Incorporating Backreaction in One-Loop Corrections in Ultra-Slow-Roll Inflation

Chao Chen; Cheng-Jun Fang; Zhen-Hong Lyu; Zong-Kuan Guo

REVIEW 3 major objections 3 minor 4 cited by

Incorporating Backreaction in One-Loop Corrections in Ultra-Slow-Roll Inflation

T0 review · 3 major / 3 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read Backreaction cancels one-loop super-horizon corrections in ultra-slow-roll inflation.

desk verdict Plausible and careful in-in treatment of backreaction, but the headline cancellation is for ζ-bar, not the observable ζ, and the missing third-order gauge transformation is load-bearing. read the letter →

arxiv 2502.09555 v2 pith:ADIPIK4T submitted 2025-02-13 gr-qc

Cheng-Jun Fang , Zhen-Hong Lyu , Chao Chen , Zong-Kuan Guo This is my paper · ORCID

classification gr-qc

keywords ultra-slow-rollinflationprimordialblackholescurvaturepowerspectrumone-loopcorrectionsbackreactionin-informalismsuper-horizonconservationzeromode

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

This paper argues that the one-loop super-horizon corrections to the curvature power spectrum found in earlier in-in calculations for ultra-slow-roll inflation (a phase used in primordial-black-hole models) are an artifact of neglecting backreaction. When the backreaction of curvature perturbations on the background is included in the same formalism, each of the three one-loop contributions is cancelled exactly by a corresponding backreaction term. The authors conclude that the previously reported enhancement of the large-scale power spectrum during a sudden ultra-slow-roll phase does not survive a complete one-loop treatment. If correct, the large-scale curvature perturbation is conserved at one-loop order, resolving a disagreement between direct in-in calculations and several consistency-based arguments.

What carries the argument

The load-bearing object is the separation of the curvature perturbation into quantized non-zero-momentum modes and a zero-momentum background component. Only $k\neq 0$ modes are quantized as harmonic oscillators; the $q=0$ mode is absorbed into the background correction $P_2 = \dot{\bar\phi}_2/\dot{\bar\phi}_0$, whose equation of motion is sourced by the bilinear expectation value $\langle\delta\phi^2\rangle$. The zero-mode Green function used to solve for $P_2$ is defined as the $q\to 0$ limit of the non-zero-mode Green function, and the standard mode-function normalization (fixed by the Wronskian) converts the source integrals into the same momentum integrals appearing in the one-loop diagrams. Because the $q=0$ mode is excluded before the field pairings (Wick contractions) are made, the induced fourth-order term acquires a characteristic $2/3$ factor relative to the quartic Hamiltonian $H_{4I}$ used in earlier work, and this same exclusion is what allows the backreaction terms to match the loop terms term by term.

What would settle it

Repeat the derivation of equations (19), (21) and (22) keeping the $q=0$ mode inside the quantized field rather than absorbing it into $P_2$; if the backreaction terms then no longer cancel the loop terms term by term, the claimed exact cancellation depends on the zero-mode subtraction and would fail if that split is not physical.

Watch

Extended reading notes

Core claim

The central claim is that the one-loop correction to the super-horizon (larger-than-Hubble) curvature power spectrum in single-field ultra-slow-roll inflation vanishes once backreaction from the perturbations is treated consistently. Working in the spatially-flat gauge, the authors split the background scalar field as $\bar\phi = \bar\phi_0 + \bar\phi_2$, where $\bar\phi_2$ is the second-order response of the background to quantum fluctuations. They then show that the third-order, induced fourth-order, and quartic self-interaction one-loop contributions, equations (19), (21) and (22), are each cancelled by terms arising from the backreaction quantity $P_2 = \dot{\bar\phi}_2/\dot{\bar\phi}_0$, with the net correction given by equation (27). The cancellations are claimed to be exact in the sharp-transition limit used in the earlier literature, and they identify the previously reported enhancement as coming from the omission of the $q=0$ mode, which here is treated as part of the background rather than as a quantum fluctuation.

Load-bearing premise

The whole cancellation rests on the claim that the zero-momentum mode can be cleanly separated from the quantized fluctuations and treated as part of the classical background; if that split is not the correct physical treatment, the exact cancellation could be an artifact of the subtraction.

