{"id":"4b742c52-7bbd-4d6d-aaf9-36204713eab0","arxiv_id":"2608.12736","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Tracing out short-wavelength modes in a radiation-dominated FLRW universe gives a non-Markovian, non-Gaussian Langevin equation for the order parameter, with explicitly computed memory and noise kernels.","lead":"This theory paper derives a non-Markovian, non-Gaussian Langevin equation for the phase-transition field in a radiation-dominated expanding universe by tracing out short-wavelength quantum fluctuations. A generalist might read it because it converts a quantum-field-theory description of cosmological vacuum decay into a form suitable for numerical simulation of gravitational waves and baryogenesis.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The plane-wave mode functions used for all kernels are invalid precisely at the vacuum-decay epoch η ≈ η0, so the quantitative memory/noise kernels do not apply to the claimed decays.","rationale":"The reader's weakest assumption identifies exactly the same load-bearing concern. The paper itself restricts the plane-wave mode function (2.14) to early times, but its physical target is the decay near η0. Eq. (2.16) shows that a²m² changes sign at η0, so the exact modes (2.17) are not oscillatory plane waves there; using stable oscillator modes removes the tachyonic instability from the environment and changes the dissipation and noise structure. This is a correctness risk for the quantitative kernels, not for the formal SK/Langevin construction, which is standard and would survive with (2.17). I therefore keep the reader's CONDITIONAL verdict. The finite-L weight-function typo in Eq. (3.47) and the vacuum-versus-thermal initial state are real secondary issues, but the mode-function regime is the most load-bearing because it directly governs the kernels in the decay epoch.","tokens_in":19048,"tokens_out":10773,"duration_ms":110199,"concrete_test":"Recompute the spatial-averaged one-loop kernel 2 Im ∫ d³r G>² of Eq. (3.31) using the exact Parabolic Cylinder mode function (2.17) for η1 and η2 within one Hubble time before η0, and compare it with the plane-wave expression (3.31). If the exact result differs by an O(1) factor or grows as η → η0, the plane-wave kernels (3.20)-(3.50) cannot quantify vacuum decay; if it matches, the approximation is vindicated.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The quantitative core of the paper, namely the averaged memory kernel in Eqs. (3.20)-(3.31) and the noise correlators in Eqs. (3.37)-(3.50), is built from the plane-wave mode function (2.14), whose own validity condition is stated in Sec. 2.2 as η0 ≫ η. But the physical regime of interest, vacuum decay in a phase transition, is precisely η near η0: from Eq. (2.16), a²m² = M²T0²(1 − η²/η0²) vanishes and then turns negative at η0, so the exact mode function (2.17) develops the tachyonic instability that drives the decay. Replacing it with the plane-wave solution (2.14), which has constant positive m̃², removes this instability entirely: the σ environment becomes a collection of stable oscillators for all η, and no spinodal or tachyonic enhancement can appear in the computed kernels. Consequently, Eqs. (3.20)-(3.50) cannot be used to quantify vacuum decay near η0; at best they describe the early-time regime η0 ≫ η, far from the transition. This does not destroy the formal SK/Langevin framework, which would survive if the exact mode functions (2.17) were used, but it does undermine the specific quantitative claim that the dissipation and noise kernels are the computable kernels of cosmological vacuum decay.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript constructs an open-quantum-system description of a real scalar order parameter undergoing a first-order phase transition in a radiation-dominated FLRW background. It splits the field Φ into a spatially homogeneous mean field φ and short-wavelength fluctuations σ, traces out σ using the Schwinger-Keldysh/in-in formalism, and derives a Langevin-type equation for φ with a non-Markov memory kernel and a generally non-Gaussian noise. For a polynomial potential with temperature-dependent mass, the influence functional is computed to second order in the σ propagators, and explicit expressions are derived for the spatially averaged memory kernel (Sec. 3.1) and for noise correlators in large- and small-scale limits (Sec. 3.2). A generalized Hubbard-Stratonovich transformation is used to represent higher powers of the quantum field φ_Δ as cumulants of a stochastic field ξ.","tokens_in":19340,"tokens_out":10757,"duration_ms":104944,"significance":"If the quantitative results were valid, the paper would provide a useful real-time alternative to Euclidean bounce