{"id":"bb021fb2-52f9-41c2-a1c7-5338591b221c","arxiv_id":"2501.09575","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The relaxational dynamics at topological Z2-gauge transitions has dynamic exponent z=2.55(6), while the XY-type transitions in the Z2-gauge XY model match the standard XY dynamics.","lead":"This paper measures how slowly Monte Carlo updates relax near phase transitions in three-dimensional Z2 gauge models. It finds the topological transition is much slower (z=2.55) than the equivalent Ising transition (z=2.02), while the XY-type transitions behave like ordinary XY magnets.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The headline z=2.55(6) conflicts at about 2.2 sigma with the published out-of-equilibrium z=2.70(3) (Ref. [82]), a discrepancy the paper notes but does not resolve, and the Conclusions overstate confirmation.","rationale":"The paper is a careful equilibrium MC study: the fits are stable over Lmin>=16, the energy/Polyakov cross-check is provided, and the XY-line results match model-A expectations. I do not think the Polyakov-loop assumption is the weakest link: the energy data give consistent z, and the Lmin stability makes an observable-specific exponent unlikely. The most load-bearing unresolved issue is the conflict between the headline equilibrium z=2.55(6) and the out-of-equilibrium z=2.70(3) of Ref. [82]. Both are advertised as the same universal dynamic exponent of the same transition; the text acknowledges but does not explain the gap, and the Conclusions sentence claiming confirmation is stronger than the data warrant. A direct out-of-equilibrium simulation with the same Metropolis dynamics would arbitrate: if it reproduces 2.55(6), Ref. [82]'s protocol had a systematic issue; if it returns about 2.70, the finite-L equilibrium fits in Sec. IV are missing corrections. This does not require rejection: the equilibrium data and analysis are internally sound, and the discrepancy is disclosed. It does justify keeping the reader's conditional verdict, pending either the proposed check or a quantitative explanation. No verdict change.","tokens_in":18216,"tokens_out":9509,"duration_ms":99265,"concrete_test":"Perform an out-of-equilibrium critical dynamic simulation of the same model with the same Metropolis update: start from fully disordered and fully ordered configurations, quench to Kc=0.761413292, and for L=16,24,32,44 measure the time-dependent Polyakov-loop autocorrelation; extract z from the scaling collapse of tau(L) or from the short-time growth of the order parameter. If the result is 2.55(6), the conflict with Ref. [82] is a methodological artifact and the paper's value stands; if it is closer to 2.70(3), the equilibrium fit is biased (e.g., by unrecognized corrections) and the central claim needs revision.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative claim is z=2.55(6) for the 3D Z2-gauge topological transition (Sec. IV, Eq. (15)). The paper itself reports that the out-of-equilibrium estimate of Ref. [82] is z=2.70(3) and states that 'such discrepancy should be further investigated', yet the Conclusions say the results confirm Ref. [82]. The difference is about 2.2 combined standard deviations (0.15 with sigma approx 0.067), so the disagreement is not negligible. Because z is a universal dynamic exponent expected to be the same for equilibrium and out-of-equilibrium critical dynamics, an unresolved external conflict with a published measurement of the same quantity is directly load-bearing for the headline number. The internal evidence (stable fits for Lmin>=16, consistency between energy and Polyakov-loop data, agreement with Ref. [81] z=2.5(3)) supports the equilibrium analysis but does not explain why a different protocol gives 2.70(3). The weakest internal assumption (tau_P ~ L^z with the bulk exponent) is mitigated by the energy-observable consistency, so the external discrepancy is the more serious unresolved threat to the claim.