Pith. sign in

REVIEW 3 major objections 5 minor 57 references

Axion inflation can pass through a long, steady mild-backreaction phase at large gauge coupling, with homogeneous backreaction still valid — and CMB non-Gaussianity caps the axion-gauge coupling times Hubble rate.

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

2026-08-04 06:04 UTC pith:NVAOXUJ7

load-bearing objection A genuinely useful diagnostic framework for when homogeneous backreaction fails, plus a clean CMB bound; the headline β=50 'mild backreaction' discovery is real but unvalidated and should be treated as a tentative example, not a settled regime. the 3 major comments →

arxiv 2608.02498 v1 pith:NVAOXUJ7 submitted 2026-08-03 astro-ph.CO

Axion Inflation: Perturbative control in the strong backreaction regime

Valerie Domcke , Alexandros Papageorgiou , Marco Peloso , Stefan Sandner This is my paper
classification astro-ph.CO
keywords axion inflationgauge field productionstrong backreactiongradient expansion formalismin-in perturbation theoryparticle production parameterCMB non-Gaussianityprimordial gravitational waves
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 asks when the standard homogeneous-backreaction approximation for axion inflation — treating the axion as uniform while keeping gauge fields inhomogeneous — remains trustworthy, and what happens when it breaks down. It develops two complementary perturbative diagnostics for the backreaction term ⟨E·B⟩: a gradient-expansion formalism extended to first order in axion gradients, and a one-loop in-in computation. Both are validated against lattice simulations at moderate couplings. At large coupling, β=50 in a quadratic-potential benchmark, the paper discovers an extended mild-backreaction phase lasting roughly 20 e-folds, with a large, nearly constant particle-production parameter ξ rising from about 4.12 to 5.05 and no oscillations — the first such regime it knows of. It also derives a model-independent CMB bound βH/Mp < 1.4×10⁻³ and shows that, under nearly constant ξ and H, the strong-backreaction region lies generically beyond perturbative control for ξ ≳ 2.5.

Core claim

The central claim is a new dynamical phase in axion inflation: at large axion-gauge coupling, the system enters a prolonged mild-backreaction regime in which gauge friction is significant but subdominant to Hubble friction, and ξ grows slowly and monotonically without the oscillatory bursts seen at lower couplings. In the β=50 example, mild backreaction starts near N≈27.6, strong backreaction only near N≈47.2, and the in-in correction to the spectral backreaction is enhanced but shows no sign of the infrared-to-ultraviolet cascade that accompanies breakdown at smaller β. The gradient-expansion evaluation sees no significant deviation in the inflationary trajectory, so homogeneous backreactio

What carries the argument

The organizing object is the first-order correction δ⁽¹⁾B to the backreaction correlator B=⟨E·B⟩ caused by axion inhomogeneities; the paper treats the ratio |δ⁽¹⁾B|/|B⁽⁰⁾| as the perturbativity criterion, with 10% as the operational onset of nonlinearity. This correction is computed by two methods: a first-order gradient expansion formalism that adds axion-gradient terms to a tower of gauge-field correlators, and a one-loop in-in calculation that takes the homogeneous-backreaction solution as the unperturbed background and the δφ F F̃ interaction as the vertex. The spectral decomposition of δ⁽¹⁾B is what reveals whether power is redistributed from infrared to ultraviolet modes.

Load-bearing premise

The one-loop in-in truncation — which keeps only the interference term between one vacuum and one sourced gauge leg, discards the [δφ,δφ]⟨A⁴⟩ piece and all two-loop sourced-sourced contributions, and sets the correction to zero for k>k_max — is assumed to capture the dominant correction at β=50, even though those high-k modes are exactly where lattice simulations at smaller β find the IR-to-UV cascade that breaks homogeneous backreaction.

What would settle it

Run a lattice simulation of the same quadratic-potential model at β=50 and check whether the gauge-field spectrum develops support beyond k_max and whether the axion trajectory departs from the homogeneous-backreaction solution before N≈47; if either happens, the extended mild-backreaction phase is an artifact of the homogeneous approximation.

