{"id":"7d2cb3bc-782e-483d-a3c5-eadf80a2ea33","arxiv_id":"2509.00545","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":3.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"First simulations of model H stochastic fluid dynamics near a critical point give a dynamic critical exponent z ≈ 3.01 and viscosity renormalization.","lead":"This paper summarizes the first simulations of model H, the stochastic fluid theory expected to describe QCD matter near a critical endpoint. A generalist should read it because it shows a practical way to simulate fluctuating fluids and extracts the dynamic critical exponent z ≈ 3.01.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Missing discretization-error/conservation-law validation for the advection step leaves the z≈3.01 claim under-determined.","rationale":"The paper's central claim is the first successful simulation of model H and a dynamic critical exponent z ≈ 3.01. The load-bearing condition is that the discrete algorithm used in Section 3 actually realizes continuum model H. The authors note the risk themselves: the advection step may violate conservation laws in the continuum limit. No convergence test or alternative discretization comparison is provided, and the quoted z has no error bars. This is precisely the reader's weakest assumption. It is a missing-evidence concern, not an internal inconsistency; the underlying published work (Refs. [4,5]) may well support the claim. I recommend a single computational check: compare z from two advection discretizations and two lattice sizes. Until such a check is reported, the standalone paper is insufficiently validated, so the existing CONDITIONAL verdict is appropriate.","tokens_in":3503,"tokens_out":9821,"duration_ms":130518,"concrete_test":"Run the full model-H simulation at the critical point with the current skew-symmetric advection discretization and with a standard second-order central-difference advection discretization, for lattice sizes L=24 and L=48, keeping all other parameters fixed. Extract the dynamic critical exponent z from the finite-size scaling of the order-parameter autocorrelation time. If the two schemes give z values differing by more than 0.05, or if z_eff shifts by more than 0.05 between L=24 and L=48 within either scheme, then the reported z≈3.01 is not a robust continuum-model-H result.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 3 describes the numerical algorithm but gives no validation that the advection discretization ('skew-symmetrized derivatives', Ref. [9]) plus the transverse projection preserve the conservation laws and fluctuation-dissipation balance needed for continuum model H. The authors explicitly acknowledge (Section 3): 'we may encounter large violations of conservation laws that follow from the ideal equations of motion in the continuum limit.' This is the step on which the central claim z ≈ 3.01 rests: if the advection term injects or removes order parameter or momentum at the lattice scale, the measured relaxation time τ(L) and the extracted z can be biased. The paper reports no lattice-spacing dependence, no comparison between advection schemes, and no conservation-law residual plots, and the z value is quoted without error bars. The claim that z is 'clearly different from mean-field behavior' is therefore not backed by evidence that the simulated model is continuum model H.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports simulations of real-time stochastic fluid dynamics in the universality class of model H, motivated by QCD critical-point phenomenology. The authors describe a numerical scheme combining conserving Metropolis updates for the dissipative and noise parts of the equations of motion with a skew-symmetrized derivative discretization for the advective terms. They report two main results: (i) a UV-sensitive renormalization of the shear viscosity, dominated by self-advection and consistent with the 'stickiness of sound' effect, and (ii) a dynamic critical exponent z ≃ 3.01 extracted from finite-volume scaling directly at the critical point, which they claim is clearly different from mean-field behavior and consistent with the epsilon expansion. They also observe a crossover from z ≃ 4 to z ≃ 3 as the physical viscosity is reduced and evidence for universality between model H and a truncated 'model H0'.","tokens_in":3746,"tokens_out":4096,"duration_ms":49790,"significance":"If the numerical method can indeed simulate the stochastic equations of model H and reproduce the expected universal dynamic scaling, this is a substantial step toward first-principles modeling of critical fluctuations in heavy-ion collisions and other fluid systems near a critical point. The paper explicitly identifies a challenging combination of issues—large thermal fluctuations on all scales, renormalization of UV divergences, and the need to preserve conservation laws and fluctuation-dissipation relations—and the proposed conserving Metropolis update is an interesting and potentially powerful idea. However, the quantitative claims made here (z ≈ 3.01, viscosity renormalization, crossover, Ising static exponents) are not actually demonstrated within this manuscript; the text repeatedly refers to the authors' previous papers [4,5] for the supporting analysis. As a standalone contribution, the paper therefore lacks the evidence needed to substantiate its central numerical results, even though