{"id":"5623f633-3d5a-4aaa-be93-04f4aabaf689","arxiv_id":"2507.01959","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Non-Markovian dissipation together with conserved angular momentum stabilizes long-range order in one- and two-dimensional nonequilibrium O(N) models.","lead":"This paper finds that in low-dimensional nonequilibrium O(N) models, non-Markovian dissipation alone cannot create long-range order, but combining it with conserved angular momentum can suppress fluctuations enough to stabilize that order. The result offers a new generic mechanism to evade the Mermin-Wagner theorem in driven systems.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Stability of the Gaussian fixed point against the omitted amplitude and L_alpha_beta sectors is asserted, not demonstrated; the central exponent can be trusted only after a one-loop or numerical check.","rationale":"The paper's strongest claim is the exponent formula and its consequence. The derivation has two pillars: (i) the Gaussian scaling analysis of the linearized equations, and (ii) the statement that nonlinearities are irrelevant. Pillar (i) is well supported by the numerical solution of the linearized equations (App. E). Pillar (ii) is the soft spot. The nonlinearity check in Table II only computes the tree-level scaling dimension of two vertices; it does not establish that one-loop corrections from the truncated sectors (sigma and L_alpha_beta) are harmless. The reader's weakest_assumption points to the same area: the two-field truncation is assumed complete. I partially agree, but I would sharpen it: the issue is not merely that the modes are omitted, but that the paper assigns scaling dimensions to L_alpha_beta by analogy rather than by solving (5d), and a direct Gaussian calculation suggests those dimensions are wrong. If the dimensions are wrong, the power-counting of the irrelevant operators is not trustworthy. The proposed concrete test—a one-loop RG or a direct numerical simulation of the full nonlinear model—would settle whether the Gaussian fixed point is actually the IR limit. Until then, the correct verdict is CONDITIONAL, which matches the reader's assessment.","tokens_in":28525,"tokens_out":57584,"duration_ms":632671,"concrete_test":"Compute the one-loop self-energy corrections to the phi and ell propagators in the full MSR action (6,8), including the amplitude mode sigma and all L_alpha_beta modes, using the scaling dimensions of Table I. If any correction to the inverse phi correlation at small q is more singular than q^{2 alpha_n/(alpha_n+alpha_L)}, or if a relevant operator (e.g., a q^2 term with positive scaling dimension) is generated, the Gaussian fixed point is unstable. As a complementary check, numerically integrate the full nonlinear Langevin equations (1) in d=2 with alpha_n=alpha_L=1/2 and measure the order parameter <n>^2 versus system size; absence of a nonzero extrapolation would indicate the truncation misses destabilizing modes.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—eta_phi=2 alpha_n/(alpha_n+alpha_L), and hence LRO for eta_phi<d—is derived from the Gaussian fixed point of the two-field sector (phi, ell). The argument requires that all other fluctuations (the amplitude mode sigma and the conserved densities L_alpha_beta with alpha,beta>1) do not alter this fixed point. The paper justifies this only by tree-level power counting of two vertices (Eq. 8 and Table II); it does not compute one-loop renormalization of the bilinear sector, nor does it integrate out sigma and L_alpha_beta. Moreover, the scaling dimensions assigned to L_alpha_beta are asserted to equal those of ell, but L_alpha_beta obeys the decoupled linear equation (5d); a direct Gaussian calculation from (5d) for sub-Ohmic gamma_L gives C_LL(q) ~ const, implying [L_alpha_beta]=d/2, not [ell]=a+d/2. This discrepancy suggests the power-counting used to declare the nonlinearities irrelevant may not be reliable. If any omitted mode generates a relevant operator at one loop, the exponent eta_phi would flow away from the Gaussian value and the LRO prediction would fail. Since the paper's verification is limited to the linearized equations (App. E), the nonlinear stability of the fixed point is the weakest load-bearing assumption.