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

If this is right

  • At couplings up to β=25, the two diagnostics agree with each other and with lattice results: homogeneous backreaction fails once axion gradient energy reaches roughly 1–5% of kinetic energy, and the one-loop spectrum shows IR-to-UV power redistribution.
  • At β=50, homogeneous backreaction can remain accurate through a ~20 e-fold mild-backreaction phase with large, slowly varying ξ — a regime that predicts sustained, non-oscillatory gauge-field production and possibly observable gravitational waves at interferometer scales.
  • Under nearly constant ξ and H, the strong-backreaction condition σ=1 lies above the perturbativity line for ξ≳2.5, meaning strong friction generically coincides with loss of perturbative control.
  • CMB non-Gaussianity imposes a model-independent bound βH/Mp ≲1.4×10⁻³ for an axion inflaton, which already restricts how large couplings can be explored consistently.
  • The in-in perturbativity criterion is presented as sufficient but not necessary: a 10% one-loop correction can appear while the background evolution remains essentially unchanged, so the gradient-expansion criterion should be used to decide when homogeneous backreaction actually breaks down.

Where Pith is reading between the lines

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

  • If the β=50 steady phase is real, gravitational-wave and primordial-black-hole predictions in the strong-backreaction literature may need revision: recurring bursts driven by ξ oscillations would be replaced by a quieter, long-duration signal — a distinction a dedicated lattice run at β=50 could settle.
  • The ~20 e-fold gap between the in-in and gradient-expansion trigger times suggests that the one-loop interference term underestimates the role of sourced-sourced contributions; a two-loop calculation with support beyond k_max would test whether the IR-to-UV cascade eventually appears at β=50.
  • Because the CMB bound forces βH/Mp<1.4×10⁻³ in single-field axion inflation, the interesting large-ξ strong-backreaction phenomenology may be reachable only in axion-spectator or multi-field setups, which bypass the bound by not sourcing the observed CMB perturbations.

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. This paper studies axion inflation with a Chern-Simons coupling to an Abelian gauge field, focusing on the strong-backreaction regime. The authors propose a perturbativity criterion based on the leading correction δB^(1) to the homogeneous-backreaction correlator ⟨E·B⟩, and implement it with two complementary methods: a first-order gradient-expansion formalism (GEF) that includes axion gradients, and a one-loop in-in computation. The two diagnostics are validated against existing lattice simulations for β = 15, 18, 20, 25, and then applied to β = 50, where the paper claims an extended mild-backreaction regime with slowly growing ξ ≈ 4.1 → 5.1 and no oscillations. The paper also derives a constant-ξ,H analytic perturbativity threshold βH/Mp ≈ 5.6×10² ξ e^{−πξ}, concluding that strong backreaction is generically non-perturbative for ξ ≳ 2.5, and a CMB normalization bound βH/Mp < 1.4×10⁻³ that is independent of the inflaton potential.

Significance. The main deliverable is a set of fast diagnostics that can guide lattice simulations and assess the validity of homogeneous backreaction in parameter regions where full lattice runs are currently impractical. The validation at β ≤ 25 is convincing: both criteria track the departure of the homogeneous approximation from the lattice, and they are consistent with the gradient-energy criterion of Ref. [37]. The analytic threshold (4.7) and the CMB bound (4.10) are simple, potentially useful results. If the β = 50 regime is real, it would be the first example of a long, steady, large-ξ mild-backreaction phase, with implications for gravitational-wave phenomenology. However, the central new claim is not yet fully supported: it rests on the GEF diagnostic in a situation where the in-in diagnostic disagrees by roughly 20 e-folds, and on a one-loop truncation that cannot probe the k > k_max modes implicated in the IR→UV cascade. The paper is honest about this limitation, but the abstract's 'discover' is stronger than the evidence presented.