calculations for cosmological first-order phase transitions, with computable noise and dissipation kernels that could feed into numerical simulations of bubble nucleation. The formal SK derivation, Keldysh rotation, and the generalized HS representation are competently executed and self-contained, and the paper is transparent about the plane-wave approximation and about the Markov approximation's limitations. However, the specific quantitative kernels are computed with mode functions that are invalid in the decay epoch, and the spatial averaging procedure is inconsistent with the declared short-wavelength split. These issues affect the central quantitative claims rather than the formal framework.","major_comments":[{"comment":"The quantitative memory and noise kernels are all built from the plane-wave mode function (2.14), but the paper itself states after Eq. (2.17) that this is valid only for early times η0 ≫ η. In the physical regime of vacuum decay, η is near η0, where Eq. (2.16) shows a²m² = M²T0²(1 − η²/η0²) crosses zero and becomes negative; the exact Parabolic Cylinder modes (2.17) then develop the tachyonic instability that drives the decay. The plane-wave approximation removes this instability, so Eqs. (3.20)–(3.50) cannot quantitatively describe vacuum decay near η0. This is a load-bearing gap: the paper should either recompute the kernels with the exact mode functions or clearly restrict the quantitative claims to the early-time regime and adjust the title and abstract accordingly.","section":"Sec. 2.2 and Eqs. (3.20)–(3.50)"},{"comment":"The spatial average of the propagator in Eq. (3.19) uses ∫d³r e^{ik·r} = (2π)³δ³(k), so the tree-level averaged memory kernel receives only the k = 0 contribution of the σ propagator. But σ is defined as the short-wavelength/environment part of the split in Eq. (2.3), and k = 0 belongs to the long-wavelength field φ, not to the traced-out environment. A spatially homogeneous φ couples only to the zero mode of σ, so if the environment genuinely excludes k = 0 the tree-level kernel (3.20) should vanish, and the interpretation of the volume-averaged noise in Sec. 3.2 is similarly affected. The paper never specifies the k-space cutoff or the precise definition of 'short-wavelength', so the averaging in (3.12) is not equivalent to tracing out the intended environment.","section":"Sec. 3.1, Eqs. (3.12)–(3.20)"},{"comment":"The claim that a PDF P(ξ) can always be reconstructed by matching all cumulants C_n is not generally valid: an arbitrary set of multi-time cumulants need not satisfy the positivity and moment constraints required for a stochastic process to exist. The same issue affects the generalized HS identity (3.5) when the kernel K is not positive. Since the Langevin equation (3.14) presupposes the existence of the stochastic noise ξ, the paper should either prove or construct such a process, or explicitly state that the noise is a formal device whose correlation functions are defined by the influence functional.","section":"Sec. 3, Eqs. (3.8)–(3.10)"}],"minor_comments":[{"comment":"The last definition in Eq. (3.3) should read J^{nl}_Δ, not J^{nl}_c.","section":"Sec. 3, Eq. (3.3)"},{"comment":"The symmetric definition should be K_n(x_{π(1)}, ..., x_{π(n)}); the text currently repeats x_{π(1)} twice.","section":"Sec. 3, Eq. (3.7)"},{"comment":"The parameters γ and ν are not fully defined: γ = m/t0 uses an m that has not been specified for the exact solution, and it would help to distinguish the constant ~m² = a²m² from the time-dependent m².","section":"Sec. 2.2, Eq. (2.17)"},{"comment":"The sentence beginning 'For an extended Kn, the suppression factor could be less stronger' is unclear and should be rephrased.","section":"Sec. 3.2, after Eq. (3.35)"},{"comment":"There are several typos, including 'first tow terms' (should be 'first two terms') and 'It’ clear' (should be 'It is clear'), and 'the limit the limit ~m→0' (should be 'the limit ~m→0').","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The formal SK/Langevin construction is promising and within the journal's scope. The main risk is that the advertised application to vacuum decay is not supported by the quantitative kernels as currently computed; the k=0 selection problem in the spatial average is a more basic inconsistency that should be resolved before publication. I do not see a circularity or attribution problem; the paper builds on standard references and is transparent about most of its assumptions."