Editorial extensions

If this is right

If the cancellation is exact, the large-scale curvature power spectrum is conserved through the super-horizon phase at one-loop order in the sharp-transition limit.
The one-loop enhancement claimed in earlier direct in-in calculations would no longer be available to boost the large-scale spectrum in ultra-slow-roll models.
The in-in formalism and the equation-of-motion approach that already found a vanishing super-horizon correction would be reconciled, since both now give conservation.
Any one-loop in-in calculation that keeps the $q=0$ mode inside the quantized field would need to justify why that mode is not part of the backreaction-treated background.

Reading between the lines

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

The same mode split could be applied to other sharp-transition models with enhanced small-scale power; if the cancellation persists, the conservation result would generalize beyond the specific ultra-slow-roll setup treated here.
A numerical evaluation of the loop integrals with a small but nonzero $q$, taken to zero after integration, would test whether the exact cancellation depends on the order of limits inherent in the zero-mode Green function definition.
A lattice simulation that lets the background respond to the amplified small-scale modes could check whether the semiclassical $P_2$ equation (33) reproduces the loop-level cancellation; agreement would support the semiclassical backreaction treatment, while disagreement would locate its limitation.
If the cancellation holds beyond the sharp-transition limit, it would suggest that consistency relations protect the super-horizon spectrum even when individual diagrams are large, making the earlier direct-in-in enhancements a boundary-term or gauge artifact.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

Thanks for the report. Here's my read.

The paper does a serious job: it sets up the in-in calculation in the spatially-flat gauge, treats the q=0 mode as a classical background (P2), derives the backreaction correction to the Hamiltonian, and shows term-by-term cancellation of the three one-loop contributions. The 2/3 factor relative to the H4I-type induced quartic term is a real subtlety, and the appendices are detailed enough that a patient referee can check the steps. The authors also cite the prior cancellation results [22] and [34] honestly; they don't oversell novelty.

That said, the central claim is weaker than the abstract suggests. The cancellation is for ζ-bar, defined in Eq. (13), and the paper uses a linear gauge identification ζ = ζ-bar at the end of inflation (footnote 3). The Discussions section explicitly concedes that the complete third-order gauge transformation was not computed. This is not a cosmetic gap: a quadratic term in the exact ζ(δφ) relation, contracted with two fluctuations, contributes to ⟨ζ²⟩ at exactly the one-loop order being canceled. The slow-roll suppression invoked elsewhere is not obviously applicable here because the second-order gauge corrections involve USR mode functions with enhanced time dependence. So the advertised conclusion — that the one-loop superhorizon correction to the observed curvature power spectrum vanishes — is not yet demonstrated. What is demonstrated is cancellation for the auxiliary variable ζ-bar in a semiclassical backreaction scheme.

Other soft spots are more minor: the physical momentum cutoffs Λ0, Λ1 are a choice that is discussed but not compared quantitatively with comoving cutoffs; slow-roll suppressed terms are dropped without a size estimate; and the backreaction treatment is semiclassical. The self-consistency structure (P2 solved from an equation sourced by the same correlator it cancels) is a mean-field reorganization, not a logical flaw, but it does mean the 'exact' cancellation is tied to the truncation.

Who's this for? Cosmologists working on loop corrections in USR and PBH formation. It's a useful method paper and it sharpens the debate, but the physical takeaway is conditional until the gauge issue is settled.

Recommendation: definitely send it to a serious referee. Ask for an estimate or computation of the third-order gauge terms, and a quantified statement about slow-roll corrections. With that, it could be a solid publication; without it, the conclusion should be limited to ζ-bar.

Referee Report

3 major / 3 minor

Summary. The manuscript investigates one-loop corrections to the primordial curvature power spectrum in a single-field ultra-slow-roll (USR) inflationary model with a sharp transition. Working in the spatially flat gauge, the authors separate the inflaton perturbation into a quantum part containing only non-zero momentum modes and a classical homogeneous background correction induced by backreaction. They derive the backreaction-corrected cubic and quartic interaction Hamiltonians, compute the one-loop contributions from the standard cubic and quartic vertices, solve for the backreaction variable P2 from its equation of motion, and show that the three one-loop corrections (19), (21), and (22) are canceled by the backreaction corrections (50), (51), and (53). The claimed result is that the superhorizon one-loop correction to the curvature power spectrum vanishes once backreaction is included, which would reconcile direct in-in calculations with conservation and consistency-relation arguments.