major comments (3)
  1. [Sec. 4.2, Sec. 5, Fig. 4 (right panel)] The central claim of an extended mild-backreaction regime at β=50 rests on the assumption that homogeneous backreaction remains valid between N≈27.6 and N≈47.2. The GEF-based criterion is triggered only at N≈47.2, while the one-loop in-in criterion triggers at N≈27.6; Sec. 5 states 'we are unable to fully explain the origin of this difference.' Because the one-loop correction (Eq. 3.9, App. A) is built from one vacuum and one sourced gauge leg, it has support only for k ≤ k_max (Eq. 4.5), which is precisely the momentum range where the IR→UV cascade is known to break homogeneous backreaction for β≤25 (Sec. 4.1, citing Fig. 2 of [31]). At β=50 no lattice or two-loop computation is available, so the absence of IR-to-UV transfer in Fig. 5 may be an artifact of the truncation. To support the abstract's 'discover' claim, the authors should provide a quantitative estimate of the two-loop sourc
  2. [App. A, Eq. (A.5) and Eq. (3.9)] The dominance argument for the one-loop in-in computation is internally inconsistent. The text states 'the contribution proportional to [δφ,δφ]⟨A⁴⟩ is subdominant and can be ignored', but Eq. (3.9) explicitly exhibits a term with both [A,A] and [δφ,δφ] commutators, and Eq. (A.5) evaluates an expression proportional to Im[δφ(τ₁)δφ*(τ₂)] times a four-gauge-field correlator of the same type. Either the equations or the text are misprinted. Since this one-loop result is one of the two diagnostics used to certify the β=50 regime, the authors must clarify which structure is retained and justify, with a quantitative estimate, why the omitted one is negligible.
  3. [Sec. 4.3, Eq. (4.6), Fig. 6] The analytic conclusion that strong backreaction is generically non-perturbative for ξ≳2.5 is based on the perturbativity condition (4.6), a ratio of one-loop to tree-level gauge power spectra at horizon crossing, while the criterion used in the dynamical examples is the correction to the integrated ⟨E·B⟩ backreaction, Eq. (4.2). The paper states these two 'are expected to yield very similar results' but provides only a single cross-check (the black star in Fig. 6). If the analytic bound is to be used as a general statement, the correspondence between (4.2) and (4.6) should be demonstrated numerically over a representative range of the (ξ, βH/Mp) plane, or the bound should be re-derived directly for δB^(1)/B^(0).
minor comments (5)
  1. [Eq. (2.13)] The inequality in (2.13) is written with ≲, while the condition it summarizes, Eq. (2.12), uses ≳. The text subsequently uses (2.13) as a weak-backreaction bound, which is consistent with the derivation, but the wording 'this condition can be written more simply as' is misleading. Please clarify the direction or state explicitly that (2.13) is the bound below which backreaction is weak.
  2. [Sec. 4.3, App. B, Eq. (4.7)] Eq. (4.7) is a semi-empirical fit: the prefactor 5.6×10² and the linear-in-ξ form of g(ξ)^{-1/2} are determined by fitting Eq. (B.4) to the numerical red line. The appendix states this, but the main text presents the expression without emphasizing its fitted nature. Please state explicitly in Sec. 4.3 that the prefactor and functional form are fitted, and give the fit range and claimed precision.
  3. [Fig. 4 (right panel)] The right panel spans N∈[10,70]; the star markers for the two criteria and the gradient-energy vertical lines are difficult to read. Adding labels with the numerical values N≈27.6 and N≈47.2 would improve the figure.
  4. [Sec. 4.1] The statement 'by construction, the modes are not enhanced beyond that scale' refers to the one-loop computation using homogeneous-backreaction mode functions, not to the full theory. Please add a clause clarifying that this is a property of the truncation.
  5. [Sec. 4.4, Eq. (4.10)] The bound (4.10) is called model-independent, but it relies on the assumption that ξ and H are approximately constant over the CMB window. This is stated in the text, yet the abstract's phrase 'independently of the choice of the axion potential' could be read as stronger. A footnote reiterating the constant-ξ,H assumption at CMB scales would prevent over-interpretation.

Circularity Check

0 steps flagged

No significant circularity: the central diagnostic is calibrated against external lattice data, and the analytic bound is an openly labeled semi-empirical fit, not a fitted quantity disguised as a prediction.

full rationale

I walked the derivation chain from the homogeneous-backreaction equations (Sec. 2) through the two perturbativity diagnostics (Sec. 3) to the β=50 example, the constant-ξ,H bound, and the CMB limit (Sec. 4). The GEF and in-in criteria are both defined as corrections to the actual backreaction correlator, not as a relabeling of any target result. The two methods are calibrated against existing lattice simulations [31,34] for β=15,18,20,25, providing external benchmarks; the β=50 extrapolation is a numerical claim within a self-consistent approximation, and its unresolved discrepancy with the GEF trigger time is openly admitted. The analytic perturbativity threshold Eq. (4.7) is explicitly described as a fitted function derived from the paper's own in-in integrals (App. B), and the conclusion that strong backreaction is non-perturbative for ξ≳2.5 follows from comparing that fitted line with a separate analytic backreaction condition (Eq. 2.13), so it is not a tautology. The CMB bound Eq. (4.10) uses an external non-Gaussianity bound and standard slow-roll relations, not an input of this paper. Self-citations to Refs. [37] and [46] supply computational tools and a known perturbativity result, but the tools are independently benchmarked and the cited result is not used as an unverified uniqueness claim. The main weaknesses—the in-in truncation's support only up to k_max, the unresolved 20 e-fold split, and the lack of a lattice check at β=50—are validation gaps and correctness risks, not circularity. I therefore find no load-bearing circular step; the appropriate score is 1.