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the relaxational critical dynamics of the three-dimensional Z2 lattice gauge model and the Z2-gauge XY model under a standard locally reversible Metropolis dynamics. Using Monte Carlo simulations at the critical points, the authors measure autocorrelation times of gauge-invariant observables and extract the dynamic critical exponent z from the finite-size scaling tau ~ L^z. For the topological transitions (the pure Z2-gauge model and the DD-DO line of the gauge XY model) they report z=2.55(6), which is significantly larger than the 3D Ising value z=2.0245(15) despite the static equivalence of the two models via duality. For the DD-O (LGW) and DO-O (LGW*) transitions of the gauge XY model they find z consistent with the standard XY value z~2.022. The paper argues that the nonlocal nature of the duality mapping explains the different dynamic universality class at the topological transitions, while gauge modes do not affect the model-A dynamics at the LGW-type transitions.","tokens_in":18296,"tokens_out":7189,"duration_ms":72555,"significance":"The main result is of clear interest to statistical mechanics and lattice gauge theory: it demonstrates that static universality via a nonlocal duality does not imply dynamic universality, and it provides the first high-precision equilibrium estimate of the dynamic critical exponent for the topological Z2-gauge transition. The paper is commendably transparent: Appendix A reports the full autocorrelation-time data, the fits are checked for stability against the choice of Lmin and of the time-definition parameters, and the Polyakov-loop results are cross-checked with energy-density data. The XY results provide concrete evidence for the conjecture that gauge modes do not alter the model-A dynamic universality class at LGW and LGW* transitions. Even if the precise value of z were to shift with future studies, the qualitative conclusion of a dynamic universality class distinct from Ising is robust and significant.","major_comments":[{"comment":"The statement in the Conclusions that the estimate z=2.55(6) 'confirm[s] earlier numerical results [82]' is not supported by the paper's own discussion in Sec. IV, where the authors report z=2.70(3) from Ref. [82] and note that the discrepancy should be further investigated. With combined uncertainty of about 0.067, the difference is roughly 2.2 standard deviations. Please rephrase the conclusion to say that the result is qualitatively consistent with the slow dynamics found in Ref. [82], and quantify the discrepancy when citing that work.","section":"Sec. VI (Conclusions) and Sec. IV"}],"minor_comments":[{"comment":"The sentence reporting an unbiased analysis leading to z=2.52(8) would benefit from a brief statement of the fit range (Lmin) and of whether scaling corrections were included, so that the reader can judge the consistency with the headline z=2.55(6).","section":"Sec. V.A"},{"comment":"In Table I, the column label 'tau x(H)' could be confused with the Hamiltonian H; consider using 'tau_x(E)' or adding a footnote to clarify that H denotes the energy density.","section":"Appendix A"},{"comment":"For the DD-DO line study at J=0.1, the value Kc approximately 0.7612 is used; given Eq. (6) the shift from Kc(J=0) is -2e-4, so the value is correct, but stating the precise value used would improve reproducibility.","section":"Sec. II.B"},{"comment":"The comparison value z=2.022(5) for the XY universality class is obtained from the epsilon expansion; the text could explicitly note that this is an analytical estimate rather than an independent numerical determination, which would help readers assess the strength of the comparison.","section":"Sec. V"}],"recommendation":"minor_revision","confidential_remarks":"The manuscript is well within the scope of the journal and the central computational results appear sound. The main issue is the unresolved tension with Ref. [82] and the overstatement in the Conclusions that the result 'confirms' that work. I do not think an independent out-of-equilibrium simulation is required for acceptance, but the authors should soften the confirmation language and perhaps add a quantitative statement of the discrepancy. The paper is otherwise carefully written and the data are presented in a way that allows independent reanalysis."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe paper is a careful equilibrium Monte Carlo study of relaxational dynamics in 3D Z2-gauge models. The headline number, z=2.55(6) for the topological transition, is credible and is a real improvement over earlier z=2.5(3). The genuinely new part is the first dynamic characterization along the DD-O, DO-O, and DD-DO lines of the Z2-gauge XY model: the two XY-type transitions match the ungauged XY value z≈2.022, while the topological line matches the pure gauge model. That static duality does not imply dynamic duality is a clean and important point.