those results may be correct.","major_comments":[{"comment":"The central quantitative claim is 'We observe a value z≃3.01 which is clearly different from mean-field behavior', but no error bars, scaling-fit details, system sizes, or lattice-spacing dependence are provided. The right panel of Fig. 2 shows a curve labeled 'model H' but with no uncertainty and no description of how z_eff is extracted from the order-parameter correlation function. Since the entire paper's significance rests on this exponent, the result is under-determined in this manuscript. The pointer to Refs. [4,5] is not a substitute, because the present paper makes a standalone claim; please either include the scaling analysis or explicitly state that this is a summary and rephrase the claim as a report of previous work.","section":"§4, Fig. 2"},{"comment":"The discretization of the advective terms is the step on which the model-H universality claim depends, but the manuscript gives no validation that the skew-symmetrized derivative scheme combined with the Metropolis update and transverse projection preserves conservation laws and fluctuation-dissipation balance at finite lattice spacing. The text itself acknowledges 'we may encounter large violations of conservation laws that follow from the ideal equations of motion in the continuum limit'. No convergence study with lattice spacing, no comparison between different advection discretizations, and no conservation-law residual plots are shown. Without such tests, the measured relaxation times and the extracted z could be biased by lattice-scale violations, so the identification of the simulated model with continuum model H is not established.","section":"§3"},{"comment":"The statement 'We have verified that m^2 can be tuned to a critical point at which the simulation reproduces the static critical exponents of the 3d Ising Model' is made without any supporting data, method, or reference to which figure shows this verification. The static critical point provides the physical condition under which the dynamic exponent is measured; if the static tuning is not shown, the dynamic scaling claim lacks its foundation. Please provide the static-exponent determination or specify that it is taken from Ref. [4] and give the relevant quantities.","section":"§4"},{"comment":"The crossover from z_eff ≈ 4 to z_eff ≈ 3 as a function of physical viscosity is described only qualitatively. The right panel of Fig. 2 is unclear: the definition of z_eff, the observable from which it is extracted, the system size, and the number of independent runs are not given. The curve labeled 'Kawasaki approximation' is not defined or referenced. Because this crossover is used to argue for universality, it needs a precise definition and an uncertainty estimate, or it should be explicitly deferred to a forthcoming/previous detailed paper.","section":"§4, Fig. 2"}],"minor_comments":[{"comment":"The index structure of the last term in Eq. (2) is difficult to parse: the expression 'πT_j δH/δπT_k' is not explicitly summed over j and k in the displayed formula, and the overall index of the term is not clear. Please clarify the notation.","section":"§2, Eq. (2)"},{"comment":"The sentence 'We follow the same procedure for ⃗π. We perform a conserving Metropolis update for all components of the momentum density, and then apply a transverse projection operator.' leaves open the question of whether the transverse projection is part of the Markov chain or a separate step applied to the output; this affects the conservation properties. Please state the order of operations and the effect on the momentum components.","section":"§3"},{"comment":"'Model H0' is used in the figure caption and the text but is never defined in the body. Please define it explicitly, stating which terms of Eq. (1)-(2) are dropped.","section":"Fig. 2"},{"comment":"Reference [3] is a arXiv preprint without a journal or DOI; if possible, update to the published version or include the arXiv identifier consistently. Also, the reference to 'Kawasaki approximation' in Fig. 2 has no citation; either add the appropriate reference or remove the label.","section":"References"},{"comment":"The sentence 'We have demonstrated that we can also measure more complicated observables, such as the relaxation rate of higher moments of the order parameter.' is not accompanied by a figure, equation, or data. Either present an example or delete the sentence, as it currently serves only as a placeholder.","section":"§4"}],"recommendation":"major_revision","confidential_remarks":"This manuscript is essentially a 4-page summary of the authors' own previous publications [4,5]. The central numerical results—especially z ≈ 3.01—are asserted with no supporting data inside the paper, and the text explicitly says it 'summarize[s] the results obtained in our recent work'. If the journal allows such proceedings-style contributions, then a major revision should require the authors to either (a) include enough quantitative detail to make the results verifiable, or (b) clearly reframe the paper as a summary and temper the claims accordingly. If the journal expects a self-contained research paper, the current scope is insufficient for publication. There is no indication of misconduct; the issue is purely about the amount of evidence presented. I recommend a major revision rather than rejection because the underlying approach may be sound and the missing validation could in principle be added from the authors' longer papers."