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a new mechanism for circumventing the Mermin-Wagner theorem in low-dimensional driven O(N) models: the combination of non-Markovian (power-law) dissipation with conserved angular momentum. The authors analyze the linearized Langevin equations for the Goldstone modes and the coupled conserved density, derive by Gaussian scaling the exponent eta_phi = 2 alpha_n/(alpha_n+alpha_L) for the equal-time correlation C_phi_phi(q) ~ q^{-eta_phi}, and conclude that eta_phi<2 gives long-range order in d=2, while eta_phi<1 gives order in d=1 for alpha_n<alpha_L. The phase diagram in Fig. 2 summarizes these regimes. The paper also argues by tree-level power counting that the leading nonlinearities do not destabilize the Gaussian fixed point. The numerical solution of the linearized equations in App. E is reported to agree with the analytic exponent.","tokens_in":28807,"tokens_out":32609,"duration_ms":339932,"significance":"If the result holds, the paper identifies a new and fairly general route to nonequilibrium long-range order, with a parameter-free prediction for the anomalous exponent eta_phi as a function of the two memory exponents, and a concrete physical setup (Fig. 1) of a 2D system coupled to diffusive wires. The analytic exponent is not obtained by fitting, and the agreement with the numerical solution of the linearized equations in App. E is a genuine cross-check. The proposed mechanism is distinct from previously known ones (flocking, shear, anisotropic relaxation). However, the proof that nonlinearities do not alter the Gaussian result is the weakest part: the stability analysis is at tree level only, and one of the key scaling assignments used in that analysis appears to be incorrect. The paper is therefore potentially significant, but the central stabilization claim is not yet fully established.","major_comments":[{"comment":"The assertion that the fields L_alpha_beta and \\tilde L_alpha_beta have the same scaling dimensions as ell and \\tilde ell is not correct. Equation (5d) is decoupled from the Goldstone sector; promoting it to a Gaussian MSR action with the same sub-Ohmic kernel gamma_L(t) ~ t^{-alpha_L} yields C_LL(q) ~ const as q -> 0, which implies [L_alpha_beta] = d/2, not alpha_n/(alpha_n+alpha_L) + d/2. Recomputing the scaling dimension of the vertex g \\tilde phi_alpha L_alpha_beta phi_beta in Eq. (8) with [L] = d/2 gives [g] = 2 alpha_n/(alpha_n+alpha_L) - d/2. This is positive for alpha_n > alpha_L in d=2 and for alpha_n > alpha_L/3 in d=1, i.e. in regions where the linearized analysis (Fig. 2) predicts long-range order. The paper must either provide a correct derivation of the L_alpha_beta scaling or test the stability of these regions by an explicit one-loop or numerical calculation.","section":"Nonlinearities, Eq. (8) and Table II"},{"comment":"The stability of the Gaussian fixed point is justified only by tree-level power counting of the two operators in Eq. (8). No one-loop renormalization of the quadratic action (6) is presented, and Appendix E verifies only the linearized equations (5), not the full nonlinear Langevin equations (1). Since the central claim of the paper is the stabilization of long-range order, i.e. that nonlinearities do not change the Gaussian exponent (7), an explicit check of the beta-functions for g and g' (or an equivalent numerical simulation) is needed before the conclusion can be accepted.","section":"Nonlinearities and Appendix E"},{"comment":"The truncation that omits the amplitude mode sigma is not fully justified for the nonlinear stability analysis. Appendix A states that eliminating sigma adiabatically 'would lead to new interactions and renormalization of existing interactions,' but the paper does not analyze these terms; it only uses the truncated equations to show that a Markovian perturbation is generated. Because sigma is massive, this omission may be innocent, but in the absence of a one-loop calculation the reader cannot exclude the possibility that sigma generates a relevant operator that shifts eta_phi away from Eq. (7).","section":"Appendix A and Eq. (5)"}],"minor_comments":[{"comment":"The phrase 'consider the effect of non-Markovian dissipation, in stabilizing long range order' contains a misplaced comma; it should read 'consider the effect of non-Markovian dissipation in stabilizing long range order.'","section":"Abstract"},{"comment":"The caption does not explain the color coding of the LRO regions nor the status of the alpha_n = 0 line; the text discusses both, so a short caption note would help.","section":"Fig. 2"},{"comment":"The phrase 'The gray line indicates the value