\n\nThe internal evidence for the main claim is solid. The fits to τ=cL^z are stable for Lmin≥16, Polyakov-loop and energy-density autocorrelations agree, and Appendix A gives enough numbers to redo the fits. The XY-line results are not just a single exponent from one observable; the rescaled data collapse with the XY z and ω is reasonable.\n\nThe soft spots are real but not fatal. The paper notes that the out-of-equilibrium estimate from Ref. [82] is z=2.70(3), about 2.2 sigma away from 2.55(6), and says the discrepancy 'should be further investigated'—then the Conclusions say the results 'confirm' Ref. [82]. That overstates the situation. You cannot both flag a conflict and call it confirmation. Since z is a universal exponent, an unresolved external conflict with a published value is directly relevant to the headline claim. This needs to be addressed in a revision, at least by softening the language and discussing possible sources (protocol differences, finite-time effects, observables). A second caveat: calling the XY-type transitions 'the same dynamic universality class' is an inference from z alone; a full classification would require more, but this is a minor overreach typical of the field. The DD-DO line data go only to L=32 with large errors, though the unbiased fit z=2.52(8) is consistent. No code or raw data are released, but the tables in Appendix A are sufficient for the main fits.\n\nWho is this for? Practitioners of lattice gauge simulations and people working on dynamic critical phenomena in systems with gauge symmetries. It deserves a serious referee: the central equilibrium result is well supported, the novelty is real, and the unresolved discrepancy is the kind of thing peer review should push on. I would accept it for review with a request to fix the conclusions and discuss the Ref. [82] conflict more carefully.","headline":"Solid equilibrium MC determination of z=2.55(6) for the topological Z2-gauge transition; the main open issue is an unresolved ~2.2 sigma conflict with an out-of-equilibrium estimate, and the conclusions overstate confirmation.","tokens_in":18999,"tokens_out":2523,"would_cite":true,"duration_ms":24141,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["82B20","82B27","82B80"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper claims that at the topological transition of the 3D $\\mathbb{Z}_2$-gauge model, the Metropolis relaxational dynamics slows down as $\\tau\\sim L^z$ with $z=2.55(6)$, considerably slower than the Ising value $z=2.0245(15)$ despite…","keywords":["topological phase transition","dynamic critical exponent","critical slowing down","lattice gauge theory","Z2-gauge symmetry","Metropolis dynamics","model A universality","XY universality class"],"falsifier":"Measure equilibrium autocorrelation times of the local plaquette energy at the $\\mathbb{Z}_2$-gauge critical point for lattices of size $L=64$ to $128$, fit $\\tau\\sim cL^z$ with corrections, and compare to the Polyakov-loop result $z=2.55(6)$; an energy-based exponent that disagrees beyond errors would show that the Polyakov loop is not the bulk critical mode, while agreement would confirm the claimed dynamic class. A second check is to repeat the out-of-equilibrium slow-crossing measurement with the same lattice sizes and observables; if it does not converge to $z=2.55(6)$, the equilibrium and nonequilibrium protocols are not probing the same exponent.","tokens_in":17856,"feed_emoji":"⏱️","tokens_out":8640,"duration_ms":74654,"temperature":0.7,"pith_summary":"This paper tries to establish that the presence of a local $\\mathbb{Z}_2$ gauge symmetry changes the universal critical dynamics of continuous phase transitions in three dimensions, not just their static properties. At the topological transition of the pure $\\mathbb{Z}_2$-gauge model, a standard locally reversible Metropolis dynamics relaxes with $z=2.55(6)$, far slower than the $z=2.0245(15)$ of the 3D Ising universality class, even though the two models have identical static critical behavior because of duality. Along the DD-DO line of the $\\mathbb{Z}_2$-gauge XY model, the same slower topological dynamic class appears. By contrast, at the DD-O and DO-O transitions of that model, where XY order exists, the dynamics belongs to the standard XY model-A universality class with $z\\approx 2.022$, independent of whether the order parameter is gauge invariant. If correct, the paper shows that