Pith. sign in

REVIEW 3 major objections 7 minor 1 cited by

Parasitic wiggles in simulated gravitational-wave spectra are nonphysical; averaging each infrared mode after the source dies recovers the true slope near the peak.

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 · grok-4.5

2026-07-31 02:00 UTC pith:SAIL2W2B

load-bearing objection Practical fix for nonphysical IR wiggles in lattice GW spectra, plus a clean negative result that PRS corrupts GW even with rescaled sources; usable, with a real but bounded soft spot on lowest-mode stationarity. the 3 major comments →

arxiv 2607.25073 v1 pith:SAIL2W2B submitted 2026-07-27 gr-qc astro-ph.COhep-ph

Fixing IR tail of gravitational waves from domain walls

Ivan Dankovsky , Dmitry Gorbunov This is my paper
classification gr-qc astro-ph.COhep-ph
keywords gravitational wavesdomain wallslattice simulationsinfrared spectrumCosmoLatticePRS scalingpulsar timing arraysearly Universe
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.

Lattice simulations of gravitational waves from early-Universe sources such as domain-wall networks produce spectra whose infrared tails are riddled with artificial oscillations. Those wiggles come from double-frequency terms that appear when one naively squares oscillating amplitudes; they are not physical and must be removed before the spectrum can be compared with data. The deep infrared is fixed by causality, but the slope just below the peak encodes the dynamics that produced the waves and is what pulsar-timing arrays actually constrain. The authors show that simply continuing the run for a few Hubble times after the source turns off, then averaging each mode’s amplitude between its last two local peaks, washes out the artifacts and yields a clean power-law fit. They also demonstrate that the popular Press–Ryden–Spergel rescaling, which lets one run longer by freezing the wall width, produces incorrect gravitational-wave spectra even after the source terms are rescaled. The procedure is generic and can be applied to any other lattice source of primordial gravitational waves.

Core claim

The infrared wiggles seen in lattice gravitational-wave spectra are nonphysical double-frequency artifacts. Extending the simulation a few Hubble times past source termination and averaging each mode’s energy density between its latest two local peaks recovers the true smooth infrared spectrum near the peak; the Press–Ryden–Spergel artificial scaling does not.

What carries the argument

Per-mode time average (eq. 15): for each infrared wave-number k one takes the arithmetic mean of Ω_GW(k,τ) over the discrete times lying between the last two local peaks of that mode’s amplitude after the source has died.

Load-bearing premise

That a simple average of the energy density between only the latest two peaks faithfully estimates the true free-wave amplitude of every infrared mode, including the lowest lattice modes that may have completed only a few oscillations.

What would settle it

Run the same domain-wall simulation both with and without the post-source free-evolution window, apply the per-mode average, and check whether the recovered infrared slope matches an independent analytic free-wave projection or a longer high-resolution run that resolves many more periods.

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

If this is right

  • Published domain-wall and other lattice GW spectra can be reprocessed with the same averaging to obtain cleaner infrared slopes for comparison with pulsar-timing data.
  • Any future lattice study of early-Universe gravitational waves should budget a few extra Hubble times of free evolution after the source is switched off.
  • The Press–Ryden–Spergel rescaling should not be used when the goal is an accurate gravitational-wave spectrum, even if energy densities are later corrected by hand.
  • The infrared slope near the peak becomes a reliable diagnostic that can discriminate among different source-termination mechanisms.

Where Pith is reading between the lines

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

  • The same double-frequency contamination is likely present in every lattice GW code that extracts spectra from the equal-time two-point function of h′, so the fix is portable beyond CosmoLattice.
  • Once the infrared slope is trustworthy, joint fits to NANOGrav and future PTA data can begin to distinguish biased walls, melting walls, and other termination scenarios rather than treating them as a single power-law template.
  • A more rigorous estimator could replace the two-peak arithmetic mean with a projection onto the homogeneous free-wave solution, potentially reducing the required free-evolution time.

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 / 7 minor

Summary. The manuscript addresses a practical problem in lattice simulations of stochastic gravitational-wave backgrounds: the infrared part of the spectrum, obtained directly from the two-point function of h' on the lattice, exhibits strong wiggles. The authors identify these as parasitic double-frequency (2k) terms in the squared oscillation amplitudes (Eqs. 8–9), which survive for kτ ≲ 1. Using CosmoLattice simulations of domain-wall networks in three settings (rigid walls, biased potential, melting walls), they propose a per-mode time-averaging procedure (Eq. 15): after the GW source terminates, each IR mode's Ω_GW(k,τ) is averaged over the discrete time samples between its latest two local maxima, yielding smooth spectra that are then fit with power laws. They also demonstrate that the Press–Ryden–Spergel (PRS) rescaling of the scalar equation, which would naively allow longer runs, produces incorrect GW energy densities and spectra, even when the scalar energy-density components are rescaled by hand to restore their correct a-dependence.

