{"id":"30992b31-210d-4644-a77f-000df701fe51","arxiv_id":"2602.15122","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"In the ν=1 QSSEP, noise-averaged entanglement after inhomogeneous quenches grows as (1/12)log t (domain wall) or (1/8)log t (free expansion), with exact realization-to-realization fluctuations from averaging quantum GHD over Brownian Fermi contours.","lead":"By adding space-time noise to a gas of free fermions, this paper derives exact rules for how entanglement spreads after a domain-wall or trap-release quench. The new 'stochastic quantum hydrodynamics' framework predicts diffusive spreading and halves the logarithmic entanglement growth rate compared with the noiseless case.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Eq. (11)'s Itô diffusion coefficient is a factor of 2 too large, making the stochastic GHD equation inconsistent with its own average, Eq. (12).","rationale":"The reader's weakest_assumption explicitly flags the factor-2 inconsistency between Eq. (11) and Eq. (12), and my independent check confirms it. This is the most concrete and load-bearing concern because the stochastic GHD equation is the formal foundation of the Fermi-contour ensemble used in Eq. (25). The QGHD input itself is supported by the deterministic complex-hopping analysis in Appendix A and by the numerical comparisons, so the strongest challenge is not the imported QGHD machinery but the incorrect Itô coefficient in the paper's own derivation. The verdict should remain CONDITIONAL: the central claim is well-supported numerically and likely correct, but the manuscript must correct the SDE and clarify the derivation before the claimed formal extension of QGHD to stochastic dynamics is fully established.","tokens_in":31346,"tokens_out":17307,"duration_ms":159852,"concrete_test":"Re-derive Eq. (11) from Eq. (9) using Itô's lemma with dξ_k^2=2D dt. If the deterministic term is D ∂x^2 n_k, confirm that Eq. (11) contains a factor-2 typo. Then, using the corrected SDE, derive the Fokker-Planck equation for the probability density of n_k and check that its mean reproduces the exact lattice density profile (19) in the diffusive limit. If the corrected SDE fails to produce the right contour distribution, the derivation of the central claim is unsound; if it succeeds, the paper requires only a typo fix.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim (25) rests on a stochastic GHD description whose fundamental evolution equation is stated incorrectly. From Eq. (9), n_k(x,t)=n_k(x-ξ_k(t),0). With dξ_k=2√D(cos k dB1 + sin k dB2) and dB^2=dt/2, one has dξ_k^2=2D dt. Itô's lemma gives d n_k = -∂x n_k dξ_k + (1/2)∂x^2 n_k dξ_k^2 = -∂x n_k dξ_k + D ∂x^2 n_k dt. The manuscript instead writes 2D ∂x^2 n_k dt in Eq. (11). Consequently, averaging Eq. (11) would yield ∂t⟨n_k⟩=2D ∂x^2⟨n_k⟩, not the advertised Eq. (12), ∂t⟨n_k⟩=D ∂x^2⟨n_k⟩. The later density profile (15) and the exact lattice result (19) confirm the coefficient is D, so the displayed SDE is internally inconsistent with the rest of the paper. This is not a cosmetic typo: Eq. (11) is advertised as the derived stochastic evolution of the occupation function, and the Fermi-contour ensemble (28) is supposed to be the collection of its single-realization solutions. As written, the derivation of the contour ensemble—and hence of Eq. (25)—cannot be reproduced. The numerics may survive a simple erratum, but the formal route to the central claim contains a concrete, correctable error.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the ν=1 QSSEP, a free-fermion chain with spatially uniform stochastic complex hopping, starting from either a domain-wall state or the ground state of a trapped gas. The central proposal is that, in the hydrodynamic limit, every noise realization is characterized by a stochastically translated Fermi contour, and that the per-realization Rényi/von Neumann entanglement entropy is exactly given by the deterministic QGHD/CFT twist-field formula evaluated on that contour. Averaging over the exactly known Brownian distribution of contours yields Eq. (25), from which the authors derive explicit asymptotic predictions: half-system entanglement grows as (1/12) log t for domain-wall melting and as (1/8) log t for free expansion, with relative fluctuations vanishing at late times. These predictions are compared with exact numerics for systems up to L=160–400 over about 10^3 noise realizations, with good agreement in the bulk/hydrodynamic regime. The paper is clearly written and the numerical checks are extensive.","tokens_in":31682,"tokens_out":9514,"duration_ms":98033,"significance":"If the central claim holds, this is the first extension of quantum generalized hydrodynamics to stochastic quantum