{"id":"94e18c80-db0f-4563-971f-4dec1fb9796c","arxiv_id":"1908.02715","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A complex Langevin simulation of 2D spin-orbit coupled bosons gives a density equation of state where mean-field underestimates density and spin-orbit coupling suppresses pseudo-condensation.","lead":"This paper reports a non-perturbative lattice simulation of two-dimensional bosons with spin-orbit coupling, using the complex Langevin method to handle the sign problem. The results show that spin-orbit coupling lowers the particle density and destroys the pseudo-condensate that forms in finite simulation boxes.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Complex Langevin correctness is the load-bearing assumption; the paper checks only Im<n>≈0, which cannot rule out a biased stationary distribution for the interacting data.","rationale":"The reader's weakest assumption matches my own: the entire interacting section rests on CL. I considered two alternative concerns—the lack of a continuum extrapolation and the possibility that the mean-field comparison is trivially dominated by zero MF density at μ≤0—but neither is as load-bearing. The continuum-extrapolation issue mostly affects quantitative precision, not the sign or qualitative shape of the claim, and the paper does not rest the strongest statement only on μ≤0; at βμ=0.4 the MF density is nonzero and the comparison is meaningful. The CL concern, by contrast, could remove the interacting results altogether if the process samples the wrong distribution. The free-field validation and the Im-density check are real supporting evidence, but they are exactly what is insufficient: the free case has a linear drift, while the interacting action has quartic drifts in complexified fields, a regime where CL failures are documented. The proposed boundary-term test is the standard criterion that would turn the conditional into an accept or reject.","tokens_in":11001,"tokens_out":8738,"duration_ms":111208,"concrete_test":"At one representative interacting point (e.g., aκ=0.4π/2, βμ=0.4, λ/a=0.5), measure the standard complex-Langevin boundary term for the density and condensate observables following Aarts-Seiler-Stamatescu: accumulate the long-time average of the drift divergence times the observable and record the large-|Im φ| tail of the complexified-field distribution. If the boundary term is nonzero within errors, or if the tail does not decay fast enough to make boundary contributions vanish, the CL averages are not the path-integral expectation values and the mean-field comparison in Figs. 4-7 is invalidated.","verdict_should_be":"UNCHANGED","load_bearing_attack":"All interacting results supporting the central claim—the density equation of state and the pseudo-condensate fractions—are produced solely by complex Langevin for a genuinely complex action. The paper's only stated correctness test is that the imaginary part of the density is statistically zero (Fig. 3), and the only quantitative validation is against the noninteracting solution (Fig. 1). Neither is sufficient. Complex Langevin can converge to a wrong stationary measure if boundary terms in the associated Fokker-Planck equation fail to vanish or if the complexified drift has run-away directions; these failure modes can leave Im<n> small while biasing the real part. The noninteracting check exercises only the quadratic part of the drift, whereas the interacting action has quartic terms whose complexified dynamics are precisely where CL correctness is most fragile. Since the headline comparison against mean field is a difference between interacting CL data and mean field, any such bias directly contaminates the strongest claim. The presence of a nonzero imaginary part would be a clear failure, but its absence is not certification.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript presents a lattice complex-Langevin study of a two-dimensional Bose gas with two pseudo-spin components, Rashba-Dresselhaus-type spin-orbit coupling, and contact interactions. The authors derive the exact noninteracting lattice density, benchmark the CL code against it, and then compute the interacting density and pressure equations of state as functions of chemical potential and spin-orbit coupling, together with a finite-volume pseudo-condensate fraction. The two central claims are that mean-field theory underestimates the average density (most visibly for stronger SOC) and that SOC depletes the finite-volume pseudo-condensate.","tokens_in":11115,"tokens_out":8605,"duration_ms":91080,"significance":"The paper's strengths are the closed-form noninteracting solution, the absence of fitted parameters, the direct evaluation of the lattice path integral, and the independent mean-field benchmark. If the interacting CL data are correct, the density and pressure equations of state represent a useful non-perturbative reference for spin-orbit coupled bosons, a regime in which sign problems preclude standard Monte Carlo. The manuscript is written clearly and the lattice treatment of SOC as a background non-Abelian field is elegant. Its central quantitative claims currently rest on CL validation that is necessary but not sufficient, on data shown without visible error bars, and on a single-volume pseudo-condensate estimator; these issues are addressable and should be fixed before