Significance. If the procedure holds up, this is a useful and genuinely generic contribution to the lattice-GW literature: the near-peak IR slope is precisely the quantity that discriminates among sources for PTA-era observations, and the wiggle problem is common to all lattice computations of sub-horizon-sourced backgrounds. The paper has several concrete strengths. The analytic diagnosis is clean: Eq. (9) makes explicit which oscillating terms survive at kτ ≲ 1, and the choice of averaging window is better motivated than it first appears — consecutive maxima of Ω_GW ∝ cos²(kτ+φ) are separated by exactly one period of the parasitic 2k term, so Eq. (15) is, in the continuous stationary limit, an exact period average rather than an ad hoc smoothing. The method is demonstrated on three physically distinct source classes, fits are reported with uncertainties and χ²_ν values, and the PRS failure is shown convincingly at two levels (the secular decrease of ρ_GW in Fig. 8, which contradicts the analytically required constancy for a scaling wall network, and the spectral distortion in Fig. 10). The PRS critique is a falsifiable, negative result of value to several groups currently using that prescripti

major comments (3)
  1. [§4, Eq. (15) and Fig. 4] The estimator (15) is an exact period average only for a mode that has reached the stationary free-oscillation regime (constant amplitude after source termination). The lowest modes — which anchor the fitted IR slopes that are the paper's quantitative deliverable (k^2.16±0.17, k^1.47±0.03, k^1.74±0.18) — satisfy this worst: Fig. 4 (bottom) shows k = 0.066 and 0.132 completing barely one cos² period between source-off (τ=20) and the end of the run (τ=75), and §4 concedes the lowest mode 'may [need] some extrapolation' without specifying it. If these amplitudes are still drifting secularly, the fitted exponents inherit an uncontrolled bias. The paper needs a stationarity/convergence test: e.g., compare the average over the latest peak-to-peak window with that over the preceding window for each mode, or extend the biased-potential run until the lowest mode completes two full periods, and st
  2. [§4, Figs. 5–7 (error bars and fit quality)] The error bars are the standard deviation of the time samples {Ω(k,τ_i)} within a single realization — they measure residual oscillation, not the statistical uncertainty of the asymptotic amplitude. Seed-to-seed scatter (the procedure in §2 describes running multiple seeds, but the number of seeds used here is never stated) is the relevant ensemble error for the fitted spectrum. The formally poor fits (χ²_ν = 3.65 in Fig. 5, 3.77 in Fig. 6) are consistent with underestimated errors or residual non-stationarity in the lowest bins, and Fig. 7 reports a two-point fit (N=2, χ²_ν=0.38) which carries essentially no uncertainty information. Either the ensemble errors should be computed and the fits redone, or the captions/text should state plainly that the quoted errors are intra-realization and the exponents' uncertainties are underestimated.
  3. [§5, Figs. 9–10] The 'correction' to the PRS run is described only as 'artificially multiplying the scalar energy density terms by properly chosen powers of the scale factor.' It is essential to state whether this rescaling is applied dynamically to the source Π^TT_ij entering Eq. (1) during the evolution, or is a post-processing of plotted energy-density components. If the latter, the constancy of ρ_GW in Fig. 9 is achieved by construction and the claim in the abstract that PRS fails 'even if the source terms are properly rescaled' is not supported by what is shown; if the former, the procedure should be written down as an equation so the result is reproducible. This is load-bearing for the abstract's strong formulation of the PRS conclusion.
minor comments (7)
  1. [Fig. 5 caption] 'The wiggling IR modes, presented on bottom panel of Ref. 4' should presumably read 'Fig. 4' (this paper), not 'Ref. 4' (Grishchuk).
  2. [Figs. 5–7 captions] The quantity N (=5, 23, 2) is never defined; presumably it is the number of points (or degrees of freedom) entering the fit. Please define it, and state the fit range in k for each case.
  3. [§2, Eq. (7) vs. lattice setup] The symbol k_IR is used both for the lattice infrared cutoff 2π/L (below Eq. 11) and for the initial-fluctuation parameter in Eq. (7), which is then set to k_IR = 0 in the simulations. The collision is confusing; please rename one of them.
  4. [§2] The number of realizations (seeds) used in each simulation set is never reported, despite ensemble averaging being introduced in §2. Please state it for each figure.
  5. [Fig. 1 and throughout] Typos and grammar: 'highly wiggle infrared parts' (abstract), 'Intead' (§1), 'KARGO' should be 'KAGRA' (§1), 'it's physical width' (§3), 'Once its comoving size became smaller' (§3), 'wiggles in the infrared part ... is even more pronounced' (§3). A careful proofreading pass is recommended.
  6. [§5] It would help the reader to add one sentence noting that PRS was designed to reproduce wall-network dynamics (Nambu-Goto limit), not GW sourcing, so the failure demonstrated here does not bear on its uses for wall evolution observables — this clarifies the scope of the critique.
  7. [Fig. 4] Marking the source-termination time and the identified local peaks used in Eq. (15) on these panels would make the averaging procedure much easier to follow visually.

