{"id":"fa486a71-5d3f-4496-adae-c9f761c6e199","arxiv_id":"2501.04397","paper_version":2,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Lyapunov exponents of soft SU(2) gluon modes give a thermalization time of about 0.5 fm/c at 600 MeV and a maximum of chaos at the deconfinement temperature.","lead":"This paper uses computer simulations of SU(2) gauge theory to measure how quickly its soft gluon modes become chaotic and thermalize, estimating about 0.5 fm/c at 600 MeV. The result matters because it connects chaos in gauge theory to the thermalization time of quark-gluon plasma in heavy-ion collisions.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Eq. 15 infers the full Lyapunov spectrum from a velocity-dependent ansatz; an O(1) error in the KS rate shifts t_th directly.","rationale":"The reader already identified constancy of the KS entropy rate as the weakest assumption. I focus on a prior step: whether Eq. (15) is actually a measurement of the KS entropy rate. The full Lyapunov spectrum is not extracted anywhere in Sec. 4; only lambda_L is. The factor 4 is assembled from a 1994 result and an assumed lambda(v) spectrum. The OTOC lambda(v) is a different object from the ordered Lyapunov spectrum, so the Pesin entropy is not directly computed. The numerical time estimate inherits this systematic uncertainty linearly. The paper's measured lambda_L(T) values and the thermal/non-thermal comparison are credible and cross-checked, and the constant-rate assumption has partial support from Fig. 3's non-thermal datapoint lying on the same curve; hence the paper should neither be rejected nor accepted as-is. It should be conditional on a direct spectrum computation, which is what the reader's condition already implies.","tokens_in":12818,"tokens_out":8180,"duration_ms":78204,"concrete_test":"Recompute the full Lyapunov spectrum for the same thermal soft-mode configurations used for Fig. 3 (N=32, T/Tc=4 and 6, Langevin-thermalized and then evolved with noise/damping switched off) using a standard Benettin/QR tangent-space method; sum the positive exponents and compare with 4 lambda_L N^3 and with the lambda(v) ansatz. If the sum differs by more than about 30%, or if the positive fraction deviates substantially from 1/3, Eq. 15 and the inferred t_th must be revised.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Equation (15) sets the KS entropy rate to -sum_i lambda_i a / N^3 approx -4 lambda_L a, where the factor 4 is obtained by assuming that about 1/3 of the 18N^3 Lyapunov exponents are positive (Ref. [12]) and that their values follow lambda(v) approx lambda_L [1 - (v/v_B)^2] with v uniformly distributed in [0, v_B]. The paper measures only the maximal exponent lambda_L from d(t) and OTOCs; it never directly computes the full Lyapunov spectrum. The velocity-dependent exponent lambda(v) of an OTOC is the growth rate of a localized perturbation moving at velocity v, not the ordered set of phase-space Lyapunov exponents entering Pesin's formula. Using one to estimate the other is a nontrivial identification that is not justified in the text. Any O(1) correction to the factor 4 changes the headline t_th = 0.8/lambda_L = 0.50(3) fm/c linearly. Sec. 6's claim that the only assumption is energy conservation is therefore inaccurate: the estimate also assumes a particular shape and fraction of the positive Lyapunov spectrum, plus a constant rate along the non-equilibrium trajectory.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript studies classical-statistical SU(2) lattice gauge theory in and out of equilibrium and reports measurements of the maximal Lyapunov exponent lambda_L of soft gluonic modes. In the thermal case, lambda_L is extracted from the gauge-invariant plaquette distance for temperatures in the 0.6-3 GeV range and found to obey lambda_L ~ 0.52 T; in the non-thermal case, an over-occupied initial condition in the self-similar scaling regime gives lambda_L = 0.66 Q_s, which is compatible with the thermal trend when the temperature is defined through the energy density. The authors also compute OTOCs for a Z_2 scalar theory near the deconfinement transition and report that lambda_L maximizes at T_c. Using Eq. (15), they convert a conjectured Kolmogorov-Sinai entropy rate of the soft modes into a thermalization time t_th ~ 0.8/lambda_L ~ 0.50(3) fm/c at T ~ 600 MeV and 0.70(5) fm/c at T ~ 450 MeV.","tokens_in":13063,"tokens_out":7409,"duration_ms":75715,"significance":"If the central claim holds, the paper offers a non-perturbative, lattice-based route from Lyapunov spectra to