1.5 3.0 4.5 6.0 φ 0.2 0.4 0.6 0.8 χ1 0.2 0.4 0.6 0.8 χ2 0.6 1.2 1.8 2.4 3.0 θ1 0.8 1.6 2.4 θ2 1.5 3.0 4.5 6.0 ∆φ 1.5 3.0 4.5 6.0 φJL 30 31 32 33...

    Pith/arXiv arXiv 2023

  47. [47]

    A. G. Abacet al.(LIGO Scientific, VIRGO, KAGRA), Astrophys. J. Lett.993, L25 (2025), arXiv:2507.08219 [astro-ph.HE]

    Pith/arXiv arXiv 2025

  48. [48]

    Abbottet al.(LIGO Scientific, KAGRA, VIRGO), Astrophys

    R. Abbottet al.(LIGO Scientific, KAGRA, VIRGO), Astrophys. J.970, 191 (2024), arXiv:2304.08393 [gr-qc]

    Pith/arXiv arXiv 2024

  49. [49]

    Abbottet al.(LIGO Scientific, VIRGO), Astrophys

    R. Abbottet al.(LIGO Scientific, VIRGO), Astrophys. J.923, 14 (2021), arXiv:2105.06384 [gr-qc]

    Pith/arXiv arXiv 2021

  50. [50]

    K. K. Y. Ng, K. W. K. Wong, T. Broadhurst, and T. G. F. Li, Phys. Rev. D97, 023012 (2018), arXiv:1703.06319 [astro-ph.CO]

    Pith/arXiv arXiv 2018

  51. [51]

    F. S. Broekgaarden, S. Banagiri, and E. Payne, As- trophys. J.969, 108 (2024), arXiv:2303.17628 [astro- ph.HE]

    Pith/arXiv arXiv 2024

  52. [52]

    G. P. Smith, A. Robertson, G. Mahler, M. Nicholl, D. Ryczanowski, M. Bianconi, K. Sharon, R. Massey, 14 J. Richard, and M. Jauzac, Mon. Not. Roy. Astron. Soc. 520, 702 (2023), arXiv:2204.12977 [astro-ph.HE]

    Pith/arXiv arXiv 2023

  53. [53]

    S.-S. Li, S. Mao, Y. Zhao, and Y. Lu, Mon. Not. Roy. Astron. Soc.476, 2220 (2018), arXiv:1802.05089 [astro- ph.CO]

    Pith/arXiv arXiv 2018

  54. [54]

    Skilling, Bayesian Analysis1, 833 (2006)

    J. Skilling, Bayesian Analysis1, 833 (2006)

    2006

  55. [55]

    A. R. A. C. Wierda, E. Wempe, O. A. Hannuksela, L. ´ e. V. E. Koopmans, and C. Van Den Broeck, Astrophys. J.921, 154 (2021), arXiv:2106.06303 [astro-ph.HE]

    Pith/arXiv arXiv 2021

  56. [56]

    Ashtonet al., Astrophys

    G. Ashtonet al., Astrophys. J. Suppl.241, 27 (2019), arXiv:1811.02042 [astro-ph.IM]

    Pith/arXiv arXiv 2019

  57. [57]

    J. S. Speagle, Mon. Not. Roy. Astron. Soc.493, 3132 (2020), arXiv:1904.02180 [astro-ph.IM]

    Pith/arXiv arXiv 2020

  58. [58]

    Qin, T.-Y

    Z. Qin, T.-Y. Sun, B.-Y. Li, J.-F. Zhang, X. Guo, and X. Zhang, (2025), arXiv:2505.20996 [gr-qc]

    Pith/arXiv arXiv 2025

  59. [59]

    Santoliquidoet al., (2025), arXiv:2504.21087 [astro- ph.HE]

    F. Santoliquidoet al., (2025), arXiv:2504.21087 [astro- ph.HE]

    arXiv 2025

  60. [60]

    M. Dax, S. R. Green, J. Gair, J. H. Macke, A. Buonanno, and B. Sch¨ olkopf, Phys. Rev. Lett.127, 241103 (2021), arXiv:2106.12594 [gr-qc]

    Pith/arXiv arXiv 2021

  61. [61]

    Bada-Nerin, O

    R. Bada-Nerin, O. Bulashenko, O. Gramaxo Freitas, and J. A. Font, Phys. Rev. D111, 084067 (2025), arXiv:2412.00566 [gr-qc]

    Pith/arXiv arXiv 2025

  62. [62]

    S. R. Green and J. Gair, Mach. Learn. Sci. Tech.2, 03LT01 (2021), arXiv:2008.03312 [astro-ph.IM]

    Pith/arXiv arXiv 2021

  63. [63]