Circularity Check

0 steps flagged

No significant circularity: IR cleaning is a post-processing estimator on free-evolution modes; power-law slopes are fits to cleaned output, not inputs.

full rationale

The paper’s load-bearing chain is: (i) analytic solution of the sourced tensor equation yields parasitic double-frequency terms in Ω_GW (eqs. 8–9); (ii) after source termination each lattice mode obeys the homogeneous wave equation, so Ω_GW(k,τ) oscillates as cos²(kτ+φ); (iii) a peak-to-peak arithmetic mean (eq. 15) therefore estimates the asymptotic free-wave amplitude; (iv) the resulting smooth IR points are then fitted with a power law. None of these steps defines the target spectrum in terms of a fitted parameter, imports a uniqueness theorem from the authors’ prior work, or renames a known empirical pattern as a derivation. Self-citations ([17],[19],[24],[26]–[28]) supply only simulation setup and prior DW phenomenology; the central claim (that eq. 15 recovers the true IR slope while PRS does not) is tested directly against unscaled CosmoLattice runs and does not rest on those citations. Power-law indices and error bars are ordinary post-processing of cleaned output. Stationarity of the lowest modes is an assumption about applicability, not a circular reduction. Score 0 is therefore appropriate.

Axiom & Free-Parameter Ledger

5 free parameters · 5 axioms · 0 invented entities

Central claims rest on standard linearized GW propagation in FLRW, the CosmoLattice discretization, and the modeling choice that free evolution plus arithmetic peak-to-peak averaging recovers the physical IR spectrum. No new physical entities. Free parameters are simulation/model knobs (bias, lattice, initial spectra, averaging window definition), not fits that define the claimed slopes.

free parameters (5)
  • bias ε in Z2-breaking term = 0.025 λ η
    Sets DW destruction timescale in the biased-potential runs; chosen as ε=0.025λη for illustration, not derived.
  • initial fluctuation parameters (α, T, k_UV, k_IR) = α=-3, k_IR=0, k_UV=1, T=T_i/5
    Ansatz (7) with α=-3, k_IR=0, k_UV=1, T=T(τ_i)/5 chosen to reach scaling; affects early transient but claimed universal late DW network.
  • averaging window N_k / last-two-peaks rule = between latest two local peaks
    Eq. (15) defines which discrete times enter Ω_GW,avg(k); the two-peak rule is a hand choice that directly controls the cleaned IR points and fit slopes.
  • source switch-off time / destruction instant = τ=20 (example)
    For rigid walls, instantaneous switch-off at τ=20 is assumed to mimic fast destruction; location of the IR slope depends on this choice.
  • lattice size N and IR cutoff k_IR=2π/L = N=1024
    N=1024 fixes mode sampling and minimum k; IR systematics and number of usable fit points depend on it.
axioms (5)
  • domain assumption Linearized TT tensor modes in conformal FLRW obey eq. (1) with GW energy density given by the period-averaged (3).
    Standard GR weak-wave assumption used throughout §§2–4; no backreaction or gauge subtleties beyond Maggiore textbook citation.
  • domain assumption Radiation-dominated expansion a∝τ holds for the duration of the runs used to extract spectra.
    Analytic Green function (2) and PRS comparison assume RD; stated in §2.
  • domain assumption After source termination, each Fourier mode of h_ij evolves as a free damped oscillator whose squared amplitude average equals the physical spectral contribution.
    Underpins why late-time averaging should wash parasitic terms; implicit in §4 procedure.
  • domain assumption DW network in scaling (or its biased/melting variants) on a periodic lattice with the stated initial conditions represents the continuum cosmological source for the IR modes of interest.
    Relies on prior lattice DW literature including authors’ own CosmoLattice studies; finite-volume and resolution limits acknowledged but not eliminated.
  • ad hoc to paper Arithmetic mean of Ω_GW(k,τ_i) between the latest two peaks is an unbiased estimator of the asymptotic mode power.
    Core methodological postulate of eq. (15); not derived from a filter theory or compared to alternative estimators in the text.