above which order in d=1 can be sustained' is misleading; since order requires eta < 1, the line at eta = 1 is the threshold above which order is lost, not sustained. Please reword.","section":"Fig. 5 caption"},{"comment":"The statement 'if alpha_i > 1 we should set it to alpha_i = 1 to obtain the correct scaling' is abrupt; it should reference Appendix B, where the super-Ohmic limit is treated.","section":"Main text, paragraph before Eq. (7)"}],"recommendation":"major_revision","confidential_remarks":"The paper presents a novel and potentially important mechanism, and the linearized exponent is well supported by the numerical solution in App. E. The main obstacle is the handling of the nonlinear sector: the L_alpha_beta scaling assignment appears incorrect, and the revised power counting may change the phase diagram. I would ask for a corrected stability analysis (one-loop or numerics) before acceptance. If the authors can show that the nonlinearities remain irrelevant with the correct [L] or that the dangerous regions are nevertheless order-stable, the paper would be a solid contribution."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a genuinely new mechanism for evading Mermin-Wagner in low dimensions. The combination of non-Markovian dissipation and conserved angular momentum produces a clean exponent eta_phi = 2 alpha_n/(alpha_n + alpha_L) and a phase diagram that is easy to state. The Gaussian-level derivation is solid: the scaling analysis is parameter-free, and the numerics on the linearized equations in App. E match the predicted exponent. The Markovian limits and the alpha=0 mass terms are handled carefully. The claim that the bath is physical is plausible but asserted; the wire-bath setup in Fig. 1 is sketched, not derived, so the connection between the memory kernels and a concrete Hamiltonian is the weakest experimental link.\n\nThe real soft spot is nonlinear stability. Irrelevance of the two g and g' vertices is shown by tree-level power counting only. The amplitude mode sigma and the non-conserved L_alpha_beta sector are dropped, and the paper asserts their scaling equals ell's rather than computing it. If any of those modes generates a relevant operator at one loop, eta_phi flows and the LRO prediction could fail. The stress-test note adds a sharper worry: from Eq. (5d) alone, a direct Gaussian calculation for sub-Ohmic gamma_L gives a C_LL(q) that is not obviously [L_alpha_beta] = [ell]. So the power-counting table may be missing something. That said, this is a gap in proof, not a demonstrated error; I don't see a smoking gun in the paper, and the linearized check gives real support.\n\nBottom line: worth a serious referee. The right outcome is probably conditional acceptance, with the nonlinear-stability section expanded to at least a one-loop calculation or a numerical RG check of the irrelevant directions. The physical motivation can stay as a conjecture. I'd bring this to reading group and would likely cite it for the mechanism.","headline":"A genuinely new mechanism -- non-Markovian dissipation plus conserved angular momentum -- with a solid Gaussian-level exponent, but the nonlinear-stability argument is thin enough that the central LRO claim is not fully closed.","tokens_in":29343,"tokens_out":2235,"would_cite":true,"duration_ms":29650,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["82B27","82C26","82C31"],"pacs":["05.40.-a","05.70.Ln"],"model":"deepseek-v4-flash","headline":"This paper establishes that long-range order in low-dimensional nonequilibrium O(N) models is stabilized by the interplay of power-law (non-Markovian) bath memory and conserved angular momenta, yielding a Goldstone correlation exponent…","keywords":["non-Markovian dissipation","long-range order","Mermin-Wagner theorem","O(N) model","Goldstone modes","conserved angular momentum","nonequilibrium steady state","Langevin dynamics"],"falsifier":"A direct test is an exact numerical solution of the full nonlinear Langevin equations in $d=1$ with conserved angular momentum and kernels satisfying $\\alpha_n<\\alpha_L$: measuring $C_{\\varphi\\varphi}(q)$ and finding it grows faster than $1/q$, or seeing the order parameter vanish with system size, would refute the predicted one-dimensional order. Measuring an exponent different from $2\\alpha_n/(\\alpha_n+\\alpha_L)$ anywhere in the $(\\alpha_n,\\alpha_L)$ plane would likewise refute the scaling