Axiom & Free-Parameter Ledger

4 free parameters · 6 axioms · 0 invented entities

The central claim rests on one convention (10% onset), one fitted prefactor (5.6×10²), and four domain assumptions about the perturbative expansion (Gaussianity, one-loop dominance, negligible metric perturbations, helicity truncation) plus the external ξ_CMB ≲ 2.5 input. No new entities are invented. The heaviest unexamined weight is the one-loop-discards-k>k_max assumption, which coincides with the regime where lattice data are absent.

free parameters (4)
  • Prefactor of analytic perturbativity threshold, g(ξ)^{-1/2} = 5.6×10² ξ = 5.6×10²
    App. B: 'After performing a fit to the numerical results, with a polynomial g(ξ)^{-1/2} we find excellent agreement for g(ξ)^{-1/2} = 5.6·10² ξ.' Eq. (4.7) is therefore semi-empirical, not derived.
  • Perturbativity onset threshold |δ^(1)B| = 0.1 |B^(0)| = 0.1 (10%)
    Eq. (4.2), Sec. 1: 'Operationally, we use a ten-percent correction as the onset of a relevant nonlinear effect.' Chosen by hand; calibrated to lattice only for β = 15–25.
  • Mild/strong backreaction thresholds σ = 0.1 / 1 = 0.1, 1
    Eq. (2.12) defines which regime is 'mild' vs 'strong'; a classification convention, but the β=50 discovery's characterization as 'mild' depends on σ = 0.1.
  • Lower cutoff ξ = 2.5 for analytic analysis = 2.5
    Sec. 4.3: below ξ = 2.5 the vacuum/enhanced scale separation fails and UV-cutoff choices give O(1) uncertainties; the claimed non-perturbativity of strong backreaction is stated for ξ ≳ 2.5.
axioms (6)
  • domain assumption Unperturbed gauge and axion fluctuations are Gaussian under homogeneous backreaction; the ⟨A⁴⟩ correlator factorizes into products of two-point functions and the enhanced-mode commutator [A,A] is negligible versus ⟨AA⟩.
    Sec. 3.2 and App. A (Eq. A.5 and surrounding text): 'the ⟨A⁴⟩ correlator can be expressed as the combinatorial sum of products of two-point functions' and the [δφ,δφ]⟨A⁴⟩ term is dropped because ⟨AA⟩ ≫ [A,A]. Load-bearing for the entire one-loop computation.
  • domain assumption The one-loop interference term (one vacuum leg, one sourced leg) is the leading nonlinear correction to ⟨E·B⟩; two-loop sourced-sourced contributions and the scalar-commutator term are subdominant.
    Sec. 3.3 and 4.2: the paper notes the one-loop term has no support beyond k_max, and that two loops would be a much better indicator of IR→UV transfer; App. A discards [δφ,δφ]⟨A⁴⟩. This assumption is decisive for the β=50 interpretation.
  • domain assumption Scalar metric perturbations are negligible compared with the direct axion–gauge interactions.
    Sec. 3 (footnote): 'We instead can ignore scalar metric perturbations… they are not sourced directly by the tachyonically enhanced gauge fields [6].'
  • domain assumption ξ_CMB ≲ 2.5 from CMB scalar non-Gaussianity, with ξ and H approximately constant over the CMB window, and H monotonically decreasing afterwards.
    Sec. 4.4, Eqs. (4.8)-(4.10): the model-independent bound βH/Mp < 1.4×10⁻³ inherits the validity assumptions of the non-Gaussianity constraint from Ref. [6] (self-cited) and the monotonicity of H. The paper states this condition explicitly.
  • domain assumption Only the tachyonically enhanced gauge helicity contributes for ξ ≥ 2.5.
    Sec. 4.3: 'We maintain only the gauge field helicity that is tachyonically enhanced…' justified by exponential suppression; used for both the analytic criterion and the parameter-space plot.
  • standard math Standard QFT in a fixed quasi–de Sitter background; Whittaker mode functions (2.9) solve Eq. (2.7) for constant ξ and H.
    Sec. 2, Eq. (2.9): the analytic mode functions are the backbone of the constant-ξ,H bounds in Sec. 4.3 and App. B.