dynamic universality in gauge systems is not fixed by static universality alone and that locally reversible algorithms at topological gauge transitions pay an extra cost in autocorrelation time.","feed_headline":"Topological gauge transitions slow critical relaxation to z=2.55(6)","feed_subtitle":"Same static exponents, different dynamics: local updates crawl at gauge transitions but match XY elsewhere.","key_machinery":"The analysis is carried by finite-size scaling of autocorrelation times at the critical point: self-consistent exponential autocorrelation times $\\tau_x$ and integrated autocorrelation times $\\tau_{x,\\mathrm{int}}$ are fitted to $\\tau=cL^z(1+c_\\omega L^{-\\omega}+\\cdots)$. The observable that carries the topological-transition signal is the Polyakov loop, a nonlocal product of $\\mathbb{Z}_2$ link variables along one direction, which couples more strongly to the slowest modes than the energy density and shows smaller scaling corrections. The dynamics is a Metropolis sweep that proposes local, reversible flips of the fundamental gauge and spin variables, realizing model-A relaxational dynamics. The contrast between the topological and XY transitions is made sharp by the nonlocality of the duality that relates the $\\mathbb{Z}_2$-gauge model to the Ising model: statics are mapped in the energy sector, but a local update in one model does not map to a local update in the other.","core_discovery":"For the topological $\\mathbb{Z}_2$-gauge transition in 3D, the paper reports $z=2.55(6)$ from equilibrium Metropolis simulations at the critical coupling, using autocorrelation times of the nonlocal Polyakov loop as the observable most strongly coupled to the slowest modes; energy-density autocorrelations give consistent but noisier results and substantially smaller time scales. The same exponent, within errors, is found along the DD-DO transition line of the $\\mathbb{Z}_2$-gauge XY model at $J=0.1$, with an unbiased fit giving $z=2.52(8)$. In contrast, along the DD-O line at $K=0.5$ and the DO-O line at $K=1$, autocorrelation times of the susceptibility of the gauge-invariant bilinear operator $Q^{ab}_x$ scale as $\\tau\\sim L^z$ with $z=2.02(4)$ and $z=1.96(8)$ respectively, fully consistent with the standard XY value $z=2.022(5)$. The paper concludes that LGW and LGW$\\times$ transitions in these gauge models belong to the same dynamic universality class as model-A dynamics in the corresponding ungauged $\\Phi^4$ theory, while the topological transitions form a distinct, slower dynamic class.","pith_inferences":["A direct equilibrium measurement of $z$ from a purely local gauge-invariant operator on larger lattices, such as the plaquette energy at $L=64$ to $128$, would discriminate between the paper's single-exponent scenario and the possibility that the Polyakov loop carries a separate, slower mode; the paper's energy data are consistent but have much larger errors.","The discrepancy between the equilibrium value $z=2.55(6)$ and the out-of-equilibrium estimate $z=2.70(3)$ cited by the authors could reflect different scaling variables in slow-crossing protocols; if the two exponents remain different when measured with the same observable and the same finite-size scaling, the notion of a single dynamic exponent for this transition would need revision.","The same classification logic suggests testing continuous gauge groups: at GFT transitions where gauge modes themselves are critical, one would not expect the simple reduction to ungauged model-A dynamics, so a separate dynamic universality class may appear there."],"forward_implications":["At the 3D $\\mathbb{Z}_2$-gauge topological transition, the dynamic exponent $z=2.55(6)$ means autocorrelation times grow roughly as $L^{2.55}$, so reaching a given precision at large $L$ requires substantially longer runs than an Ising simulation at the same lattice size.","The $\\mathbb{Z}_2$-gauge XY model inherits the same slow topological dynamic class along its DD-DO line, so the slower dynamics is a property of the topological $\\mathbb{Z}_2$-gauge universality class, not of the pure model alone.","The DD-O and DO-O transitions of the $\\mathbb{Z}_2$-gauge XY model show standard XY model-A dynamics with $z\\approx 2.022$, so the Metropolis dynamics on gauge-dependent variables behaves like the model-A Langevin dynamics of the effective $\\Phi^4$ theory for gauge-invariant order parameters.","Because the duality between