Significance. If correct, the paper offers a systematic way to incorporate backreaction effects within the in-in formalism and would resolve a live controversy about USR one-loop corrections. The manuscript is carefully organized, with detailed appendices, an explicit treatment of the Gibbons-Hawking-York boundary term, and a transparent statement of its approximations. The main result, however, is conditional on an incomplete third-order gauge transformation and on a specific treatment of zero modes, so its significance for the observable curvature perturbation ζ is not yet established.

major comments (3)

[Discussions; footnote 3] Footnote 3 and the Discussions explicitly state that the complete third-order gauge transformation is not included and that the calculation is performed for ζ̄ rather than the observed ζ. The central cancellation in Eqs. (50), (51), and (53) is therefore a statement about ζ̄. A nonlinear relation of the form ζ = ζ̄ + O(ζ̄^2), when inserted into ⟨ζ_p ζ_−p⟩, produces one-loop contractions of exactly the order of the terms being canceled. Because USR mode functions can have enhanced time dependence, a slow-roll suppression argument is not obviously applicable to those terms. The paper should either compute these gauge contributions and show that they are subleading, or explicitly restrict the abstract and title to ζ̄.
[Appendix B, Eq. (B8)] The factor 2/3 in the induced fourth-order correction, and with it the cancellation of Eq. (21) with Eq. (51), follows from the identity ∫_{q≠0} d^3q δ^(3)(q)=0. This is a distributional statement about the internal loop momentum q, not about the k=0 Fourier mode of ζ̄; the configurations with k1+k2=0 that are removed contain pairs of non-zero modes whose product has zero total momentum. In a finite-volume regularization the q=0 contribution is proportional to the box volume and does not vanish, so the factor may change. The authors should justify the exclusion with an explicit regulator, for example box normalization with the k=0 mode treated as a classical background, and show that the 2/3 factor is regulator-independent.
[Eqs. (33)-(38) and (46)-(48)] The backreaction variable P2 is solved from Eq. (33), whose source is the same type of one-loop correlator ⟨ζ^2⟩ that P2 is later used to cancel in Eqs. (50)-(53). This is a legitimate self-consistent mean-field reorganization if the source is evaluated at the correct perturbative order, but the paper does not state explicitly at which order ⟨ζ^2⟩ is to be substituted and why the procedure is not a tautology. In addition, the derivation uses physical cutoffs from Eq. (41) and relies on the cancellation in Eq. (C4) between cutoff-derivative terms and the B contribution; the Discussions acknowledge that the choice of cutoffs is open. The authors should verify that the final cancellations are independent of the cutoff scheme, or identify the class of cutoffs for which they hold.

minor comments (3)

[One-loop corrections section] In the paragraph preceding Eq. (19), the text reads 'We can approximate η as η = ∆η δ(τ − τe)'; the formulas require η′ = ∆η δ(τ − τe), so this is a typographical error.
[Eqs. (19)-(20)] The notation ⟨⟨ ... ⟩⟩ is used in Eqs. (19), (21), and (22) before it is defined in Eq. (20); please move the definition earlier or add a gloss at first use.
[Title and text] There are typographical issues in the title line ('Ult ra-Slow-Roll') and in the paragraph before Eq. (21) ('comv-ing gauge') that should be corrected.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the backreaction field is solved from the action's equations of motion, not fitted to cancel the loop corrections.

full rationale

The central cancellation is a computed identity, not an input. The paper expands the action in δφ, derives the backreaction/tadpole equation (11) from δδφS1 + ⟨δδφS3⟩ = 0, defines ζbar and P2 via (13), computes the one-loop contributions (19), (21), and (22) with the in-in formalism, and independently solves for P2 through the sourced equation of motion (33) using the Green's function (37). The final cancellations (50), (51), and (53) follow from explicit integrals and the sharp-transition approximations; P2 is not a free parameter chosen to reproduce the loop terms. The equation-of-motion input from Ref. [22] is external to this author list and functions as a cross-check, not as a self-citation chain. The zero-mode subtraction and the resulting 2/3 factor are physical assumptions about how q=0 modes are separated, and the incomplete third-order gauge transformation acknowledged in the Discussion is a limitation of the presented calculation, but neither constitutes a circular reduction of the claimed result to its own premises.