pith-pipeline@v1.2.0-grok45-kimik3 · 16451 in / 3465 out tokens · 62667 ms · 2026-07-31T02:00:36.663412+00:00 · methodology

0 comments
read the original abstract

Numerical simulations of gravitational waves (GW) production during violent evolution of matter inhomogeneities in the early Universe yield highly wiggle infrared parts of the spectra. Obtained directly from the two-point correlation function, these wiggles are nonphysical, corresponding to the parasitic, double frequency terms in naively averaged squared oscillation amplitudes of GW, and hence must be washed out. The deep infrared behavior can be predicted on general grounds, e.g. fixed by causality considerations. However, what matters for real observations, e.g. like that of NANOGrav, and what can be only inferred from numerical simulations, is the infrared slope near the maximum of the spectrum. It reflects the dynamics responsible for the GW production in its heyday, and hence must be accurately predicted. We illustrate the problem with numerical simulations of the Domain Wall network performed with the help of code CosmoLattice. We suggest a numerical procedure to smooth out these parasitic wiggles, which allows us to recover the true spectrum. Being quite generic, it may be applied to numerical simulations of other hypothetical sources of GW possibly operating in the early Universe. The procedure requires to extend the simulation by a few Hubble times after termination of the GW production. We checked with numerical simulations, that the technically natural long-time extension of simulations, which becomes available via artificial scaling of different parts in the scalar sector equations provided by PRS prescription, gives wrong GW spectra even if the source terms are properly rescaled.

Figures

Figures reproduced from arXiv: 2607.25073 by Dmitry Gorbunov, Ivan Dankovsky.

Figure 1
Figure 1. Figure 1: GW spectra in model with DW (14) at subsequent conformal times from τ = 1 to τ = 20 with step ∆τ = 1/2. Rhombs and stars indicate temporary conformal momenta corresponding to the Hubble parameter and the DW width, respectively. The problem is illustrated with plots in [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 3
Figure 3. Figure 3: Typical prediction of GW spectra produced by melting domain walls [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figure 2
Figure 2. Figure 2: GW spectra at conformal times from τ = 5 to τ = 20 with step ∆τ = 1 (lines are indicated by colors changing from blue to yellow) in the model with DWs originated from the biased double-well potential, where DW network finally disappears; the GW spectrum is saturated by τ = 20. Rhombs and stars indicate temporary conformal momenta corresponding to the Hubble parameter and the DW width, respectively. 24, 25]… view at source ↗
Figure 6
Figure 6. Figure 6: GW spectra for the case of melting domain walls, where the GW [PITH_FULL_IMAGE:figures/full_fig_p005_6.png] view at source ↗
Figure 4
Figure 4. Figure 4: Time evolution of GW amplitudes for a set of modes. [PITH_FULL_IMAGE:figures/full_fig_p005_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: GW spectra for the case of biased potential with [PITH_FULL_IMAGE:figures/full_fig_p005_5.png] view at source ↗
Figure 9
Figure 9. Figure 9: The same energy density components as in Fig. [PITH_FULL_IMAGE:figures/full_fig_p006_9.png] view at source ↗
Figure 8
Figure 8. Figure 8: Time evolution of GW energy density and various terms of scalar [PITH_FULL_IMAGE:figures/full_fig_p006_8.png] view at source ↗
Figure 10
Figure 10. Figure 10: GW spectra calculated with the same physical parameters of the [PITH_FULL_IMAGE:figures/full_fig_p006_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 1 Pith paper

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

  1. Domain walls through different cosmologies

    astro-ph.CO 2026-07 accept novelty 6.5

    Domain-wall network area scales as S ≈ 2ξV/τ with ξ≈1.2 across cosmologies from dust to near-Minkowski, so the particle horizon—not H⁻¹—sets the correlation length and GW peak.