the $\\mathbb{Z}_2$-gauge model and the Ising model is nonlocal, no local algorithm on the Ising side can reproduce the gauge-model dynamics; local updates in the two dual models belong to different dynamic universality classes.","The paper conjectures that at all LGW and LGW$\\times$ transitions in gauge systems, relaxational critical dynamics is the same as in the corresponding LGW $\\Phi^4$ theory."],"supporting_citations":[{"why":"Establishes the duality between the 3D $\\mathbb{Z}_2$-gauge model and the Ising model, which gives equal static critical behavior and motivates the dynamic comparison.","marker":"[13]"},{"why":"Provides the precise Ising dynamic exponent $z=2.0245(15)$ that serves as the baseline for the slower gauge-model value.","marker":"[99]"},{"why":"Supplies the phase diagram and critical parameters of the 3D $\\mathbb{Z}_2$-gauge $N$-vector model used to locate the DD-O, DO-O, and DD-DO transition lines.","marker":"[71]"},{"why":"Provides the classification of transitions into LGW, LGW$\\times$, GFT, and topological classes that organizes the study.","marker":"[7]"},{"why":"Defines model-A relaxational dynamics and the general framework of dynamic universality classes used throughout.","marker":"[79]"},{"why":"Earlier equilibrium study of the same model whose result $z=2.5(3)$ is compared with the new estimate.","marker":"[81]"},{"why":"Reports out-of-equilibrium estimate $z=2.70(3)$, whose discrepancy with the equilibrium value the paper flags for further investigation.","marker":"[82]"},{"why":"Shows how a stochastic gauge fixing uncovers the nongauge-invariant spin order parameter at DO-O transitions, supporting the LGW$\\times$ classification.","marker":"[72]"}],"fun_headline_variants":["Gauge symmetry slows critical relaxation to z=2.55","Topological Z2 gauge transitions: critical dynamics slower than Ising","Different dynamic classes for topological vs XY gauge transitions","Z2 gauge critical dynamics: z=2.55 at topological transitions","Relaxational dynamics at Z2 gauge transitions: z differs by class"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The estimate $z=2.55(6)$ rests on the assumption that the autocorrelation time of the nonlocal Polyakov loop is governed by the same critical mode as the transition's critical slowing down, so that $\\tau_P\\sim L^z$ with the bulk dynamic exponent; if the Polyakov loop had a different scaling or a distinct slow mode, the fitted exponent would not be the dynamic exponent of the bulk transition.","fun_headline_variants_meta":{"raw":{"variants":["Gauge symmetry slows critical relaxation to z=2.55","Topological Z2 gauge transitions: critical dynamics slower than Ising","Different dynamic classes for topological vs XY gauge transitions","Z2 gauge critical dynamics: z=2.55 at topological transitions","Relaxational dynamics at Z2 gauge transitions: z differs by class"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000706,"raw_usage":{"total_tokens":3262,"prompt_tokens":1102,"completion_tokens":2160,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":718,"completion_tokens_details":{"reasoning_tokens":2072}},"tokens_in":718,"tokens_out":2160,"duration_ms":14999,"temperature":1.0,"reasoning_tokens":2072,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T19:52:53.257913+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure equilibrium autocorrelation times of the local plaquette energy at the $\\mathbb{Z}_2$-gauge critical point for lattices of size $L=64$ to $128$, fit $\\tau\\sim cL^z$ with corrections, and compare to the Polyakov-loop result $z=2.55(6)$; an energy-based exponent that disagrees beyond errors would show that the Polyakov loop is not the bulk critical mode, while agreement would confirm the claimed dynamic class. A second check is to repeat the out-of-equilibrium slow-crossing measurement with the same lattice sizes and observables; if it does not converge to $z=2.55(6)$, the equilibrium and nonequilibrium protocols are not probing the same exponent.","supporting_citations":[{"cited_title":"Ben-Av, D","cited_arxiv_id":null,"evidence_quote":"Earlier equilibrium study of the same model whose result $z=2.5(3)$ is compared with the new estimate."},{"cited_title":"Bonati, A","cited_arxiv_id":null,"evidence_quote":"Shows how a stochastic gauge fixing uncovers the nongauge-invariant spin order parameter at DO-O transitions, supporting the LGW$\\times$ classification."}],"review_version":1}