a thermalization time in a gauge theory, with a concrete number relevant for heavy-ion phenomenology. The measurements of lambda_L in over-occupied SU(2) and in the Z_2 critical theory are useful additions to the existing literature, and the benchmark lambda_L/T ~ 0.52 below the MSS bound, together with consistency with lambda_L ~ g^2 E/6, lends credibility to the numerical method. However, the quantitative thermalization-time claim rests on an unverified identification between OTOC velocity-dependent exponents and the phase-space Lyapunov spectrum entering Pesin's formula, and on the assumption that the KS rate remains constant along the non-equilibrium trajectory. These gaps make the result interesting but not yet established at the level claimed in the abstract.","major_comments":[{"comment":"The KS entropy rate is not measured directly; it is inferred by multiplying lambda_L by a factor 4. This factor assumes that about 1/3 of the 18N^3 Lyapunov exponents are positive (citing Ref. [12]) and that the positive exponents follow lambda(v) = lambda_L[1 - (v/v_B)^2] with v uniformly distributed in [0, v_B]. The paper measures only the maximal exponent lambda_L from d(t) and from OTOCs; it never computes the ordered phase-space Lyapunov spectrum. The velocity-dependent exponent lambda(v) obtained from an OTOC butterfly cone is the growth rate of a localized perturbation, not the ordered set of exponents entering the Pesin identity. An O(1) uncertainty in the factor 4 propagates linearly into the headline t_th = 0.8/lambda_L = 0.50(3) fm/c. The authors should either compute the full spectrum and evaluate the sum in Eq. (15) explicitly, or present a quantitative systematic-error estimate for the factor 4.","section":"Sec. 5, Eq. (15)"},{"comment":"The estimate assumes that the KS entropy production rate remains at its final-thermal-state value while the system evolves from the over-occupied self-similar attractor to the equilibrium state. The text states in Sec. 6 that 'the only assumption that goes into our calculation is the conservation of energy density of the soft modes during the entire evolution', but Eq. (15) already contains the spectrum-shape and positivity-fraction assumptions, and the constancy of dot_s_KS along the non-equilibrium trajectory is an additional assumption for which no evidence is provided. This should be stated explicitly and its effect on t_th should be assessed, or the thermalization-time claim should be correspondingly weakened.","section":"Sec. 5 and Sec. 6"},{"comment":"The abstract and Sec. 5 claim that 'spectra of positive Lyapunov exponents is observed' and that the authors 'extract the spectra of positive Lyapunov exponents'. The manuscript presents measurements of the maximal exponent only; no full spectrum is shown or computed. This overstatement is connected to the central claim because Eq. (15) is precisely the step where the full spectrum is replaced by an ansatz. Please correct the wording and distinguish what is measured from what is modeled.","section":"Abstract and Sec. 5"}],"minor_comments":[{"comment":"The word 'precession' should be 'precision' in the sentence describing the Gauss-law constraint.","section":"Sec. 3.1"},{"comment":"The expression for the energy density, '6/N^3 sum_k |k| aT/|k|', is confusing because the |k| factors cancel; please rewrite it as the intended sum over oscillators or explain the notation.","section":"Sec. 3.2"},{"comment":"The KS entropy rate is defined with a minus sign in Eq. (15), whereas the standard KS entropy rate is the (positive) sum of the positive Lyapunov exponents. Please clarify the sign convention and how the integration leading to t_th is performed.","section":"Sec. 5, Eq. (15)"},{"comment":"Reference [8] is titled 'Viscosity, black holes, and quantum field theory', which appears to be an incorrect title for the Kolmogorov-Sinai entropy work; please verify and correct this reference.","section":"References"},{"comment":"For the Z_2 OTOC results it would be helpful to state the lattice size, the number of thermal configurations, and how the quoted scaling exponents and their uncertainties were obtained.","section":"Sec. 4"}],"recommendation":"major_revision","confidential_remarks":"The stress-test concern lands: Eq. (15) is the load-bearing step and it is not justified by the data shown. The paper is otherwise sound in its lambda_L measurements and the Z_2 near-T_c analysis is a nice contribution. I would not reject the paper, but the quantitative thermalization-time claim should be revised to either include a direct Lyapunov-spectrum computation or carry a prominent, quantified caveat about the factor 4 and the constancy of the KS rate. The abstract should also be aligned with what is actually measured."