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear X,\n\nQuick read of arXiv:2608.12736. The paper applies the Schwinger-Keldysh open-quantum-system machinery to a radiation-dominated FLRW first-order phase transition: split the scalar Φ into a spatially homogeneous mean field φ and short-wavelength σ, integrate out σ, and get a non-Markovian Langevin equation with a memory kernel and colored non-Gaussian noise. The formal structure is executed carefully. The Keldysh rotation, influence functional to second order, and generalized Hubbard-Stratonovich trick are standard, but the presentation is clear, and the explicit 1-loop kernels for a polynomial potential—Bessel/Hankel function forms after spatial averaging—are worked out in detail. The conformal-time mapping χ = aσ to flat-space theory for RD is correct, and the paper is honest that the general framework would survive using the exact mode functions.\n\nThe soft spots are real and one is load-bearing. The quantitative kernels (3.20)–(3.50) are computed with the plane-wave mode function (2.14), whose validity is stated as η0 ≫ η. But the application the title announces—vacuum decay—happens at η ≈ η0, where a²m² = M²T0²(1 − η²/η0²) changes sign. The exact mode (2.17) is a parabolic cylinder function with tachyonic instability; the plane wave has no instability, so the kernels miss exactly the physics that matters for decay. The formal derivation may survive the fix, but the numbers in Sec. 3 do not describe decays. That should be stated as a limitation, not buried.\n\nI also agree there is a tension between the pure-vacuum initial state and the thermally-corrected potential in Eq. (2.2). The m²(T) and g(T) inputs come from finite-temperature equilibrium; tracing out σ from |Ω⟩ at T = 0 while using those inputs is not coherent. It might be harmless if only the formal structure is claimed, but it is a conceptual gap.\n\nTwo smaller points: the weight function in Eq. (3.47) is not a typo—I rechecked the change of variables and it is correct. There is a likely typo in Eq. (3.3): J^{nl}_c should be J^{nl}_Δ. Minor.\n\nOverall: this is a competent reformulation of known open-EFT results in a cosmological setting, with new but regime-limited kernel expressions, not a breakthrough. The central argument that the Langevin structure is well-defined survives, but the quantitative claim that these are the kernels of cosmological vacuum decay does not. A serious referee should not desk-reject it—it deserves a round of review where the mode-function question is forced into the open, and would be a useful step with the exact modes used. I wouldn't cite it in its current form.\n\nBest","headline":"A competent but limited SK/Langevin derivation for a radiation-dominated FLRW phase transition; the formal framework is sound, but the computed kernels sit in a regime where the physics is not the vacuum decay the title points to.","tokens_in":19862,"tokens_out":3781,"would_cite":false,"duration_ms":33357,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper derives a Langevin-type equation of motion for the mean field of a phase transition in a radiation-dominated expanding universe, with the noise and memory kernels obtained from tracing out short-wavelength fluctuations.","keywords":["Schwinger-Keldysh formalism","vacuum decay","first-order phase transition","Langevin equation","non-Markovian memory kernel","non-Gaussian noise","open quantum systems","radiation-dominated FLRW universe"],"falsifier":"Compute the kernels of Eqs. (3.20)–(3.50) with the exact Parabolic Cylinder mode functions (2.17) instead of plane waves near and after the critical time $\\eta_0$; if the imaginary part of the spatially integrated propagator no longer matches the early-time sine form or develops growing contributions, the quantitative Langevin equation is not valid for vacuum decay, even though the formal framework could survive. A complementary lattice test would be to evolve the full $\\Phi$ field in the radiation-dominated background and compare the measured noise autocorrelation of the spatially averaged field with Eq. (3.50).","tokens_in":18813,"feed_emoji":"🌌","tokens_out":11957,"duration_ms":111978,"temperature":0.7,"pith_summary":"The paper sets out to describe cosmological vacuum decay—the tunnelling of an order-parameter field from a false vacuum during a first-order phase transition—as a real-time open-quantum-system problem rather than a Euclidean tunnelling calculation. It splits the phase-transition field $\\Phi$ into a spatially uniform mean field $\\phi$ and short-wavelength fluctuations $\\sigma$, traces out $\\sigma$ as an environment, and claims the resulting effective dynamics of $\\phi$ is a classical Langevin-type stochastic equation with a non-Markovian memory kernel and colored, non-Gaussian noise. The memory and noise kernels are computed explicitly for a polynomial potential in a radiation-dominated FLRW universe, using the fact that the rescaled short-wavelength modes behave like plane waves with a constant