pith-pipeline@v1.3.0-daily-deepseek · 26226 in / 27049 out tokens · 269007 ms · 2026-08-04T06:04:47.371923+00:00 · methodology

0 comments
read the original abstract

We study the strong backreaction regime in axion inflation, in which the friction from Abelian gauge fields generated via the pseudo-scalar interaction $\frac{\beta}{4 M_p} \phi F {\tilde F}$ is comparable to Hubble friction. The non-linear dynamics associated with the gauge fields render this regime phenomenologically particularly interesting, but also notoriously difficult to study with perturbative methods. Quantifying the perturbative control through the direct impact on the spectral backreaction using (i) a first-order gradient-expansion formalism including axion gradients, and (ii) a one-loop in-in calculation, we discover an extended mild backreaction regime at large couplings $\beta$, with a relatively large, nearly constant particle production parameter $\xi$. In passing, we point out that CMB non-Gaussianity bounds impose a general upper limit on the product of axion gauge field coupling and Hubble parameter during inflation, $\beta H/M_p < 1.4\cdot 10^{-3}$, independently of the choice of the axion potential.

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

57 extracted references · 50 linked inside Pith

  1. [1]

    Freese, J.A

    K. Freese, J.A. Frieman and A.V. Olinto,Natural inflation with pseudo - Nambu-Goldstone bosons,Phys. Rev. Lett.65(1990) 3233

    1990

  2. [2]

    Turner and L.M

    M.S. Turner and L.M. Widrow,Inflation Produced, Large Scale Magnetic Fields,Phys. Rev. D 37(1988) 2743

    1988

  3. [3]

    Garretson, G.B

    W.D. Garretson, G.B. Field and S.M. Carroll,Primordial magnetic fields from pseudoGoldstone bosons,Phys. Rev. D46(1992) 5346 [hep-ph/9209238]

    Pith/arXiv arXiv 1992

  4. [4]

    Anber and L

    M.M. Anber and L. Sorbo,N-flationary magnetic fields,JCAP10(2006) 018 [astro-ph/0606534]

    Pith/arXiv arXiv 2006

  5. [5]

    Barnaby and M

    N. Barnaby and M. Peloso,Large Nongaussianity in Axion Inflation,Phys. Rev. Lett.106 (2011) 181301 [1011.1500]

    Pith/arXiv arXiv 2011

  6. [6]

    Barnaby, R

    N. Barnaby, R. Namba and M. Peloso,Phenomenology of a Pseudo-Scalar Inflaton: Naturally Large Nongaussianity,JCAP04(2011) 009 [1102.4333]

    Pith/arXiv arXiv 2011

  7. [7]

    Sorbo,Parity violation in the Cosmic Microwave Background from a pseudoscalar inflaton, JCAP06(2011) 003 [1101.1525]

    L. Sorbo,Parity violation in the Cosmic Microwave Background from a pseudoscalar inflaton, JCAP06(2011) 003 [1101.1525]

    Pith/arXiv arXiv 2011

  8. [8]

    Cook and L

    J.L. Cook and L. Sorbo,Particle production during inflation and gravitational waves detectable by ground-based interferometers,Phys. Rev. D85(2012) 023534 [1109.0022]

    Pith/arXiv arXiv 2012

  9. [9]

    Barnaby, E

    N. Barnaby, E. Pajer and M. Peloso,Gauge Field Production in Axion Inflation: Consequences for Monodromy, non-Gaussianity in the CMB, and Gravitational Waves at Interferometers, Phys. Rev. D85(2012) 023525 [1110.3327]

    Pith/arXiv arXiv 2012

  10. [10]

    Domcke, M

    V. Domcke, M. Pieroni and P. Bin´ etruy,Primordial gravitational waves for universality classes of pseudoscalar inflation,JCAP06(2016) 031 [1603.01287]

    Pith/arXiv arXiv 2016

  11. [11]