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: the λ_L measurements are credible and cross-checked; the headline thermalization time is a reasonable estimate but it is built on an inferred Lyapunov spectrum, and the paper overstates what it assumes. Still worth refereeing.\n\nWhat is new: this is the first estimate of a thermalization time for soft SU(2) modes from the Kolmogorov-Sinai entropy rate, and the Z2 OTOC at the critical point is a clean use of universality, even if similar peaks in λ_L at phase transitions were seen earlier in spin and scalar systems. The T >> T_c result λ_L/T ~ 0.52 matches previous independent calculations, and the non-equilibrium λ_L sits on the same temperature curve, a sensible consistency check.\n\nThe soft spots are in Sec. 5. Equation (15) is not a measurement of the KS rate; it is an estimate obtained by assuming one-third of the 18N^3 Lyapunov exponents are positive and that λ(v) = λ_L (1 - (v/v_B)^2) with v uniformly distributed. The paper measures only λ_L, so the factor 4 is an ansatz, and any O(1) error goes linearly into t_th = 0.8/λ_L. The abstract's 'spectra of positive Lyapunov exponents' is similarly a bit stronger than what is actually computed. The statement that 'the only assumption is conservation of energy density' is inaccurate: you also need the final-state spectrum to represent the entire trajectory and the spectral shape. These assumptions are plausible but should be stated honestly.\n\nSecondary issues: no code or data is provided, and the Z2 critical study is at a single lattice spacing, so continuum scaling is not demonstrated. These are minor relative to the main claim.\n\nNet: the paper deserves a serious referee. The core measurements are reproducible in principle and cross-checked, and the thermalization time is a physically interesting estimate for heavy-ion phenomenology. I would encourage a revision that either measures the Lyapunov spectrum or explicitly labels Eq. (15) as a model-dependent estimate, and drops the 'only assumption' language. I would cite the λ_L/T value with confidence, but the thermalization time with a caveat.","headline":"Solid λ_L measurements for SU(2) soft modes, but the headline thermalization time rests on an inferred Lyapunov spectrum and an overstated 'only assumption' claim.","tokens_in":13617,"tokens_out":4641,"would_cite":true,"duration_ms":39986,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["81T25","81T13","37D45","81T80"],"pacs":["11.15.Ha","12.38.Mh","05.45.-a"],"model":"deepseek-v4-flash","headline":"Chaos in the soft gluon modes of SU(2) gauge theory sets the thermalization time to about 0.50(3) fm/c at $T \\sim 600$ MeV.","keywords":["thermalization","Lyapunov exponent","SU(2) gauge theory","soft gluon modes","Kolmogorov-Sinai entropy","out-of-time-ordered correlator","lattice gauge theory","quark-gluon plasma"],"falsifier":"Along a non-equilibrium trajectory, compute the maximal Lyapunov exponent $\\lambda_L(t)$ at several times before equilibrium; if it differs by more than the quoted uncertainty from the thermal-state value used in $t_{\\rm th} = 0.8/\\lambda_L(T)$, the constant-rate integration fails. Alternatively, compute the thermodynamic entropy density $s(t)$ directly from the evolving gluon distribution $f(p,t)$ and compare $\\dot{s}(t)$ with the Kolmogorov-Sinai entropy rate; a mismatch would invalidate the identification.","tokens_in":12611,"feed_emoji":"⚛️","tokens_out":10239,"duration_ms":83532,"temperature":0.7,"pith_summary":"This paper tries to establish that the thermalization of a non-Abelian gauge plasma is governed by the chaotic dynamics of its infrared soft (magnetic) modes, and that a thermalization time can be derived from their Lyapunov spectrum. In lattice simulations of SU(2) gauge theory, the authors find positive maximal Lyapunov exponents in both thermal equilibrium and an over-occupied non-thermal self-similar state, which indicates phase-space mixing. Using the Kolmogorov-Sinai entropy rate obtained from the positive Lyapunov exponents, they estimate $t_{\\rm th} = 0.8/\\lambda_L(T)$, about 0.50(3) fm/c at $T \\sim 600$ MeV and 0.70(5) fm/c at $T \\sim 450$ MeV. Near the deconfinement transition, the maximal Lyapunov exponent of the long-wavelength critical modes peaks at $T_c$, as computed through a $Z_2$ scalar field theory in the same universality class. If