analysis.","tokens_in":28315,"feed_emoji":"🧲","tokens_out":9926,"duration_ms":100612,"temperature":0.7,"pith_summary":"This paper claims that the equilibrium rule known as the Mermin-Wagner theorem — continuous symmetries cannot break spontaneously in one or two dimensions — can be bypassed in a generic class of driven systems by combining two ingredients: a bath with a long power-law memory and a conserved angular momentum. Memory alone is shown to be insufficient; the conserved charges provide slow diffusive modes that translate temporal memory into effective long-range spatial couplings. The paper derives the momentum-space singularity of Goldstone-mode fluctuations, $\\eta_\\varphi = 2\\alpha_n/(\\alpha_n+\\alpha_L)$, and shows that it drops below the divergence threshold ($\\eta_\\varphi<2$ in $d=2$, $\\eta_\\varphi<1$ in $d=1$) when the bath kernels are non-Markovian and the angular momentum is conserved. Nonlinearities are argued to be irrelevant in the ordered windows, and a fluctuation-dissipation test shows the steady state is genuinely nonthermal, so the mechanism is not a hidden equilibrium ordering.","feed_headline":"Conserved charge plus bath memory stabilizes order in 1D and 2D","feed_subtitle":"Non-Markovian baths and slow conserved modes suppress the fluctuations that normally destroy order at low dimensions.","key_machinery":"The machinery is the coupled set of linearized Langevin equations for the Goldstone modes and the conserved angular-momentum densities, together with the scaling analysis of their Martin-Siggia-Rose (MSR) action, a path-integral reformulation of the stochastic equations. Under rescaling $x\\to x/b$, the requirement that all non-diffusive terms stay marginal fixes the field dimensions in Table I and gives the dynamical exponent $z=2/(\\alpha_n+\\alpha_L)$ and the correlation exponent $\\eta_\\varphi=2\\alpha_n/(\\alpha_n+\\alpha_L)$. The same scaling powers evaluated on the two leading nonlinear vertices determine the stability of the Gaussian fixed point, which is where the linear prediction becomes a claim about the actual (nonlinear) model.","core_discovery":"On the paper's own terms, the central claim is that the linearized dynamics of a single Goldstone mode $\\varphi_\\alpha$ and its conserved angular-momentum partner $\\ell_\\alpha$, with memory kernels $\\gamma_n(t)\\propto \\theta(t)/t^{\\alpha_n}$ and $\\gamma_L(t)\\propto \\theta(t)/t^{\\alpha_L}$, is governed at the Gaussian fixed point by the exponent $\\eta_\\varphi = 2\\alpha_n/(\\alpha_n+\\alpha_L)$ for the equal-time correlation $C_{\\varphi\\varphi}(q)\\sim 1/q^{\\eta_\\varphi}$. Because $\\eta_\\varphi<d$ is the condition for finite order-parameter fluctuations, order survives in $d=2$ whenever either kernel is sub-Ohmic and in $d=1$ whenever $\\alpha_n<\\alpha_L$; without angular-momentum conservation the exponent returns to $\\eta_\\varphi=2$ and the standard divergences apply. The same framework shows the leading nonlinear vertices are irrelevant in the regions where linear analysis allows order, and that the effective temperature of the conserved system vanishes as $|\\omega|^{\\alpha_L}$ at low frequency, an unambiguous departure from thermal equilibrium.","pith_inferences":["A design principle the paper leaves implicit is that one can engineer such order by coupling a low-dimensional system to a bath with a broad, power-law distribution of relaxation times (for example, diffusive wires) while keeping the relevant angular momentum conserved; the exponents $\\alpha_n$ and $\\alpha_L$ are then set by the bath's spectral density.","Since $\\eta_\\varphi$ varies continuously with $\\alpha_n/\\alpha_L$, the analysis implies a tunable family of correlation exponents that could be measured in cold-atom or trapped-ion simulators with engineered non-Markovian reservoirs, a connection the paper does not make.","The mechanism may also support quasi-long-range order or algebraic phases at the boundary $\\alpha_n=\\alpha_L$ in $d=1$, and it may persist under weak additional Markovian dissipation, though those regimes are not analyzed in the paper.","A natural numerical test beyond the paper is to integrate the full nonlinear Langevin equations and extract $C_{\\varphi\\varphi}(q)$ across the $(\\alpha_n,\\alpha_L)$ plane; matching $2\\alpha_n/(\\alpha_n+\\alpha_L)$ would confirm the mechanism, while any deviation would reveal missing soft modes."],"forward_implications":["In $d=2$, any sub-Ohmic memory ($\\alpha_n<1$ or $\\alpha_L<1$) pushes $\\eta_\\varphi$ below 2, so Goldstone fluctuations are finite and the broken-symmetry state survives.","In $d=1$, order is stable when $\\alpha_n<\\alpha_L$ since $\\eta_\\varphi<1$; at $\\alpha_n\\ge\\alpha_L$ the correlations diverge faster than $1/q$ and order is lost.","If angular momentum is not conserved ($a=0$), the mechanism fails: $\\eta_\\varphi=2$ and the Mermin-Wagner divergences reappear in both one and two dimensions.","The steady state in the ordered regime is not a Gibbs state; the low-frequency effective temperature vanishes as $|\\omega|^{\\alpha_L}$, so the long-wavelength modes equilibrate at zero effective temperature.","The leading nonlinearities are irrelevant exactly where the linear analysis predicts order, so the Gaussian fixed point is stable and the long-range order should survive beyond the linearized approximation."],"supporting_citations":[{"why":"Supplies the equilibrium no-order theorem that the paper's mechanism is designed to circumvent.","marker":"[1]"},{"why":"Extends the no-order result to finite-temperature and superfluid settings.","marker":"[2]"},{"why":"Provides a canonical driven model in which two-dimensional order appears.","marker":"[3]"},{"why":"Gives the hydrodynamic theory of flocking order, a known nonequilibrium mechanism.","marker":"[4]"},{"why":"Shows anisotropic relaxation with a conserved order parameter can yield low-dimensional order; the paper contrasts its own mechanism with this one.","marker":"[6]"},{"why":"Establishes that long-range spatial interactions stabilize order, the equilibrium analogy motivating non-Markovian memory.","marker":"[10]"},{"why":"Supplies the Ohmic/sub-Ohmic/super-Ohmic classification that defines the non-Markovian regime.","marker":"[12]"},{"why":"Provides the dynamical critical-phenomena classification; the Markovian limit of the model maps onto model G or model A.","marker":"[13]"},{"why":"Used for the gapped Goldstone modes in the $\\alpha=0$ limit.","marker":"[14]"}],"fun_headline_variants":["Bath memory and conserved modes silence fluctuations: order in 1D and 2D","Non-Markovian baths plus conserved slow modes bypass Mermin-Wagner","Memory-friction and conserved modes conjure long-range order in low D","1D and 2D order from non-Markovian baths and conserved currents","Mermin-Wagner circumvented: memory and conserved modes stabilize order"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The prediction assumes that the only long-wavelength modes that matter are the Goldstone mode and its conserved partner, and that the amplitude mode and all other angular-momentum components stay gapped or decoupled; if any of those omitted modes becomes soft or couples resonantly, the exponent and the ordered windows could change.","fun_headline_variants_meta":{"raw":{"variants":["Bath memory and conserved modes silence fluctuations: order in 1D and 2D","Non-Markovian baths plus conserved slow modes bypass Mermin-Wagner","Memory-friction and conserved modes conjure long-range order in low D","1D and 2D order from non-Markovian baths and conserved currents","Mermin-Wagner circumvented: memory and conserved modes stabilize order"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001224,"raw_usage":{"total_tokens":5028,"prompt_tokens":938,"completion_tokens":4090,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":554,"completion_tokens_details":{"reasoning_tokens":3985}},"tokens_in":554,"tokens_out":4090,"duration_ms":33600,"temperature":1.0,"reasoning_tokens":3985,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T20:40:05.150696+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A direct test is an exact numerical solution of the full nonlinear Langevin equations in $d=1$ with conserved angular momentum and kernels satisfying $\\alpha_n<\\alpha_L$: measuring $C_{\\varphi\\varphi}(q)$ and finding it grows faster than $1/q$, or seeing the order parameter vanish with system size, would refute the predicted one-dimensional order. Measuring an exponent different from $2\\alpha_n/(\\alpha_n+\\alpha_L)$ anywhere in the $(\\alpha_n,\\alpha_L)$ plane would likewise refute the scaling analysis.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Shows anisotropic relaxation with a conserved order parameter can yield low-dimensional order; the paper contrasts its own mechanism with this one."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Used for the gapped Goldstone modes in the $\\alpha=0$ limit."}],"review_version":1}