    Bastero-Gil and A.T

    M. Bastero-Gil and A.T. Manso,Parity violating gravitational waves at the end of inflation, JCAP08(2023) 001 [2209.15572]

    Pith/arXiv arXiv 2023

  12. [12]

    ¨Ozsoy, A

    O. ¨Ozsoy, A. Papageorgiou and M. Fasiello,Scale-dependent chirality as a smoking gun for Abelian gauge fields during inflation,JCAP12(2024) 008 [2405.14963]

    Pith/arXiv arXiv 2024

  13. [13]

    Linde, S

    A. Linde, S. Mooij and E. Pajer,Gauge field production in supergravity inflation: Local non-Gaussianity and primordial black holes,Phys. Rev. D87(2013) 103506 [1212.1693]

    Pith/arXiv arXiv 2013

  14. [14]

    Bugaev and P

    E. Bugaev and P. Klimai,Axion inflation with gauge field production and primordial black holes,Phys. Rev. D90(2014) 103501 [1312.7435]

    Pith/arXiv arXiv 2014

  15. [15]

    Garcia-Bellido, M

    J. Garcia-Bellido, M. Peloso and C. Unal,Gravitational waves at interferometer scales and primordial black holes in axion inflation,JCAP12(2016) 031 [1610.03763]

    Pith/arXiv arXiv 2016

  16. [16]

    Garcia-Bellido, M

    J. Garcia-Bellido, M. Peloso and C. Unal,Gravitational Wave signatures of inflationary models from Primordial Black Hole Dark Matter,JCAP09(2017) 013 [1707.02441]

    Pith/arXiv arXiv 2017

  17. [17]

    ¨Ozsoy and G

    O. ¨Ozsoy and G. Tasinato,Inflation and Primordial Black Holes,Universe9(2023) 203 [2301.03600]

    Pith/arXiv arXiv 2023

  18. [18]

    Franciolini, N

    G. Franciolini, N. Ijaz and M. Peloso,Primordial black hole dark matter from axion inflation, 2604.27496

    Pith/arXiv arXiv

  19. [19]

    Kawasaki, K

    M. Kawasaki, K. Murai and S. Tsuchida,Lattice study of primordial black hole formation in bumpy axion inflation,2607.01780

    Pith/arXiv arXiv

  20. [20]

    Anber and L

    M.M. Anber and L. Sorbo,Naturally inflating on steep potentials through electromagnetic dissipation,Phys. Rev. D81(2010) 043534 [0908.4089]

    Pith/arXiv arXiv 2010

  21. [21]

    Domcke, V

    V. Domcke, V. Guidetti, Y. Welling and A. Westphal,Resonant backreaction in axion inflation, JCAP09(2020) 009 [2002.02952]. – 29 –

    Pith/arXiv arXiv 2020

  22. [22]

    Peloso and L

    M. Peloso and L. Sorbo,Instability in axion inflation with strong backreaction from gauge modes,JCAP01(2023) 038 [2209.08131]

    Pith/arXiv arXiv 2023

  23. [23]

    Cheng, W

    S.-L. Cheng, W. Lee and K.-W. Ng,Numerical study of pseudoscalar inflation with an axion-gauge field coupling,Phys. Rev. D93(2016) 063510 [1508.00251]

    Pith/arXiv arXiv 2016

  24. [24]

    Notari and K

    A. Notari and K. Tywoniuk,Dissipative Axial Inflation,JCAP12(2016) 038 [1608.06223]

    Pith/arXiv arXiv 2016

  25. [25]

    Dall’Agata, S

    G. Dall’Agata, S. Gonz´ alez-Mart ´ ın, A. Papageorgiou and M. Peloso,Warm dark energy,JCAP 08(2020) 032 [1912.09950]

    Pith/arXiv arXiv 2020

  26. [26]

    Garcia-Bellido, A

    J. Garcia-Bellido, A. Papageorgiou, M. Peloso and L. Sorbo,A flashing beacon in axion inflation: recurring bursts of gravitational waves in the strong backreaction regime,JCAP01 (2024) 034 [2303.13425]

    Pith/arXiv arXiv 2024

  27. [27]

    Barbon, N

    M. Barbon, N. Ijaz and M. Peloso,Axion inflation in the regime of homogeneous backreaction, JCAP06(2026) 013 [2510.17207]

    arXiv 2026

  28. [28]