Reference graph

Works this paper leans on

25 extracted references · 17 linked inside Pith · cited by 1 Pith paper

  1. [2]

    CosmoLattice: A modern code for lattice simulations of scalar and gauge field dynamics in an expanding universe,

    D. G. Figueroa, A. Florio, F. Torrenti, and W. Valkenburg, “CosmoLattice: A modern code for lattice simulations of scalar and gauge field dynamics in an expanding universe,”Comput. Phys. Commun.283 (2023) 108586,arXiv:2102.01031 [astro-ph.CO]

    Pith/arXiv arXiv 2023

  2. [3]

    Dynamical Evolution of Domain Walls in an Expanding Universe,

    W. H. Press, B. S. Ryden, and D. N. Spergel, “Dynamical Evolution of Domain Walls in an Expanding Universe,” Astrophys. J.347(1989) 590–604

    1989

  3. [4]

    Amplification of gravitational waves in an isotropic universe,

    L. P. Grishchuk, “Amplification of gravitational waves in an isotropic universe,”Sov. Phys. JETP40no. 3, (1975) 409–415

    1975

  4. [5]

    Spectrum of relict gravitational radiation and the early state of the universe,

    A. A. Starobinsky, “Spectrum of relict gravitational radiation and the early state of the universe,”JETP Lett. 30(1979) 682–685

    1979

  5. [6]

    V . A. Rubakov and D. S. Gorbunov,Introduction to the Theory of the Early Universe: Hot big bang theory. World Scientific, Singapore, 2017

    2017

  6. [7]

    Cosmological backgrounds of gravitational waves,

    C. Caprini and D. G. Figueroa, “Cosmological backgrounds of gravitational waves,”Class. Quant. Grav. 35no. 16, (2018) 163001,arXiv:1801.04268 [astro-ph.CO]. [8]KAGRA, VIRGO, LIGO ScientificCollaboration, R. Abbottet al., “Open Data from the Third Observing Run of LIGO, Virgo, KAGRA, and GEO,”Astrophys. J. Suppl.267no. 2, (2023) 29,arXiv:2302.03676 [gr-qc]...

    Pith/arXiv arXiv 2018

  7. [10]

    Search for an Isotropic Gravitational-wave Background with the Parkes Pulsar Timing Array,

    D. J. Reardonet al., “Search for an Isotropic Gravitational-wave Background with the Parkes Pulsar Timing Array,”Astrophys. J. Lett.951no. 1, (2023) L6, arXiv:2306.16215 [astro-ph.HE]

    Pith/arXiv arXiv 2023

  8. [11]

    Searching for the Nano-Hertz Stochastic Gravitational Wave Background with the Chinese Pulsar Timing Array Data Release I,

    H. Xuet al., “Searching for the Nano-Hertz Stochastic Gravitational Wave Background with the Chinese Pulsar Timing Array Data Release I,”Res. Astron. Astrophys.23 no. 7, (2023) 075024,arXiv:2306.16216 [astro-ph.HE]. [12]EPTA, InPTA:Collaboration, J. Antoniadiset al., “The second data release from the European Pulsar Timing Array - III. Search for gravitat...

    Pith/arXiv arXiv 2023

  9. [14]

    The NANOGrav 15 yr Data Set: Running of the Spectral Index,

    G. Agazieet al., “The NANOGrav 15 yr Data Set: Running of the Spectral Index,”Astrophys. J. Lett.978 no. 2, (2025) L29,arXiv:2408.10166 [astro-ph.HE]

    Pith/arXiv arXiv 2025

  10. [15]

    Present and future ofCosmoLattice,

    D. G. Figueroa, A. Florio, and F. Torrenti, “Present and future ofCosmoLattice,”Rept. Prog. Phys.87no. 9, (2024) 094901,arXiv:2312.15056 [astro-ph.CO]

    Pith/arXiv arXiv 2024

  11. [16]

    Maggiore,Gravitational Waves

    M. Maggiore,Gravitational Waves. V ol. 1: Theory and Experiments. Oxford University Press, 2007

    2007

  12. [17]

    Cosmic domain walls on a lattice: Illusive effects of initial conditions,

    I. Dankovsky, S. Ramazanov, E. Babichev, D. Gorbunov, and A. Vikman, “Cosmic domain walls on a lattice: Illusive effects of initial conditions,”Phys. Rev. D112 no. 12, (2025) 123521,arXiv:2509.25367 [hep-ph]

    arXiv 2025

  13. [18]

    On the estimation of gravitational wave spectrum from cosmic domain walls,

    T. Hiramatsu, M. Kawasaki, and K. Saikawa, “On the estimation of gravitational wave spectrum from cosmic domain walls,”JCAP02(2014) 031,arXiv:1309.5001 [astro-ph.CO]