correct, this connects a microscopic chaos measure to the early-time hydrodynamization of the quark-gluon plasma.","feed_headline":"Soft gluon chaos sets thermalization at 0.50 fm/c","feed_subtitle":"Lattice-computed Lyapunov exponents yield a first-principles time scale for forming the quark-gluon plasma at 600 MeV.","key_machinery":"The load-bearing machinery is the Lyapunov spectrum of the soft modes. In the thermal regime, the dynamics of magnetic gluons is described by an effective Langevin equation in which hard modes act as a heat bath; two nearby gauge configurations are evolved with noise and damping switched off, and the maximal Lyapunov exponent is extracted from the gauge-invariant separation $d(t) = (1/2N_p)\\sum_P |\\mathrm{tr}\\,U_P(t) - \\mathrm{tr}\\,U'_P(t)| \\sim e^{\\lambda_L t}$. For the non-thermal state, classical-statistical evolution starts from an over-occupied gluon distribution and the exponent is measured inside the self-similar scaling regime. The Kolmogorov-Sinai entropy rate is then obtained through the Pesin identity as the sum of all positive Lyapunov exponents, and the thermalization time follows from $t_{\\rm th} = 0.8/\\lambda_L(T)$. Near $T_c$, the critical-mode Lyapunov exponent is extracted from the out-of-time-ordered correlator of a $Z_2$ scalar field, justified by the shared universality class with SU(2) deconfinement.","core_discovery":"The central claim is that the soft gluon modes of SU(2) gauge theory form a chaotic dynamical system whose positive Lyapunov exponents quantify how fast information about the initial state is lost. In thermal equilibrium at high temperature, the maximal Lyapunov exponent grows linearly with temperature, $\\lambda_L/T \\approx 0.52$, and the measured values respect the conjectured bound $\\lambda_L \\le 2\\pi T$. In a non-thermal over-occupied state in the self-similar scaling regime, $\\lambda_L$ is comparable in magnitude to that of a thermal state with the same energy density. Summing the positive Lyapunov exponents gives the Kolmogorov-Sinai entropy rate, and integrating the entropy difference between the non-thermal and thermal states yields the thermalization time $t_{\\rm th} = 0.8/\\lambda_L(T)$, about 0.50(3) fm/c at $T \\sim 600$ MeV. Near the deconfinement transition, the maximal Lyapunov exponent of critical modes, obtained from the out-of-time-ordered correlator of a classical $Z_2$ scalar field theory, reaches its maximum at $T_c$, with temperature power laws $T^{6.8(7)}$ below and $T^{-2.8(3)}$ above.","pith_inferences":["A direct numerical test would be to compute $\\lambda_L(t)$ at several times along the non-equilibrium trajectory; if it drifts away from the thermal-state value before equilibrium, the constant-rate integration underlying $t_{\\rm th}$ breaks down.","Because the near-$T_c$ analysis relies on the $Z_2$ universality class, repeating the out-of-time-ordered-correlator measurement in SU(3) near its deconfinement transition would show whether the $\\lambda_L$ peak at $T_c$ is generic or specific to SU(2).","The entropy production in the paper is information-theoretic (Kolmogorov-Sinai entropy); comparing it with the thermodynamic entropy computed from the evolving distribution $f(p,t)$ would connect the chaos measure to a measurable entropy current.","The $0.5$ fm/c estimate is for a static, non-expanding box; applying the same method to a Bjorken-expanding glasma geometry with a time-dependent effective temperature could shift the estimate toward the range inferred from heavy-ion data."],"forward_implications":["The soft modes of SU(2) gauge theory are chaotic both in thermal equilibrium and in the non-thermal self-similar state, so thermalization of the infrared sector can be viewed as chaotic phase-space mixing rather than purely perturbative scattering.","Starting from an over-occupied gluon state, a thermal state at $T \\sim 600$ MeV is reached in about $0.50(3)$ fm/c, and one at $T \\sim 450$ MeV in about $0.70(5)$ fm/c.","At high temperatures the maximal Lyapunov exponent behaves as $\\lambda_L/T \\approx 0.52$, consistent with the bound $\\lambda_L \\le 2\\pi T$; the butterfly velocity is about $0.8c$ and the configuration-space diffusion coefficient falls roughly as $T^{-1}$.","Near deconfinement, $\\lambda_L$ of the critical modes is maximal at $T_c$, so phase-space spreading is strongest at the transition; the diffusion coefficient $D$ is temperature-independent below $T_c$ and decreases sharply above it.","The obtained thermalization time is shorter than perturbative bottom-up