    Baker, J

    M.J. Baker, J. Iguaz Juan and L. Sorbo,Stability in Axion Inflation with Strong Backreaction from a Massive Vector Boson,2607.22893

    Pith/arXiv arXiv

  29. [29]

    Caravano, E

    A. Caravano, E. Komatsu, K.D. Lozanov and J. Weller,Lattice simulations of Abelian gauge fields coupled to axions during inflation,Phys. Rev. D105(2022) 123530 [2110.10695]

    Pith/arXiv arXiv 2022

  30. [30]

    Caravano, E

    A. Caravano, E. Komatsu, K.D. Lozanov and J. Weller,Lattice simulations of axion-U(1) inflation,Phys. Rev. D108(2023) 043504 [2204.12874]

    Pith/arXiv arXiv 2023

  31. [31]

    Figueroa, J

    D.G. Figueroa, J. Lizarraga, A. Urio and J. Urrestilla,Strong Backreaction Regime in Axion Inflation,Phys. Rev. Lett.131(2023) 151003 [2303.17436]

    Pith/arXiv arXiv 2023

  32. [32]

    Caravano and M

    A. Caravano and M. Peloso,Unveiling the nonlinear dynamics of a rolling axion during inflation,JCAP01(2025) 104 [2407.13405]

    Pith/arXiv arXiv 2025

  33. [33]

    Sharma, A

    R. Sharma, A. Brandenburg, K. Subramanian and A. Vikman,Lattice simulations of axion-U(1) inflation: gravitational waves, magnetic fields, and scalar statistics,JCAP05 (2025) 079 [2411.04854]

    Pith/arXiv arXiv 2025

  34. [34]

    Figueroa, J

    D.G. Figueroa, J. Lizarraga, N. Loayza, A. Urio and J. Urrestilla,Nonlinear dynamics of axion inflation: A detailed lattice study,Phys. Rev. D111(2025) 063545 [2411.16368]

    arXiv 2025

  35. [35]

    Lizarraga, C

    J. Lizarraga, C. L´ opez-Mediavilla and A. Urio,Comparative study of the strong backreaction regime in axion inflation: the effect of the potential,JCAP11(2025) 020 [2505.19950]

    Pith/arXiv arXiv 2025

  36. [36]

    Jamieson, A

    D. Jamieson, A. Caravano and E. Komatsu,Primordial power spectrum and bispectrum from lattice simulations of axion-U(1) inflation,Phys. Rev. D112(2025) 103531 [2507.22285]

    arXiv 2025

  37. [37]

    Domcke, Y

    V. Domcke, Y. Ema and S. Sandner,Perturbatively including inhomogeneities in axion inflation,JCAP03(2024) 019 [2310.09186]

    Pith/arXiv arXiv 2024

  38. [38]

    Sobol, E.V

    O.O. Sobol, E.V. Gorbar and S.I. Vilchinskii,Backreaction of electromagnetic fields and the Schwinger effect in pseudoscalar inflation magnetogenesis,Phys. Rev. D100(2019) 063523 [1907.10443]

    Pith/arXiv arXiv 2019

  39. [39]

    Gorbar, K

    E.V. Gorbar, K. Schmitz, O.O. Sobol and S.I. Vilchinskii,Gauge-field production during axion inflation in the gradient expansion formalism,Phys. Rev. D104(2021) 123504 [2109.01651]

    Pith/arXiv arXiv 2021

  40. [40]

    Durrer, O

    R. Durrer, O. Sobol and S. Vilchinskii,Backreaction from gauge fields produced during inflation,Phys. Rev. D108(2023) 043540 [2303.04583]

    Pith/arXiv arXiv 2023

  41. [41]

    von Eckardstein, M

    R. von Eckardstein, M. Peloso, K. Schmitz, O. Sobol and L. Sorbo,Axion inflation in the strong-backreaction regime: decay of the Anber-Sorbo solution,JHEP11(2023) 183 [2309.04254]. [42]Planckcollaboration,Planck 2015 results. XVII. Constraints on primordial non-Gaussianity, Astron. Astrophys.594(2016) A17 [1502.01592]. – 30 –

    Pith/arXiv arXiv 2023

  42. [43]