    Pith/arXiv arXiv 2014

  14. [19]

    Revisiting evolution of domain walls and their gravitational radiation with CosmoLattice,

    I. Dankovsky, E. Babichev, D. Gorbunov, S. Ramazanov, and A. Vikman, “Revisiting evolution of domain walls and their gravitational radiation with CosmoLattice,” JCAP09(2024) 047,arXiv:2406.17053 [astro-ph.CO]. 7

    Pith/arXiv arXiv 2024

  15. [20]

    Cosmological Consequences of the Spontaneous Breakdown of Discrete Symmetry,

    Y . B. Zeldovich, I. Y . Kobzarev, and L. B. Okun, “Cosmological Consequences of the Spontaneous Breakdown of Discrete Symmetry,”Zh. Eksp. Teor . Fiz. 67(1974) 3–11

    1974

  16. [21]

    Gravitational waves from domain wall collapse, and application to nanohertz signals with QCD-coupled axions,

    N. Kitajima, J. Lee, K. Murai, F. Takahashi, and W. Yin, “Gravitational waves from domain wall collapse, and application to nanohertz signals with QCD-coupled axions,”Phys. Lett. B851(2024) 138586, arXiv:2306.17146 [hep-ph]

    Pith/arXiv arXiv 2024

  17. [22]

    Near-peak spectrum of gravitational waves from collapsing domain walls,

    B. Cyr, S. Cotterill, and R. Battye, “Near-peak spectrum of gravitational waves from collapsing domain walls,” Phys. Rev. D113no. 4, (2026) 043549, arXiv:2504.02076 [astro-ph.CO]

    Pith/arXiv arXiv 2026

  18. [23]

    The spectrum of gravitational waves from annihilating domain walls,

    A. Notari, F. Rompineve, and F. Torrenti, “The spectrum of gravitational waves from annihilating domain walls,” JCAP07(2025) 049,arXiv:2504.03636 [astro-ph.CO]

    Pith/arXiv arXiv 2025

  19. [24]

    Biased domain walls: faster annihilation, weaker gravitational waves,

    E. Babichev, I. Dankovsky, D. Gorbunov, S. Ramazanov, and A. Vikman, “Biased domain walls: faster annihilation, weaker gravitational waves,”JCAP10 (2025) 103,arXiv:2504.07902 [hep-ph]

    arXiv 2025

  20. [25]

    Biased Domain Wall Networks and their Gravitational Waves,

    D. Barbini, A. Notari, O. Pujolàs, F. Rompineve, and F. Torrentí, “Biased Domain Wall Networks and their Gravitational Waves,”arXiv:2607.18107 [astro-ph.CO]

    Pith/arXiv arXiv

  21. [26]

    Beyond freeze-in: Dark matter via inverse phase transition and gravitational wave signal,

    S. Ramazanov, E. Babichev, D. Gorbunov, and A. Vikman, “Beyond freeze-in: Dark matter via inverse phase transition and gravitational wave signal,”Phys. Rev. D105no. 6, (2022) 063530,arXiv:2104.13722 [hep-ph]

    Pith/arXiv arXiv 2022

  22. [27]

    Gravitational shine of dark domain walls,

    E. Babichev, D. Gorbunov, S. Ramazanov, and A. Vikman, “Gravitational shine of dark domain walls,” JCAP04no. 04, (2022) 028,arXiv:2112.12608 [hep-ph]

    Pith/arXiv arXiv 2022

  23. [28]

    Numerical analysis of melting domain walls and their gravitational waves,

    I. Dankovsky, S. Ramazanov, E. Babichev, D. Gorbunov, and A. Vikman, “Numerical analysis of melting domain walls and their gravitational waves,”JCAP02(2025) 064,arXiv:2410.21971 [hep-ph]

    Pith/arXiv arXiv 2025

  24. [29]

    Evading the cosmological domain wall problem,

    S. E. Larsson, S. Sarkar, and P. L. White, “Evading the cosmological domain wall problem,”Phys. Rev. D55 (1997) 5129–5135,arXiv:hep-ph/9608319

    Pith/arXiv arXiv 1997

  25. [30]

    Evolution of domain wall networks: The Press-Ryden-Spergel algorithm,

    L. Sousa and P. P. Avelino, “Evolution of domain wall networks: The Press-Ryden-Spergel algorithm,”Phys. Rev. D81(2010) 087305,arXiv:1101.3350 [hep-th]. 8

    Pith/arXiv arXiv 2010