    S. R. Green, C. Simpson, and J. Gair, Phys. Rev. D102, 104057 (2020), arXiv:2002.07656 [astro-ph.IM]

    Pith/arXiv arXiv 2020

  64. [64]

    Gupteet al., (2024), arXiv:2404.14286 [gr-qc]

    N. Gupteet al., (2024), arXiv:2404.14286 [gr-qc]

    Pith/arXiv arXiv 2024

  65. [65]

    M. Dax, S. R. Green, J. Gair, M. P¨ urrer, J. Wildberger, J. H. Macke, A. Buonanno, and B. Sch¨ olkopf, Phys. Rev. Lett.130, 171403 (2023), arXiv:2210.05686 [gr-qc]

    Pith/arXiv arXiv 2023

  66. [67]

    M. Dax, S. R. Green, J. Gair, N. Gupte, M. P¨ urrer, V. Raymond, J. Wildberger, J. H. Macke, A. Buo- nanno, and B. Sch¨ olkopf, Nature639, 49 (2025), arXiv:2407.09602 [gr-qc]

    Pith/arXiv arXiv 2025

  67. [68]

    Tambalo, M

    G. Tambalo, M. Zumalac´ arregui, L. Dai, and M. H.-Y. Cheung, Phys. Rev. D108, 103529 (2023), arXiv:2212.11960 [astro-ph.CO]

    Pith/arXiv arXiv 2023

  68. [69]

    Urrutia and V

    J. Urrutia and V. Vaskonen, Mon. Not. Roy. Astron. Soc. 509, 1358 (2021), arXiv:2109.03213 [astro-ph.CO]

    Pith/arXiv arXiv 2021

  69. [70]

    Gil Choi, S

    H. Gil Choi, S. Jung, P. Lu, and V. Takhistov, Phys. Rev. Lett.133, 101002 (2024), arXiv:2311.17829 [astro- ph.CO]

    Pith/arXiv arXiv 2024

  70. [71]

    Savastano, G

    S. Savastano, G. Tambalo, H. Villarrubia-Rojo, and M. Zumalacarregui, Phys. Rev. D108, 103532 (2023), arXiv:2306.05282 [gr-qc]

    Pith/arXiv arXiv 2023

  71. [72]

    Vujeva, J

    L. Vujeva, J. M. Ezquiaga, D. Gilman, S. Goyal, and M. Zumalac´ arregui, (2025), arXiv:2510.14953 [astro- ph.CO]

    arXiv 2025

  72. [73]

    Zumalac´ arregui, (2024), arXiv:2404.17405 [gr-qc]

    M. Zumalac´ arregui, (2024), arXiv:2404.17405 [gr-qc]

    Pith/arXiv arXiv 2024

  73. [74]

    C. R. Keeton, (2001), arXiv:astro-ph/0102341

    Pith/arXiv arXiv 2001

  74. [75]

    Brando, S

    G. Brando, S. Goyal, S. Savastano, H. Villarrubia-Rojo, and M. Zumalac´ arregui, Phys. Rev. D111, 024068 (2025), arXiv:2407.04052 [gr-qc]

    Pith/arXiv arXiv 2025

  75. [76]

    Hinshaw and L

    G. Hinshaw and L. M. Krauss, Astrophysical Journal, Part 1 (ISSN 0004-637X), vol. 320, Sept. 15, 1987, p. 468-476.320, 468 (1987)

    1987

  76. [77]

    D. L. Jow, S. Foreman, U.-L. Pen, and W. Zhu, Mon. Not. Roy. Astron. Soc.497, 4956 (2020), arXiv:2002.01570 [astro-ph.HE]

    Pith/arXiv arXiv 2020

  77. [78]

    Mishra, A

    A. Mishra, A. K. Meena, A. More, and S. Bose, Monthly Notices of the Royal Astronomical Society531, 764 (2024), https://academic.oup.com/mnras/article- pdf/531/1/764/57707112/stae836.pdf

    2024

  78. [79]

    J. F. Navarro, C. S. Frenk, and S. D. M. White, Astro- phys. J.490, 493 (1997), arXiv:astro-ph/9611107

    Pith/arXiv arXiv 1997

  79. [80]

    Thrane and C

    E. Thrane and C. Talbot, Publications of the Astronom- ical Society of Australia36(2019), 10.1017/pasa.2019.2

    doi:10.1017/pasa.2019.2 2019

  80. [81]

    J. C. L. Chan, E. Seo, A. K. Y. Li, H. Fong, and J. M. Ezquiaga, Phys. Rev. D111, 084019 (2025), arXiv:2411.13058 [gr-qc]

    Pith/arXiv arXiv 2025

Showing first 80 references.