dynamics, and it provides an exact, parameter-free description of diffusive entanglement growth and its fluctuations. The paper has notable strengths: the contour ensemble is derived from the Brownian statistics of the noise, the density propagator is checked against the exact lattice result Eq. (19), the single-realization QGHD formulas are independently benchmarked in Appendix A against deterministic complex-hopping evolution, and all final predictions are compared to exact numerics with no fitted parameters. The main formal problem is the factor-of-two inconsistency in the displayed stochastic evolution equation, Eq. (11); this is a concrete, correctable error but it sits at the base of the advertised derivation, so it must be fixed before the paper can be accepted in its present form.","major_comments":[{"comment":"The displayed Itô equation has the wrong diffusion coefficient. From the definition dξ_k = 2√D(cos k dB_1 + sin k dB_2) with dB_i^2 = dt/2, one obtains dξ_k^2 = 2D dt. Itô's lemma applied to n_k(x,t) = n_k(x−ξ_k(t),0) gives dn_k = −n'_k dξ_k + (1/2)n''_k dξ_k^2 = −n'_k dξ_k + D n''_k dt, not 2D n''_k dt. As written, averaging Eq. (11) would yield ∂_t⟨n_k⟩ = 2D ∂_x^2⟨n_k⟩, contradicting Eq. (12), the density profile Eq. (15), and the exact lattice propagator Eq. (19), whose hydrodynamic limit is the D-diffusion kernel. Since Eq. (11) is advertised as the derived stochastic evolution of the occupation function and Fig. 1 explicitly refers to it, the derivation as displayed cannot be reproduced. The error is local and correctable: replace 2D by D and re-check the Itô computation.","section":"Sec. 3, Eq. (11)"},{"comment":"The Itô computation is also formal for the domain-wall initial condition, because n_k(x,0) = θ(−x) is not differentiable, so ∂_x and ∂_x^2 appearing in Eq. (11) act on a step function. This is separate from the factor-of-two issue. The authors should state the regularized version of the argument—for example, start from a smooth initial contour as in the free-expansion setup with finite β and then take the hydrodynamic/β→∞ limit, or formulate Eq. (11) weakly. This regularity point does not invalidate the numerics, but it is needed if Eq. (11) is to serve as the rigorous basis for the stochastic GHD description.","section":"Sec. 3, Eqs. (9)–(11)"}],"minor_comments":[{"comment":"The term e^{−ℓ^2/(4t)}/(12) log t is typographically ambiguous; it should be written as (e^{−ℓ^2/(4t)}/12) log t to avoid being read as e^{−ℓ^2/(4t)}/(12 log t).","section":"Eq. (35)"},{"comment":"The expression 'Υ + log 2 / 3' is ambiguous; please write (Υ+log 2)/3 or Υ+(log 2)/3 explicitly as intended, and check the resulting numerical constant κ1.","section":"Eqs. (31), (53)"},{"comment":"The authors correctly state that the boundary-induced entanglement growth near x = −L/2 is not captured by the hydrodynamic framework. This is an important scope limitation; it should be stated in the conclusions and abstract so that the claimed agreement is understood to hold in the bulk/hydrodynamic regime only.","section":"Sec. 4, Fig. 6"}],"recommendation":"major_revision","confidential_remarks":"The main technical issue is the factor-of-two error in Eq. (11); it is correctable but should be treated as a real inconsistency in the derivation, not merely a typo, since the displayed equation contradicts its own average. The paper relies substantially on the companion preprints [23] and [46]; I recommend that the editor confirm these are available and that the relevant results are stated with sufficient detail. The paper is within the scope of the journal and the numerical verification is convincing once the formal issue is fixed."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a real advance for the non-equilibrium free-fermion subfield. The paper extends quantum generalized hydrodynamics to the ν=1 QSSEP, where each noise realization is a deterministic complex-hopping evolution, and averages the QGHD per-realization entanglement entropy over the Brownian Fermi-contour ensemble. The new results are the exact noise-averaged entanglement growth — (1/12) log t for domain-wall melting, (1/8) log t for free expansion — plus the variance asymptotics and the deterministic complex-hopping formula in Appendix A. The per-realization entropy formula itself is borrowed from the known deterministic results, so the genuinely new content is the stochastic contour ensemble and the averaging. That is worth saying plainly, not as a criticism: the averaging is the paper's contribution.