mass before the critical time. A sympathetic reader should care because the derivation offers a first-principles real-time route to the stochastic dynamics of the order parameter, with dissipation and noise both traceable to the same $\\sigma$ propagators, and it points toward numerical simulations of bubble nucleation and expansion without Euclidean bounce assumptions.","feed_headline":"Cosmic vacuum decay is a nonlocal stochastic equation","feed_subtitle":"A first-principles derivation shows the order parameter feels its own past and non-Gaussian noise.","key_machinery":"The load-bearing object is the influence functional $F[\\phi_+,\\phi_-]$ obtained by tracing over $\\sigma$, reorganized into the Keldysh basis of a classical component $\\phi_c$ and a difference component $\\phi_\\Delta$. Its quadratic terms define the retarded memory kernel $D=\\Theta(\\eta_1-\\eta_2)\\,\\mathrm{Im}\\,G$ and the noise kernel $N=\\frac12\\,\\mathrm{Re}\\,G$, where the $G$'s are connected correlators of composite operators built from powers of the $\\sigma$ propagator; a generalized Hubbard-Stratonovich transformation then trades all $\\phi_\\Delta$ polynomials for a random field $\\xi$ whose cumulants are exactly the $K_n$ of the effective action. In the radiation-dominated background, rescaling $\\chi=a\\sigma$ turns the mode equation into a flat-space form with constant mass $\\tilde{m}$, so plane-wave propagators $e^{-iE_k\\Delta\\eta}/(2E_k)$ carry the calculation; spatial averages of their powers produce the sine and Bessel/Hankel integral kernels that appear in the final equation of motion, and a finite-volume cutoff $L$ regularizes the noise correlations through spherical Bessel functions.","core_discovery":"The central claim is that the reduced dynamics of the mean field is governed by Eq. (3.14): $\\phi'' + 2H\\phi' + \\tilde{m}^2 \\phi + \\frac{a\\tilde{g}}{2}\\phi^2 + \\frac{a^2\\lambda}{6}\\phi^3 + a^2 J^\\mathrm{lin}_\\Delta{}^\\top \\int_0^\\eta d\\eta'\\, a(\\eta')^4 \\bar{D}(\\eta,\\eta') J^\\mathrm{lin}_c(\\eta') = a^2 \\bar{\\xi}(\\eta)$, with $\\tilde{m}=am$. The kernel $\\bar{D}$ is built from the imaginary part of the $\\sigma$ two-point functions, and the noise $\\bar{\\xi}$ has correlation functions built from their real parts. Because the potential of a phase-transition field necessarily contains cubic (and quartic) vertices, the noise is generically non-Gaussian; because $\\bar{D}$ is not proportional to $\\delta(\\eta-\\eta')$, the dynamics is non-Markovian. In the worked example of a radiation-dominated background with high-temperature mass, the tree-level kinetic contribution to the memory kernel vanishes (since $a''=0$), while the surviving tree-level and one-loop contributions are expressed in terms of elementary sines and integrals of Hankel functions. The formal structure survives a less aggressive scale split where $\\phi$ carries spatial dependence, giving an equation that could be simulated to follow bubble nucleation and expansion together.","pith_inferences":["If the Langevin description is right, the Euclidean bounce action should emerge as a limiting object: in the semiclassical limit, the most probable noise histories that drive $\\phi$ out of the false vacuum should be related to the bounce configuration, giving a real-time derivation of decay rates that could be compared with instanton results.","The structure $D=\\mathrm{Im}\\,G$, $N=\\frac12\\mathrm{Re}\\,G$ suggests a generalized fluctuation-dissipation relation in the expanding background; one could define an effective temperature from the ratio of noise to dissipation and check whether it tracks the Hubble temperature or the mass parameter.","The noise correlator depends on the chosen coarse-graining length $L$ (Hubble radius vs bubble radius), so predicted nucleation rates carry a scheme dependence; comparing results at different $L$ would quantify the uncertainty of the open-system reduction.","The oscillatory non-Markovian memory has period roughly $1/\\tilde{m}$, so heavy fields cannot be treated with a Markov approximation; numerical implementations will need memory-buffer algorithms, and the paper's own caveat that non-Gaussian colored noise does not give a closed local Fokker-Planck equation marks this as a genuine barrier."],"forward_implications":["If the framework is correct, the order parameter during a first-order phase transition cannot be described by a local differential equation: its acceleration at time $\\eta$ depends on the whole history of $\\phi$ through the convolution with $\\bar{D}$, so any local approximation must be justified against the memory timescale.","Non-Gaussianity of the noise is not optional: it follows from the existence of at least two vacua, because the potential