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The first thing you should know: this is a short conference-style summary of the authors' own already-published results (PRL 133, 032301 and PRD 111, 034026). The text says so explicitly. So if you read it expecting new data, you'll be disappointed. But as a summary it's well written: the model H equations are cleanly displayed, the Metropolis-based algorithm is sketched, and the main physical claims—viscosity renormalization, the dynamic exponent z≈3.01, and the finite-size crossover from z≈4 to z≈3—are stated without obfuscation. The figures are illustrative rather than quantitative.\n\nThe underlying accomplishment is genuinely significant: first simulation of model H, if the cited work is correct. There's no obvious reason to doubt it—the results are consistent with epsilon-expansion expectations and the static exponents are checked against the 3D Ising model. The paper gets credit for being upfront about its own scope.\n\nThe soft spot is also real. The stress-test note points to Section 3, where the authors acknowledge that the advection discretization can produce large conservation-law violations, and then say only that they used skew-symmetrized derivatives from Ref. [9]. In this manuscript there is no convergence test, no conservation-law residual, no lattice-spacing dependence, and the quoted z≈3.01 has no error bar. So as a standalone paper, the central quantitative claim is under-determined. But here's the catch: this is a summary, not the research paper. The actual evidence is presumably in Refs [4,5], which went through peer review. If you judge this as a proceedings contribution, it's perfectly acceptable. If it's meant to be a citable primary source, then no—but it doesn't present itself that way.\n\nI agree with the reader's conditional verdict. The paper is not wrong on its face; it's just thin. The authors could easily improve it by adding a few lines of scaling analysis or pointing to verification plots in the earlier work. As it stands, the useful takeaway is 'read Refs [4,5].' For a reader who wants a quick orientation to what model H simulation is about, this is a decent entry point. It's not a paper I would cite for the z value itself, and I wouldn't give it to a student as the primary source. But it deserves a serious referee in the sense that the underlying work is important and the summary is honest. My recommendation: if this appears in proceedings, accept as-is; if submitted as a regular article, ask the authors to redelegate the technical evidence to the cited papers or include at least one scaling plot with error bars.","headline":"A clear, honest summary of the authors' own prior model H simulations, but the quantitative evidence lives elsewhere—treat it as a briefing, not a self-contained proof.","tokens_in":816,"tokens_out":702,"would_cite":false,"duration_ms":35202,"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":"This paper reports the first numerical simulation of model H and measures the dynamic critical exponent z ≈ 3.01 at the critical point.","keywords":["model H","stochastic fluid dynamics","dynamic critical exponent","QCD critical point","critical slowing down","Metropolis algorithm","viscosity renormalization","finite-size scaling"],"falsifier":"Run the finite-size scaling measurement at several different lattice spacings: if the extracted z shifts systematically away from ≈3.01, or if global momentum conservation is violated as resolution increases, the central claim fails.","tokens_in":3431,"feed_emoji":"🌊","tokens_out":5817,"duration_ms":63412,"temperature":0.7,"pith_summary":"Model H is the stochastic hydrodynamic theory expected to govern long-wavelength fluctuations near a liquid-gas-type critical endpoint, including a possible critical point in the QCD phase diagram. Until now it had never been simulated numerically, because fluctuations on all scales make renormalization and fluctuation-dissipation balance hard to preserve on a lattice. This paper presents a discretization scheme built on a conserving Metropolis update with transverse projection and skew-symmetrized derivatives, and simulates the real-time dynamics of the order parameter and fluid momentum. At the critical point the measured dynamic critical exponent is z ≈ 3.01, clearly different from mean-field behavior and consistent with epsilon-expansion predictions. The simulations also reproduce the predicted renormalization of shear viscosity, the 'stickiness of sound,' and show a finite-size crossover from z ≈ 4 to z ≈ 3.","feed_headline":"Simulating model H yields critical exponent z ≈ 3.01","feed_subtitle":"Stochastic fluid simulations near a critical endpoint match epsilon-expansion predictions for critical slowing down.","key_machinery":"The conserving Metropolis update is the central mechanism: a trial update moves a charge between neighboring lattice sites, accepted with probability min(1, e^{-ΔH/T}). Its first moment realizes the diffusion equation and its second moment reproduces the noise correlation, so fluctuation-dissipation balance is built in. Transverse projection removes the longitudinal component of momentum, and skew-symmetrized derivatives keep the advection step conservative despite large sub-lattice fluctuations.","core_discovery":"Model