Assumptions & free parameters 1 free parameters · 6 assumptions · 0 invented entities

The calculation introduces no new physical entities or fitted constants. It rests on a specific separation of zero modes into the background, a physical-cutoff regularization, a Hartree-type semiclassical backreaction, and the sharp-transition approximation; the central cancellation is derived within these assumptions rather than from a fully gauge-invariant and cutoff-independent framework.

free parameters (1)

Physical IR and UV cutoffs Lambda_0 and Lambda_1 = not numerically specified
Chosen physical energy scales in Eq. (41) to regulate the momentum integrals and to compute the time derivative in Eq. (40); the cancellation is claimed to be independent of their exact values, but the split between loop and backreaction depends on this scheme.

assumptions (6)

domain assumption Only non-zero momentum modes are quantized as harmonic oscillators; the q=0 mode is treated as part of the classical background.
Footnote 1 and the split phi-bar = phi-bar_0 + phi-bar_2 in Eq. (10); the entire backreaction method depends on this separation.
domain assumption Physical cut-offs do not introduce additional time scales in de Sitter spacetime, so their time derivatives are physical.
Invoked at Eq. (41) with a citation to [32]; used to derive the time derivative of the zeta-squared expectation value in Eq. (40).
domain assumption Semiclassical Hartree-type backreaction is sufficient for the one-loop cancellation.
Discussions section: 'we adopted a semiclassical approach to backreaction [25], leaving further exploration of quantum backreaction to future work.'
domain assumption The sharp USR-to-SR transition is approximated by eta'(tau) = Delta_eta delta(tau-tau_e) with Delta_eta approximately -6, and the contribution at tau_s is neglected.
Section 'One-loop corrections', following [1]; used to evaluate the loop integrals (19) and (48).
domain assumption Slow-roll suppressed terms and higher-order constraint corrections can be neglected.
Stated after Eq. (11) and again in Discussions; if these terms contribute at the same order, the exact cancellation could change.
domain assumption The approximations Im[zeta_q(tau_0) zeta_q*'(tau_e)] approximately 1/(4 a^2 epsilon) and Im[zeta_q(tau_0) zeta_q*(tau_e)] approximately 0 hold.
Eq. (49) and Appendix C, inherited from [1]; used to evaluate P2^(1).

how reviews work

0 comments

Cite this review

Pith. "Pith review of Incorporating Backreaction in One-Loop Corrections in Ultra-Slow-Roll Inflation." pith.science (2026). https://pith.science/paper/ADIPIK4T

@misc{pith2026250209555,
  author       = {Pith},
  title        = {Pith review of: Incorporating Backreaction in One-Loop Corrections in Ultra-Slow-Roll Inflation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ADIPIK4T}},
  note         = {Machine review of arXiv:2502.09555}
}

read the original abstract

We investigate the one-loop quantum correction to the power spectrum of primordial curvature perturbations in the ultra-slow-roll (USR) inflationary scenario, incorporating the backreaction effect from curvature perturbations. In the spatially-flat gauge, we expand the background inflaton field up to second order and identify the one-loop level backreaction term in the action. Utilizing a gauge transformation, we derive the comoving curvature interaction Hamiltonian in the presence of the backreaction term and calculate the one-loop correction using the in-in formalism. Our results reveal that the one-loop super-horizon corrections previously reported in the literature are canceled by the backreaction contributions. This finding underscores the importance of accounting for the backreaction effects in the analysis of quantum corrections during USR inflation.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 4 Pith papers

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

Scale-Dependent Loop Corrections to the Inflationary Power Spectrum
astro-ph.CO 2026-03 conditional novelty 7.0 of 10

One-loop gravitational corrections in inflationary models with scale-dependent features are renormalizable and vanish on large and small scales, preserving perturbativity of CMB-fit feature models.
Efficient training of photonic quantum generative models
quant-ph 2026-03 unverdicted novelty 5.0 of 10

Photonic quantum generative models can be trained classically via maximum mean discrepancy, with deployment corresponding to boson sampling.
Evolution of Linear Perturbations under Time-Dependent Hubble Friction I: SR-USR-SR Inflation
gr-qc 2026-02 conditional novelty 5.0 of 10

Analytic asymptotics show the dip in the SR-USR-SR curvature power spectrum comes from cancellation between two growing modes, not a constant-versus-growing cancellation.
One-loop corrections to infrared GWs is forbidden by symmetries
gr-qc 2025-08 conditional novelty 5.0 of 10

In non-attractor single-field inflation, one-loop corrections to superhorizon gravitational waves cancel, and a Ward identity explains why.