\n\nWhat I like: the central predictions are checked against exact numerics with no free parameters, the density diffusion coefficient is pinned down by an independent exact lattice computation, and the complex-hopping appendix gives a separate consistency check. The paper is honest about boundary effects in the free-expansion protocol and about the intermediate-time corrections. The citation pattern leans heavily on the authors' own companion papers, but the cited results are directly relevant and the circularity burden is low.\n\nSoft spots, in order. First, Eq. (11) is wrong as written: from Eq. (9) and the stated Brownian increments, Itô gives D ∂²_x n_k dt, not 2D ∂²_x n_k dt. Averaging the displayed equation would give a diffusion equation with coefficient 2D, contradicting Eq. (12) and the exact lattice result. The rest of the derivation does not actually use Eq. (11) as an independent evolution equation—the contour ensemble comes from Eq. (9)—so this is a correctable typo in a central displayed equation, not a fatal flaw. But it is the kind of thing that must be fixed before publication. Second, the paper assumes the deterministic QGHD/Fisher-Hartwig per-realization formula remains valid when the Fermi contour is random; it validates this numerically but does not derive it. That is a genuine gap, though a standard one in this literature. Third, the abstract's \"full entanglement statistics\" overstates what is explicitly shown: the paper computes the mean and variance, and the method in principle gives all moments, but \"full\" is too strong. These are all fixable; none of them shakes the main result.\n\nMy take: this paper deserves a serious referee. The core claim is well supported, the mathematical structure is clear, and the one concrete internal inconsistency is localized. Send it to peer review with instructions to check the Itô factor carefully and to require an explicit statement about the validity conditions of the imported QGHD formula for stochastic contours.","headline":"A solid extension of QGHD to stochastic free fermions, with clean numerics and one central typo (Eq. 11 has a factor-2 Itô diffusion coefficient) that needs fixing before publication.","tokens_in":806,"tokens_out":2368,"would_cite":true,"duration_ms":91224,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Averaging the per-realization QGHD entropy over Brownian Fermi contours gives the exact noise-averaged entanglement: growth is (1/12) log t for a melting domain wall and (1/8) log t for free expansion.","keywords":["quantum symmetric simple exclusion process","stochastic free fermions","quantum generalized hydrodynamics","entanglement entropy","Fermi contour","domain-wall melting","free expansion","diffusive transport"],"falsifier":"Extract the late-time slope of the half-system entanglement for the domain wall: if ⟨S₀(t)⟩ − C does not approach (1/12) log t as both L and t grow with t ≪ L², the central claim fails. Equivalently, a single fixed noise realization compared to the per-realization formula would expose any missing stochastic correction beyond the QGHD prediction.","tokens_in":31232,"feed_emoji":"⚛️","tokens_out":7754,"duration_ms":80182,"temperature":0.7,"pith_summary":"This paper claims that a one-dimensional chain of free fermions with spatially uniform but time-random hopping amplitudes—the ν=1 quantum symmetric simple exclusion process—can be described at long times by a stochastic version of generalized hydrodynamics. Starting from a domain wall or a trapped gas, the coarse-grained occupation function obeys a diffusion equation after noise averaging, while each individual noise realization is a ballistic front with a random displacement. The paper then argues that the full, noise-averaged entanglement statistics are obtained exactly by averaging the quantum generalized hydrodynamics (twist-field) prediction for each realization's Fermi contour. If correct, this gives an exact account of diffusive entanglement growth, with the half-system entropy growing as (1/12) log t for domain-wall melting and (1/8) log t for free expansion, and with relative fluctuations that vanish at late times. It is the first extension of quantum generalized hydrodynamics to stochastic dynamics.","feed_headline":"Noisy fermion chain: entanglement grows as (1/12) log t","feed_subtitle":"Full statistics follow from averaging per-realization CFT predictions; free expansion gives (1/8) log t.","key_machinery":"The Brownian Fermi-contour ensemble is the central object: each noise realization moves the initial contour x0(k) to x0(k) + 2ρ sin(k + φ), with ρ Rayleigh-distributed and φ uniform. The contour's intersections with the subsystem cut are the Fermi points at which chiral twist fields are inserted; the twist-field correlation, together with a Fisher–Hartwig non-universal constant, supplies the single-realization entropy, and the ensemble average over ρ and φ reproduces the noise-averaged Rényi and von