H—stochastic hydrodynamics for an order parameter coupled to transverse momentum—has not previously been simulated, and this paper reports simulations of it in two and three dimensions. The order parameter is the specific entropy, the momentum is projected transverse, and the dynamics are generated by a conserving Metropolis update whose first and second moments reproduce the diffusion and noise terms of the continuum equations. Tuning the mass parameter to the critical point reproduces the static 3d Ising exponents; the finite-size scaling of the order-parameter correlation time then gives a dynamical critical exponent z ≈ 3.01, consistent with epsilon-expansion predictions and clearl","pith_inferences":["Editorial extension: the Metropolis-plus-projection construction should generalize to other stochastic hydrodynamic theories with different conserved charges, but the paper does not claim this.","Editorial extension: because the paper flags possible violations of ideal conservation laws and does not show a lattice-spacing convergence study, the continuum limit of z ≈ 3.01 remains an open check.","Editorial extension: once the advertised mapping to QCD parameters is completed, z ≈ 3 implies critical slowing-down stronger than mean-field estimates in a heavy-ion context; this translation is not made in the paper."],"forward_implications":["If z ≈ 3.01 is correct, the relaxation time near a QCD critical point grows as ξ^3 rather than ξ^2, changing predicted fluctuation observables in heavy-ion collisions.","The algorithm lifts a longstanding barrier: model H can now be studied quantitatively in real time, not just through analytic approximations.","The observed finite-size crossover from model-B-like z ≈ 4 to model-H z ≈ 3 implies that in a finite fireball the effective exponent depends on viscosity and system size.","Reproducing the 'stickiness of sound' viscosity renormalization means the lattice scheme captures non-perturbative momentum transport, not just order-parameter diffusion.","Embedding the framework into an expanding relativistic fluid is the stated next step; the current results provide the benchmark dynamics to embed."],"supporting_citations":[{"why":"Defines the model H equations and the dynamic universality classes; this is the theory being simulated.","marker":"[1]"},{"why":"Identifies the QCD critical point with the model H universality class, supplying the physics motivation.","marker":"[2]"},{"why":"Earlier work by the same authors presenting the simulation method and first results; this paper summarizes and extends it.","marker":"[4]"},{"why":"Companion paper giving the two- and three-dimensional critical dynamics results, including the z measurement details.","marker":"[5]"},{"why":"Justifies identifying the order parameter with specific entropy near a liquid-gas endpoint, needed for physical interpretation.","marker":"[6]"},{"why":"Provides the observation that a conserving Metropolis update's first and second moments reproduce diffusion and noise; foundation of the algorithm.","marker":"[8]"},{"why":"Supplies the skew-symmetrized derivative discretization used to keep advection conservative; load-bearing for the numerical scheme.","marker":"[9]"},{"why":"Predicted the UV-sensitive viscosity renormalization inversely proportional to the bare viscosity ('stickiness of sound') that the simulation reproduces.","marker":"[10]"}],"fun_headline_variants":["First model H simulation gives z≈3.01","Stochastic fluid simulations nail critical slowing down","Model H simulated for first time, z≈3.01 found","First simulations of model H yield z≈3.01","Critical exponent z≈3.01 from first model H runs"],"cache_read_input_tokens":2688,"weakest_assumption_plain":"The lattice discretization is assumed to faithfully represent the continuum model H equations, preserving conservation laws and fluctuation-dissipation balance; the paper does not show a convergence or conservation-law validation.","fun_headline_variants_meta":{"raw":{"variants":["First model H simulation gives z≈3.01","Stochastic fluid simulations nail critical slowing down","Model H simulated for first time, z≈3.01 found","First simulations of model H yield z≈3.01","Critical exponent z≈3.01 from first model H runs"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000738,"raw_usage":{"total_tokens":3029,"prompt_tokens":538,"completion_tokens":2491,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":282,"completion_tokens_details":{"reasoning_tokens":2411}},"tokens_in":282,"tokens_out":2491,"duration_ms":18351,"temperature":1.0,"reasoning_tokens":2411,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T13:28:00.488539+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the finite-size scaling measurement at several different lattice spacings: if the extracted z shifts systematically away from ≈3.01, or if global momentum conservation is violated as resolution increases, the central claim fails.","supporting_citations":[{"cited_title":"Transits of the QCD Critical Point","cited_arxiv_id":"1811.05081","evidence_quote":"Justifies identifying the order parameter with specific entropy near a liquid-gas endpoint, needed for physical interpretation."},{"cited_title":"Morinishi, T","cited_arxiv_id":null,"evidence_quote":"Supplies the skew-symmetrized derivative discretization used to keep advection conservative; load-bearing for the numerical scheme."}],"review_version":1}