Reference graph

Works this paper leans on

36 extracted references · 5 canonical work pages · cited by 4 Pith papers

[22]

Kawaguchi, S

R. Kawaguchi, S. Tsujikawa, and Y. Yamada, JHEP 12, 095 (2024) , arXiv:2407.19742 [hep-th]

arXiv 2024
[34]

Huenupi, E

J. Huenupi, E. Hughes, G. A. Palma, and S. Sypsas, Phys. Rev. D 110, 123536 (2024) , arXiv:2406.07610 [hep-th]

arXiv 2024
[1]

Kristiano and J

J. Kristiano and J. Yokoyama, Phys. Rev. Lett. 132, 221003 (2024) , arXiv:2211.03395 [hep-th]

arXiv 2024
[2]

Fumagalli, (2023), arXiv:2305.19263 [astro-ph.CO]

J. Fumagalli, (2023), arXiv:2305.19263 [astro-ph.CO]

arXiv 2023
[3]

J. M. Bardeen, Physical Review D 22, 1882 (1980)

1980
[4]

Kodama and M

H. Kodama and M. Sasaki, Progress of Theoretical Physics Supplement 78, 1 (1984)

1984
[5]

D. H. Lyth, K. A. Malik, and M. Sasaki, JCAP 05, 004 (2005) , arXiv:astro-ph/0411220

arXiv 2005
[6]

Senatore and M

L. Senatore and M. Zaldarriaga, JHEP 09, 148 (2013) , arXiv:1210.6048 [hep-th]

arXiv 2013

Show all 36 references

[7]

Assassi, D

V. Assassi, D. Baumann, and D. Green, JHEP 02, 151 (2013) , arXiv:1210.7792 [hep-th]

2013 arXiv
[8]

Kristiano and J

J. Kristiano and J. Yokoyama, Phys. Rev. D 109, 103541 (2024) , arXiv:2303.00341 [hep-th]

2024 arXiv
[9]

(11) As we will demonstrate later, the action acquires correc- tions involving ¯φ2 due to backreaction eﬀects, resulting in a corresponding correction to the two-point function

is simpliﬁed to ¨ ¯φ2 + 3H ˙¯φ2 + V(2)( ¯φ0) ¯φ2 = − 12 V(3)( ¯φ0) ⟨ δφ2⟩ . (11) As we will demonstrate later, the action acquires correc- tions involving ¯φ2 due to backreaction eﬀects, resulting in a corresponding correction to the two-point function. One-loop corrections. T...
[10]

Firouzjahi, Phys

H. Firouzjahi, Phys. Rev. D 109, 043514 (2024) , arXiv:2311.04080 [astro-ph.CO]

2024 arXiv
[11]

Firouzjahi, JCAP 10, 006 (2023) , arXiv:2303.12025 [astro-ph.CO]

H. Firouzjahi, JCAP 10, 006 (2023) , arXiv:2303.12025 [astro-ph.CO]

2023 arXiv
[12]

Kristiano and J

J. Kristiano and J. Yokoyama, JCAP 10, 036 (2024) , arXiv:2405.12145

2024 arXiv
[13]

Sheikhahmadi and A

H. Sheikhahmadi and A. Nassiri-Rad, (2024), arXiv:2411.18525 [astro-ph.CO]

2024 arXiv
[14]

Franciolini, A

G. Franciolini, A. Iovino, Junior., M. Taoso, and A. Urbano, Phys. Rev. D 109, 123550 (2024) , arXiv:2305.03491 [astro-ph.CO]

2024 arXiv
[15]

Choudhury, M

S. Choudhury, M. R. Gangopadhyay, and M. Sami, Eur. Phys. J. C 84, 884 (2024) , arXiv:2301.10000 [astro-ph.CO] . 11

2024 arXiv
[16]

M. W. Davies, L. Iacconi, and D. J. Mulryne, JCAP 04, 050 (2024) , arXiv:2312.05694 [astro-ph.CO]

2024 arXiv
[17]

Iacconi, D

L. Iacconi, D. Mulryne, and D. Seery, JCAP 06, 062 (2024) , arXiv:2312.12424 [astro-ph.CO]

2024 arXiv
[18]

Y. Tada, T. Terada, and J. Tokuda, JHEP 01, 105 (2024) , arXiv:2308.04732 [hep-th]

2024 arXiv
[19]

Maity, H

S. Maity, H. V. Ragavendra, S. K. Sethi, and L. Sriramkumar, JCAP 05, 046 (2024) , arXiv:2307.13636 [astro-ph.CO]

2024 arXiv
[20]

Riotto, (2023), arXiv:2301.00599 [astro-ph.CO]

A. Riotto, (2023), arXiv:2301.00599 [astro-ph.CO]

2023 arXiv
[21]