publication.","major_comments":[{"comment":"The interacting results that support the central claim are produced solely by complex Langevin for a genuinely complex action (Section III). The only correctness checks are the noninteracting benchmark (Fig. 1) and the vanishing imaginary part of the density (Fig. 3, Section IV B). Neither is sufficient: CL can converge to a wrong stationary measure with small imaginary parts if boundary terms in the associated Fokker-Planck equation do not vanish, and the noninteracting check exercises only the quadratic part of the drift while the quartic interaction is where CL failures are most likely. Please add an interacting-regime benchmark (e.g., exact diagonalization on a small lattice, a determinant or worldline QMC calculation in a parameter window, or the standard boundary-term criterion) and report Langevin-time discretization and thermalization diagnostics.","section":"Section III and Section IV B, Fig. 3"},{"comment":"The central density figures show CL points without visible error bars, whereas the pressure data in Fig. 6 are accompanied by bootstrap errors. Because the main claim is a quantitative comparison between CL and mean-field densities, this omission prevents the reader from judging whether the deviations are statistically significant, particularly for stronger SOC where differences are small. Please include error bars (or state explicitly that they are smaller than the symbols and report numerical uncertainties), the number of independent configurations, and the autocorrelation times used.","section":"Section IV B, Figs. 4, 5, 8, and 9"},{"comment":"The pseudo-condensate claim is based on a single finite-volume estimator, R_{ss'} = G_{ss'}(aN_x/2)/G_{ss'}(0), at N_x=20 (Eq. (24), Fig. 7). Since the pseudo-condensate is itself a finite-volume effect in 2D, the observed decay of R with κ cannot establish that SOC 'destroys' the pseudo-condensate without a finite-size scaling analysis. Please show R(κ) for at least one additional volume or provide a scaling argument that the zero-κ value survives the thermodynamic limit while the finite-κ value does not.","section":"Section IV B, Eq. (24), and Fig. 7"}],"minor_comments":[{"comment":"The interaction terms are typeset as 'λ +g /8' and 'λ−g /8'; they should be (λ+g)/8 and (λ−g)/8 to avoid ambiguity.","section":"Eq. (4)"},{"comment":"The sentence 'For complex fields, both the real and imaginary parts become complex' is confusing; the standard complex-Langevin complexification of an already complex scalar field should be described more precisely.","section":"Section III"},{"comment":"The mean-field density in Eq. (21) is per flavor, while Figs. 4 and 5 plot total density; please state the factor-of-2 convention in the caption or text.","section":"Eq. (21) and Figs. 4 and 5"},{"comment":"The discussion of the βμ≥0 condensation instability would benefit from a brief statement of how the instability was detected (e.g., runaway Langevin behavior) and how such configurations were excluded from the analysis.","section":"Section IV A"},{"comment":"There are minor typos: 'distincion' in Section IV C and 'F.A and acknowledges' in the acknowledgments.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The paper is within scope for a quantum-gas journal and the central numerical results are potentially valuable. I would not reject the manuscript outright, but the CL validation issue is serious enough that I cannot recommend acceptance until it is addressed. There are no concerns about novelty disclosure or citation patterns."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nQuick take: this paper is the first to get non-perturbative thermodynamic data for 2D spin-orbit coupled bosons with contact interactions, using complex Langevin on a lattice. The density and pressure equations of state and the pseudo-condensate depletion are genuinely new. If you work on ultracold gases or lattice methods for sign problems, this is worth knowing about.\n\nThe good parts: the noninteracting case is solved exactly on the lattice, and the CL results match it. The vanishing imaginary part of the density is consistent with a correct CL calculation, and the bootstrap errors on the pressure are a nice touch. The paper is clearly written and does not oversell its claims.\n\nThe soft spots are real, and they are concentrated where it matters. The only correctness tests for CL are the noninteracting benchmark and Im<n>≈0. That is weak. Im<n> being zero is necessary but not sufficient; CL can converge to the wrong stationary distribution with small imaginary parts, especially when the action has quartic terms, which is exactly the interacting case here. The noninteracting test only exercises the quadratic part of the drift, so it does not certify the interacting data. Since the central comparison is between interacting CL data and mean field, a bias in CL would directly contaminate the strongest claim. I would want to see a nontrivial cross-check, such as a direct comparison at weak coupling, a different algorithm for a subset of parameters, or at least a careful discussion of the CL criteria from Aarts et al.\n\nTwo other issues, more minor. The density figures show no visible error bars; if they are smaller than the symbols, the text should say so. And the pseudo-condensate fraction is a single finite-volume ratio at half the box, with no finite-size scaling or volume dependence. The statement that SOC destroys the pseudo-condensate is therefore suggestive, not established. There is also no continuum extrapolation; all results are at one lattice spacing.