    Domcke and K

    V. Domcke and K. Mukaida,Gauge Field and Fermion Production during Axion Inflation, JCAP11(2018) 020 [1806.08769]

    Pith/arXiv arXiv 2018

  43. [44]

    von Eckardstein, K

    R. von Eckardstein, K. Schmitz and O. Sobol,Gravitational waves from axion inflation in the gradient expansion formalism. Part I. Pure axion inflation,JHEP01(2026) 018 [2508.00798]

    Pith/arXiv arXiv 2026

  44. [45]

    Gorbar, K

    E.V. Gorbar, K. Schmitz, O.O. Sobol and S.I. Vilchinskii,Hypermagnetogenesis from axion inflation: Model-independent estimates,Phys. Rev. D105(2022) 043530 [2111.04712]

    Pith/arXiv arXiv 2022

  45. [46]

    Peloso, L

    M. Peloso, L. Sorbo and C. Unal,Rolling axions during inflation: perturbativity and signatures, JCAP09(2016) 001 [1606.00459]

    Pith/arXiv arXiv 2016

  46. [47]

    Adshead, J.T

    P. Adshead, J.T. Giblin, M. Pieroni and Z.J. Weiner,Constraining axion inflation with gravitational waves from preheating,Phys. Rev. D101(2020) 8 [1909.12842]

    Pith/arXiv arXiv 2020

  47. [48]

    Adshead, J.T

    P. Adshead, J.T. Giblin, M. Pieroni and Z.J. Weiner,Constraining Axion Inflation with Gravitational Waves across 29 Decades in Frequency,Phys. Rev. Lett.124(2020) 17 [1909.12843]

    Pith/arXiv arXiv 2020

  48. [49]

    Micha and I.I

    R. Micha and I.I. Tkachev,Turbulent thermalization,Phys. Rev. D70(2004) 043538 [hep-ph/0403101]

    Pith/arXiv arXiv 2004

  49. [50]

    Podolsky, G.N

    D.I. Podolsky, G.N. Felder, L. Kofman and M. Peloso,Equation of state and beginning of thermalization after preheating,Phys. Rev. D73(2006) 023501 [hep-ph/0507096]

    Pith/arXiv arXiv 2006

  50. [51]

    Adshead, J.T

    P. Adshead, J.T. Giblin, T.R. Scully and E.I. Sfakianakis,Gauge-preheating and the end of axion inflation,JCAP12(2015) 034 [1502.06506]

    Pith/arXiv arXiv 2015

  51. [52]

    Cuissa and D.G

    J.R.C. Cuissa and D.G. Figueroa,Lattice formulation of axion inflation. Application to preheating,JCAP06(2019) 002 [1812.03132]

    Pith/arXiv arXiv 2019

  52. [53]

    Sobol, R

    O.O. Sobol, R. von Eckardstein, E.R. Koch, S.V. Gurevich, U. Thiele and K. Schmitz, Nonlinear physics of axion inflation,JHEP06(2026) 166 [2603.02570]

    arXiv 2026

  53. [54]

    Ballardini, M

    M. Ballardini, M. Braglia, F. Finelli, G. Marozzi and A.A. Starobinsky,Energy-momentum tensor and helicity for gauge fields coupled to a pseudo-scalar inflaton,Phys. Rev. D100 (2019) 123542 [1910.13448]

    Pith/arXiv arXiv 2019

  54. [55]

    Animali, P

    C. Animali, P. Conzinu and G. Marozzi,On adiabatic renormalization with a physically motivated infrared cut-off,JCAP05(2022) 026 [2201.05602]

    Pith/arXiv arXiv 2022

  55. [56]

    Durrer, R

    R. Durrer, R. von Eckardstein, D. Garg, K. Schmitz, O. Sobol and S. Vilchinskii,Scalar perturbations from inflation in the presence of gauge fields,Phys. Rev. D110(2024) 043533 [2404.19694]

    Pith/arXiv arXiv 2024

  56. [57]

    Linde,Axions in inflationary cosmology,Phys

    A.D. Linde,Axions in inflationary cosmology,Phys. Lett. B259(1991) 38

    1991

  57. [58]

    Almeida, N

    J.P.B. Almeida, N. Bernal, D. Bettoni and J. Rubio,Chiral gravitational waves and primordial black holes in UV-protected Natural Inflation,JCAP11(2020) 009 [2007.13776]. – 31 –

    Pith/arXiv arXiv 2020