then contains cubic and higher vertices, so two-point noise statistics alone are insufficient to characterise the stochastic forcing.","The same $\\sigma$ propagators determine both dissipation (imaginary part) and noise (real part), giving a concrete microscopic link between the two that could be tested by measuring the noise spectrum from simulations.","On coarse-graining scales much larger than the Hubble radius, the noise correlations are suppressed as $1/V$, so in that limit the effective dynamics is deterministic and the mean field stays trapped in the false vacuum; stochastic decay requires choosing a finite coarse-graining scale such as the Hubble radius.","Extending the split to a spatially dependent $\\phi$ yields a Langevin equation for the full field, which the paper proposes as a semi-classical tool for simulating vacuum bubble nucleation and expansion simultaneously."],"supporting_citations":[{"why":"Supplies the closed-time-path (Schwinger-Keldysh) real-time formalism on which the whole effective-action derivation rests.","marker":"[50, 51]"},{"why":"Defines the influence functional used to encode the environment's back-reaction on the mean field.","marker":"[71]"},{"why":"Prior derivation of dissipation and noise for vacuum decay in quantum field theory, the direct precursor the paper extends to a radiation-dominated FLRW background.","marker":"[57]"},{"why":"The Euclidean bounce method for vacuum decay rates, the standard approach the paper's real-time Langevin description is built to complement.","marker":"[33, 34]"},{"why":"Gives the finite-temperature effective mass $m^2 = M^2(T^2 - T_0^2)$ and $g = -AT$ that fix the polynomial potential used in the worked example.","marker":"[69]"},{"why":"Hubbard-Stratonovich transformation that converts the quadratic $\\phi_\\Delta$ term in the effective action into Gaussian noise in the Langevin equation.","marker":"[79, 80]"},{"why":"Generalizations of the Hubbard-Stratonovich trick that the paper adapts to produce non-Gaussian random fields and higher noise cumulants.","marker":"[81, 82]"},{"why":"Provides the Keldysh-basis organization of the open functional and the proof that odd and even powers of the quantum field are real and imaginary, used to write the influence functional.","marker":"[70]"}],"fun_headline_variants":["Vacuum decay is a nonlocal stochastic process","Non-Markovian noise drives cosmological vacuum decay","Schwinger-Keldysh predicts nonlocal noise for vacuum decay","Cosmic vacuum decay: memory and non-Gaussian noise","Phase transitions in FLRW are nonlocal and stochastic"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The quantitative memory and noise kernels are computed assuming the short-wavelength fluctuations oscillate like simple plane waves, which is justified only well before the critical time when the field's squared mass turns negative; close to that time the exact modes become unstable, so the computed kernels could be wrong even though the Langevin structure might survive.","fun_headline_variants_meta":{"raw":{"variants":["Vacuum decay is a nonlocal stochastic process","Non-Markovian noise drives cosmological vacuum decay","Schwinger-Keldysh predicts nonlocal noise for vacuum decay","Cosmic vacuum decay: memory and non-Gaussian noise","Phase transitions in FLRW are nonlocal and stochastic"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000362,"raw_usage":{"total_tokens":1977,"prompt_tokens":990,"completion_tokens":987,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":606,"completion_tokens_details":{"reasoning_tokens":907}},"tokens_in":606,"tokens_out":987,"duration_ms":9272,"temperature":1.0,"reasoning_tokens":907,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T00:20:31.454849+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the kernels of Eqs. (3.20)–(3.50) with the exact Parabolic Cylinder mode functions (2.17) instead of plane waves near and after the critical time $\\eta_0$; if the imaginary part of the spatially integrated propagator no longer matches the early-time sine form or develops growing contributions, the quantitative Langevin equation is not valid for vacuum decay, even though the formal framework could survive. A complementary lattice test would be to evolve the full $\\Phi$ field in the radiation-dominated background and compare the measured noise autocorrelation of the spatially averaged field with Eq. (3.50).","supporting_citations":[{"cited_title":"Dissipation, noise and vacuum decay in quantum field theory","cited_arxiv_id":"hep-ph/0101052","evidence_quote":"Prior derivation of dissipation and noise for vacuum decay in quantum field theory, the direct precursor the paper extends to a radiation-dominated FLRW background."}],"review_version":1}