Neumann entropies.","core_discovery":"In the hydrodynamic limit, the noise-averaged Rényi entropy equals the average of the single-realization conformal-field-theory entropy over the Brownian Fermi-contour ensemble. For the domain wall, each realization is labeled by a Rayleigh-distributed radius ρ, and its von Neumann entropy is S_ℓ(ρ) = (1/6) log[2ρ(1 − (ℓ/2ρ)^2)^{3/2}] + κ₁ for ℓ < 2ρ, the same formula as the deterministic model with the replacement 2ρ → t. Averaging gives ⟨S_ℓ(t)⟩ ≃ e^{−ℓ²/(4t)}/12 log t + S(ℓ/√t), which reduces to (1/12) log t + const at ℓ = 0. For free expansion, an additional uniform phase φ appears, split Fermi seas with four Fermi points can arise, and the hard-wall limit gives (1/8) log t + const. The","pith_inferences":["Beyond the paper itself, because each per-realization entropy depends only on ρ and φ, the late-time distribution of entanglement is a one- or two-parameter distribution; measuring its skewness or third cumulant would test the mechanism more sharply than the mean does.","The boundary effect near x = −L/2, which the paper explicitly flags as outside its hydrodynamic description, could plausibly be incorporated by regularizing the Fermi contour at the hard wall and including elastic reflection—an extension the paper suggests but does not derive.","The same contour-averaging prescription should carry over to symmetry-resolved entropies and entanglement asymmetry, which are listed as future work: if the per-realization contour fixes the entanglement spectrum, those quantities follow from the identical ensemble.","For ν > 1, where the noise is spatially periodic rather than homogeneous, the contour will no longer be a rigid translation, so the exact factorization into (ρ, φ) will break; the present result therefore serves as a benchmark rather than a full solution of the general ν-QSSEP family."],"forward_implications":["Average density profiles are exactly diffusive: the domain-wall front becomes an error-function profile with width √(Dt), so transport is not ballistic on average.","Half-system entanglement grows logarithmically with half the deterministic slope: (1/12) log t for domain-wall melting and (1/8) log t for free expansion, instead of (1/6) and (1/4) in the clean model.","Entanglement is self-averaging in the hydrodynamic limit: the relative standard deviation of the half-system entropy vanishes as t → ∞, so the average equals the typical value.","The single-realization formulas also describe deterministic dynamics with a fixed complex hopping phase φ; the stochastic result arises from averaging those fixed-phase evolutions.","The same averaging procedure gives all higher moments and all Rényi indices, not only the von Neumann entropy."],"fun_headline_variants":["Quantum GHD goes stochastic: exact entanglement statistics","Stochastic fermions: entanglement entropy from CFT per realization","Domain-wall melting in noisy fermions: (1/12) log t entanglement","First QGHD for stochastic systems: entanglement averaged exactly","Noisy quantum transport: entanglement statistics from single-realization CFT"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The load-bearing premise is that averaging the deterministic twist-field formula over the Brownian Fermi contour reproduces the noise-averaged entanglement entropy, with no additional stochastic corrections; a secondary fragile step is the Itô differentiation of the step-like occupation function, where a factor-of-two inconsistency appears.","fun_headline_variants_meta":{"raw":{"variants":["Quantum GHD goes stochastic: exact entanglement statistics","Stochastic fermions: entanglement entropy from CFT per realization","Domain-wall melting in noisy fermions: (1/12) log t entanglement","First QGHD for stochastic systems: entanglement averaged exactly","Noisy quantum transport: entanglement statistics from single-realization CFT"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000605,"raw_usage":{"total_tokens":2680,"prompt_tokens":786,"completion_tokens":1894,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":530,"completion_tokens_details":{"reasoning_tokens":1807}},"tokens_in":530,"tokens_out":1894,"duration_ms":13393,"temperature":1.0,"reasoning_tokens":1807,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-02T22:57:30.326385+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Extract the late-time slope of the half-system entanglement for the domain wall: if ⟨S₀(t)⟩ − C does not approach (1/12) log t as both L and t grow with t ≪ L², the central claim fails. Equivalently, a single fixed noise realization compared to the per-realization formula would expose any missing stochastic correction beyond the QGHD prediction.","supporting_citations":[],"review_version":1}