\n\nNone of this makes the paper wrong. The qualitative direction, SOC reduces density and suppresses off-diagonal long-range order, is plausible and likely to survive. But the quantitative claims are not fully supported.\n\nMy recommendation: send it to peer review, not desk reject. A good referee can push the authors to strengthen the CL validation and add the missing error/finite-size information. The paper is useful to a specific community and deserves referee time.\n\nRegards.","headline":"First non-perturbative equation-of-state data for 2D spin-orbit coupled bosons, with a plausible physics message but a load-bearing complex Langevin step that is not verified well enough.","tokens_in":11687,"tokens_out":3157,"would_cite":false,"duration_ms":35080,"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":"For a two-dimensional gas of spin-orbit coupled bosons with contact interactions, non-perturbative simulations show that mean-field theory underestimates the density and that spin-orbit coupling destroys the finite-volume pseudo-condensate.","keywords":["spin-orbit coupled bosons","complex Langevin","equation of state","pseudo-condensate","lattice field theory","sign problem","ultracold atoms","non-perturbative methods"],"falsifier":"A direct check would be to compute $\\langle n\\rangle$ for the same lattice action at $\\lambda/a=0.5$, $a\\kappa=0.3\\pi/2$, and $\\beta\\mu\\approx0.4$ with an independent sign-problem-free method; if that method disagrees with the complex Langevin value beyond the quoted errors, the interacting equation of state—and therefore the mean-field comparison—is not established.","tokens_in":10756,"feed_emoji":"🌀","tokens_out":14462,"duration_ms":144307,"temperature":0.7,"pith_summary":"This paper establishes the non-perturbative thermal equation of state of a two-dimensional Bose gas with spin-orbit coupling and repulsive contact interactions, using complex Langevin simulations on a $20^2\\times64$ lattice. The central finding is that mean-field solutions underestimate the average density, with the discrepancy growing as the spin-orbit coupling increases or the chemical potential turns positive. The simulations also show that a pseudo-condensate, visible as equal-spin off-diagonal long-range order in the finite volume, is destroyed by the spin-orbit coupling. Because the spin-orbit term makes the Euclidean action complex, standard Monte Carlo sampling is impossible here, so the work doubles as a test of complex Langevin for synthetic-gauge-field cold-atom systems.","feed_headline":"Mean-field theory underestimates spin-orbit boson density","feed_subtitle":"Non-perturbative simulations show the gap grows with coupling and spin-orbit terms erase the finite-volume condensate.","key_machinery":"The central object is the lattice action in which the spin-orbit coupling enters as a constant background SU(2) gauge field, $v_j=e^{-i\\kappa_j\\sigma_j}$, so that arbitrarily large couplings can be studied on a small lattice. The calculation is carried by complex Langevin dynamics: the fields are complexified and evolved in a fictitious time under Gaussian noise, sampling the complex weight $e^{-S}$; the mean-field result is the noiseless limit of the same equations. The free (quadratic) part is diagonalized in a helicity basis, a change of variables in which the two free dispersion branches decouple, producing eigenvalues $\\lambda_\\pm(\\vec p,\\omega,\\mu,\\kappa)$ whose zero-crossing locates the condensation instability and whose lattice sum gives the exact noninteracting density used as a baseline.","core_discovery":"On a periodic lattice with $N_x=20$, $N_\\tau=64$, $\\xi=1/8$, and same-species coupling $\\lambda/a=0.5$, the paper obtains the density equation of state $n(\\beta\\mu,\\kappa)$ for isotropic ($\\kappa_x=\\kappa_y=\\kappa$) and anisotropic ($\\kappa_y=\\eta_{\\rm soc}\\kappa_x$) spin-orbit coupling. Density decreases monotonically with $\\kappa$, reaching a minimum at $a\\kappa=\\pi/2$, the lattice image of $\\kappa\\to\\infty$; for $\\mu\\leq0$, where the mean-field density is zero, the simulations give a finite density. Noiseless Langevin (mean field) systematically lies below the complex-Langevin density, with the gap widening for larger $\\kappa$ and positive $\\beta\\mu$. The equal-spin correlation at half the lattice size, $R_{\\uparrow\\uparrow}=G_{\\uparrow\\uparrow}(aN_x/2)/G_{\\uparrow\\uparrow}(0)$, is nonzero at $\\kappa=0$ and $\\mu>0$—the finite-volume pseudo-condensate—and drops to zero for $a\\kappa\\gtrsim0.1\\pi$; opposite-spin correlations are statistically zero. With anisotropic spin-orbit coupling the density and the pseudo-condensate decay more slowly with $|\\vec\\kappa|$, and at $\\eta_{\\rm soc}=0$ the density displays a periodicity of $2\\pi/N_x$ in the coupling.","pith_inferences":["A volume-scaling study at fixed temperature would show whether the pseudo-condensate destruction by spin-orbit coupling survives the thermodynamic limit or is purely a finite-volume effect.","Because the lattice formulation packages the spin-orbit coupling as a constant background SU(2) gauge field, the same complex-Langevin setup should transfer to spin-orbit