estimates ($\\gtrsim 2.5$ fm/c) and consistent with the early hydrodynamization time inferred in heavy-ion collisions."],"supporting_citations":[{"why":"Supplies the gauge-invariant distance method for extracting $\\lambda_L$ and the scale-matching estimate for the thermalization time.","marker":"[15]"},{"why":"Defines the over-occupied self-similar attractor regime used as the non-thermal initial state.","marker":"[14]"},{"why":"Provides the fraction of positive Lyapunov exponents (about 1/3) used in the Kolmogorov-Sinai entropy density estimate.","marker":"[12]"},{"why":"States the Pesin identity linking the Kolmogorov-Sinai entropy rate to the sum of positive Lyapunov exponents.","marker":"[9]"},{"why":"Gives the effective stochastic evolution of soft non-Abelian fields that the thermal Langevin algorithm implements.","marker":"[28]"},{"why":"Provides the perturbative color conductivity used to set the damping strength in the thermal soft-mode evolution.","marker":"[29]"},{"why":"Establishes the $Z_2$/3D-Ising universality class of SU(2) deconfinement, justifying the scalar-field proxy for critical modes.","marker":"[31]"},{"why":"Sets the chaos bound $\\lambda_L \\le 2\\pi T$ against which the measured high-temperature Lyapunov exponents are checked.","marker":"[33]"}],"fun_headline_variants":["Soft gluon chaos yields 0.50 fm/c thermalization time","Chaotic soft modes set quark-gluon plasma timescale","First Lyapunov measurement near deconfinement transition","SU(2) chaos: from soft modes to thermalization time","Lyapunov spectrum gives 0.50 fm/c for QGP thermalization"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the Kolmogorov-Sinai entropy rate computed from the positive Lyapunov exponents of the final thermal state equals the physical entropy production rate of the soft modes throughout the non-equilibrium evolution, so that one can integrate this single rate from the over-occupied attractor to equilibrium.","fun_headline_variants_meta":{"raw":{"variants":["Soft gluon chaos yields 0.50 fm/c thermalization time","Chaotic soft modes set quark-gluon plasma timescale","First Lyapunov measurement near deconfinement transition","SU(2) chaos: from soft modes to thermalization time","Lyapunov spectrum gives 0.50 fm/c for QGP thermalization"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000526,"raw_usage":{"total_tokens":2602,"prompt_tokens":1069,"completion_tokens":1533,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":685,"completion_tokens_details":{"reasoning_tokens":1451}},"tokens_in":685,"tokens_out":1533,"duration_ms":11857,"temperature":1.0,"reasoning_tokens":1451,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T21:34:22.465953+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Along a non-equilibrium trajectory, compute the maximal Lyapunov exponent $\\lambda_L(t)$ at several times before equilibrium; if it differs by more than the quoted uncertainty from the thermal-state value used in $t_{\\rm th} = 0.8/\\lambda_L(T)$, the constant-rate integration fails. Alternatively, compute the thermodynamic entropy density $s(t)$ directly from the evolving gluon distribution $f(p,t)$ and compare $\\dot{s}(t)$ with the Kolmogorov-Sinai entropy rate; a mismatch would invalidate the identification.","supporting_citations":[{"cited_title":"Gong, Lyapunov spectra in su (2) lattice gauge theory, Physical Review D 49 (5) (1994) 2642","cited_arxiv_id":null,"evidence_quote":"Provides the fraction of positive Lyapunov exponents (about 1/3) used in the Kolmogorov-Sinai entropy density estimate."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"States the Pesin identity linking the Kolmogorov-Sinai entropy rate to the sum of positive Lyapunov exponents."},{"cited_title":"Bodeker, On the e ffective dynamics of soft nonAbelian gauge fields at finite temperature, Phys","cited_arxiv_id":null,"evidence_quote":"Gives the effective stochastic evolution of soft non-Abelian fields that the thermal Langevin algorithm implements."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the perturbative color conductivity used to set the damping strength in the thermal soft-mode evolution."},{"cited_title":"Svetitsky, L","cited_arxiv_id":null,"evidence_quote":"Establishes the $Z_2$/3D-Ising universality class of SU(2) deconfinement, justifying the scalar-field proxy for critical modes."},{"cited_title":"Maldacena, S","cited_arxiv_id":null,"evidence_quote":"Sets the chaos bound $\\lambda_L \\le 2\\pi T$ against which the measured high-temperature Lyapunov exponents are checked."}],"review_version":1}