6 Finally we solve P (2) 2 from Eq

in the limit p ≪ k, it cancels out with ( 51). 6 Finally we solve P (2) 2 from Eq. ( 39) P (2) 2 (τ0) = 4 ∫ τ0 −∞ dτ [a2(τ )ǫ(τ )]2V(4)( ¯φ0(τ )) × Im[¯ζq(τ0)¯ζ⋆ q (τ )] ∫ d3k (2π)3 ⏐ ⏐¯ζk(τ ) ⏐ ⏐2 , q → 0 . (52) Its contribution to the correlation function is given by ⟨ ⟨¯ζp ...
[23]

Fumagalli, (2024), arXiv:2408.08296 [astro-ph.CO]

J. Fumagalli, (2024), arXiv:2408.08296 [astro-ph.CO]

2024 arXiv
[24]

Inomata, Phys

K. Inomata, Phys. Rev. Lett. 133, 141001 (2024) , arXiv:2403.04682 [astro-ph.CO]

2024 arXiv
[25]

V. F. Mukhanov, L. R. W. Abramo, and R. H. Brandenberger, Phys. Rev. Lett. 78, 1624 (1997) , arXiv:gr-qc/9609026

1997 arXiv
[26]

L. R. W. Abramo, R. H. Brandenberger, and V. F. Mukhanov, Phys. Rev. D 56, 3248 (1997) , arXiv:gr-qc/9704037

1997 arXiv
[27]

Schander and T

S. Schander and T. Thiemann, Front. Astron. Space Sci. 0, 113 (2021) , arXiv:2106.06043 [gr-qc]

2021 arXiv
[28]

Caravano, G

A. Caravano, G. Franciolini, and S. Renaux-Petel, (2024), arXiv:2410.23942 [astro-ph.CO]

2024 arXiv
[29]

Caravano, K

A. Caravano, K. Inomata, and S. Renaux- Petel, Phys. Rev. Lett. 133, 151001 (2024) , arXiv:2403.12811 [astro-ph.CO]

2024 arXiv
[30]

Syu, D.-S

W.-C. Syu, D.-S. Lee, and K.-W. Ng, Phys. Rev. D 101, 025013 (2020) , arXiv:1907.13089 [gr-qc]

2020 arXiv
[31]

Cheng, D.-S

S.-L. Cheng, D.-S. Lee, and K.- W. Ng, Phys. Lett. B 827, 136956 (2022) , arXiv:2106.09275 [astro-ph.CO]

2022 arXiv
[32]

Cheng, D.-S

S.-L. Cheng, D.-S. Lee, and K.-W. Ng, JCAP 03, 008 (2024) , arXiv:2305.16810 [astro-ph.CO]

2024
[33]

G. W. Gibbons and S. W. Hawking, Physical Review D 15, 2738 (1977)

1977
[35]

Kristiano and J

J. Kristiano and J. Yokoyama, (2025), arXiv:2504.18514 [hep-th]

2025 arXiv
[36]

Inomata, Phys

K. Inomata, Phys. Rev. D 111, 103504 (2025) , arXiv:2502.08707 [astro-ph.CO]

2025 arXiv

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.

[1] [22]

Kawaguchi, S

R. Kawaguchi, S. Tsujikawa, and Y. Yamada, JHEP 12, 095 (2024) , arXiv:2407.19742 [hep-th]

arXiv 2024

[2] [34]

Huenupi, E

J. Huenupi, E. Hughes, G. A. Palma, and S. Sypsas, Phys. Rev. D 110, 123536 (2024) , arXiv:2406.07610 [hep-th]

arXiv 2024

[3] [1]

Kristiano and J

J. Kristiano and J. Yokoyama, Phys. Rev. Lett. 132, 221003 (2024) , arXiv:2211.03395 [hep-th]

arXiv 2024

[4] [2]

Fumagalli, (2023), arXiv:2305.19263 [astro-ph.CO]

J. Fumagalli, (2023), arXiv:2305.19263 [astro-ph.CO]

arXiv 2023

[5] [3]

J. M. Bardeen, Physical Review D 22, 1882 (1980)

1980

[6] [4]

Kodama and M

H. Kodama and M. Sasaki, Progress of Theoretical Physics Supplement 78, 1 (1984)

1984

[7] [5]

D. H. Lyth, K. A. Malik, and M. Sasaki, JCAP 05, 004 (2005) , arXiv:astro-ph/0411220

arXiv 2005

[8] [6]

Senatore and M

L. Senatore and M. Zaldarriaga, JHEP 09, 148 (2013) , arXiv:1210.6048 [hep-th]

arXiv 2013