coupled fermions and to rotating bosons without new algorithmic ingredients.","If ultracold-atom experiments measure this density equation of state, mean-field fits would infer a smaller density at fixed $\\mu$ and $\\kappa$; the deviation is a clean quantitative signature of beyond-mean-field physics in synthetic spin-orbit coupled gases."],"forward_implications":["For $\\mu\\leq0$, where the mean-field density is zero, the interacting density is nonzero, so mean-field estimates of the phase boundary in this regime are unreliable.","The density and pressure equations of state reported here are direct, parameter-free predictions that can be compared with measurements on synthetic spin-orbit coupled Bose gases.","At fixed chemical potential, increasing the spin-orbit coupling lowers the density monotonically, with the minimum at $a\\kappa=\\pi/2$, so stronger coupling makes the gas more dilute at the same $\\mu$.","A pseudo-condensate appears in the finite volume at $\\kappa=0$ and is destroyed for $a\\kappa\\gtrsim0.1\\pi$, showing that spin-orbit coupling suppresses off-diagonal long-range order already at small coupling.","For anisotropic spin-orbit coupling, both the density and the pseudo-condensate fraction decay more slowly with $|\\vec\\kappa|$, and at $\\eta_{\\rm soc}=0$ the density develops a lattice-induced period $2\\pi/N_x$."],"supporting_citations":[{"why":"Introduces stochastic quantization, the foundation from which complex Langevin is derived.","marker":"[38]"},{"why":"Extends stochastic quantization to complex actions by complexifying the fields, the step that makes this simulation possible.","marker":"[40]"},{"why":"Justifies keeping the Langevin noise real, the prescription used throughout the simulations.","marker":"[47]"},{"why":"Supplies the adaptive step-size integration used to solve the Langevin equations.","marker":"[39]"},{"why":"Establishes complex Langevin for repulsive bosons with a sign problem, the immediate precedent for this calculation.","marker":"[24]"},{"why":"Provides the continuum eigenvalues for isotropic spin-orbit coupling used to locate the condensation instability and interpret the large-coupling limit.","marker":"[22]"},{"why":"Reports a similar spin-orbit induced destruction of condensate order in three dimensions, the comparison invoked for the pseudo-condensate result.","marker":"[49]"}],"fun_headline_variants":["Spin-orbit coupling erases pseudo-condensates in 2D bosons","Mean-field fails for spin-orbit bosons, Langevin corrects","Spin-orbit coupling boosts density gap beyond mean-field","Pseudo-condensates die under spin-orbit coupling","Langevin shows mean-field misses spin-orbit boson density"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the complex Langevin process converges to the correct quantum expectation values for this complex action; the paper's only direct check is that the imaginary part of the density is statistically zero.","fun_headline_variants_meta":{"raw":{"variants":["Spin-orbit coupling erases pseudo-condensates in 2D bosons","Mean-field fails for spin-orbit bosons, Langevin corrects","Spin-orbit coupling boosts density gap beyond mean-field","Pseudo-condensates die under spin-orbit coupling","Langevin shows mean-field misses spin-orbit boson density"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00083,"raw_usage":{"total_tokens":3638,"prompt_tokens":970,"completion_tokens":2668,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":586,"completion_tokens_details":{"reasoning_tokens":2576}},"tokens_in":586,"tokens_out":2668,"duration_ms":18950,"temperature":1.0,"reasoning_tokens":2576,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T14:36:11.737100+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A direct check would be to compute $\\langle n\\rangle$ for the same lattice action at $\\lambda/a=0.5$, $a\\kappa=0.3\\pi/2$, and $\\beta\\mu\\approx0.4$ with an independent sign-problem-free method; if that method disagrees with the complex Langevin value beyond the quoted errors, the interacting equation of state—and therefore the mean-field comparison—is not established.","supporting_citations":[{"cited_title":"Parisi and Y.-s","cited_arxiv_id":null,"evidence_quote":"Introduces stochastic quantization, the foundation from which complex Langevin is derived."},{"cited_title":"Parisi, Phys","cited_arxiv_id":null,"evidence_quote":"Extends stochastic quantization to complex actions by complexifying the fields, the step that makes this simulation possible."},{"cited_title":"Third-order perturbative lattice and complex Langevin analyses of the finite-temperature equation of state of non-relativistic fermions in one dimension","cited_arxiv_id":"1702.04666","evidence_quote":"Establishes complex Langevin for repulsive bosons with a sign problem, the immediate precedent for this calculation."},{"cited_title":"Renormalization of interactions of ultracold atoms in simulated Rashba gauge fields","cited_arxiv_id":"1107.3162","evidence_quote":"Provides the continuum eigenvalues for isotropic spin-orbit coupling used to locate the condensation instability and interpret the large-coupling limit."},{"cited_title":"Ozawa and G","cited_arxiv_id":null,"evidence_quote":"Reports a similar spin-orbit induced destruction of condensate order in three dimensions, the comparison invoked for the pseudo-condensate result."}],"review_version":1}