{"id":"ddbbb0c3-efad-49a6-b27d-c3d2bc097291","arxiv_id":"2411.14687","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Bosonic and fermionic free systems show identical statistical distributions and variance scaling for long-time entanglement fluctuations after a quantum quench.","lead":"This paper shows that the long-time entanglement fluctuations of a quenched chain of coupled harmonic oscillators follow the same statistical rules as free-fermion systems, with a sub-Gaussian upper tail and a sub-Gamma lower tail. The result suggests a universal, statistics-independent behavior of entanglement noise in free systems, relevant for quantum simulation and thermalization studies.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The dynamical-to-ensemble reduction hinges on an unproved frequency-incommensurability assumption; near-resonant or commensurate parameters are not covered and could invalidate the claimed parameter-independent universality.","rationale":"The reader's identified weakest assumption is the incommensurability of the normal-mode frequencies, and I agree that this is the most load-bearing point. Section III builds the bridge from deterministic unitary time evolution to random variables on the torus T^N: Eq. (62) writes O(t) = O(omega t), Eq. (64) asserts rational independence, and Eq. (65) uses the ergodic theorem to convert long-time averages into averages over the product measure. Without that conversion, the concentration-of-measure machinery of Sec. IV—modified log-Sobolev inequality, product measure, and tail bounds—has no object to act on. The paper provides no proof that the specific dispersion omega_k = omega sqrt(1 + (4K/omega^2) sin^2(k/2)) is generically incommensurate, nor does it discuss the effect of exact resonances or near-degeneracies. The numerical experiments use a few selected parameter sets and therefore cannot establish the broad 'irrespective of microscopic parameters' statement. This concern does not by itself invalidate the results for generic parameters, so the existing CONDITIONAL verdict is appropriate; the condition should explicitly include a proof or a precise generic assumption for the incommensurability and a quantification of near-resonance effects on the convergence in Eq. (65).","tokens_in":28737,"tokens_out":12056,"duration_ms":138977,"concrete_test":"For L=124 and parameter ratios K/omega^2 in {0.5, 1, 3, and one algebraic value chosen to create a resonance}, compute the N distinct frequencies exactly and test for integer relations sum_k x_k omega_k = 0 using the LLL algorithm. Then generate S(t) for a resonant or near-resonant case over T ~ 10^6 oscillation periods and compare the empirical upper-tail probability P_+(epsilon) with the uniform-torus prediction Eq. (113). If the resonant case deviates measurably from the non-resonant cases, the ergodic equivalence Eq. (65) fails for that parameter and the claimed parameter-independent universality must be restricted.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The entire statistical theory rests on the equivalence between the long-time series O(t) and the torus-averaged quantity O(φ) with φ uniform on T^N. This equivalence is established in Eq. (65) only under the incommensurability condition stated in Eq. (64), which the paper simply assumes after introducing it: 'We shall assume this property in the remainder of this work.' For the oscillator chain the relevant frequencies are omega_k = omega sqrt(1 + (4K/omega^2) sin^2(k/2)), and rational independence of these N = L/2 + 1 numbers is not automatic. Exact resonances occur for special parameter values (e.g., K=0), and near-resonances can make the convergence in Eq. (65) arbitrarily slow, so finite-time numerical time series need not sample the full torus. Since the abstract and Sec. VII claim universality 'irrespective of ... microscopic parameters', the unproved incommensurability assumption is a load-bearing gap: if it fails or is only approximate for some parameters, the mapping to the product measure P and all subsequent concentration results in Sec. IV do not apply.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the long-time entanglement dynamics of a quenched coupled harmonic oscillator chain, treating the time series of entanglement entropy and Rényi entropies as emergent mesoscopic fluctuations. It claims that, for any entanglement probe and any microscopic parameters, the fluctuations obey the same asymmetric distribution found earlier for free fermions: a sub-Gaussian upper tail and a sub-Gamma lower tail, with variance scaling $\\mathrm{Var}(O)=a/L+bL_A^3/L^2$ that becomes universal after rescaling. The argument proceeds by (i) mapping the time evolution to an ergodic orbit on a high-dimensional torus, (ii) importing the concentration-of-measure machinery of the authors' previous work, and (iii) deriving the variance scaling from the Toeplitz structure of the bosonic covariance matrix. The analytical predictions are compared with extensive exact numerics including a data collapse for the variance.","tokens_in":29002,"tokens_out":6754,"duration_ms":69574,"significance":"If correct, the central claim is significant: it would establish a boson-fermion universality in mesoscopic entanglement fluctuations, extending a previously fermionic theory to a canonical continuous-variable bosonic model and showing that the universal statistics are insensitive to particle statistics. The paper is also valuable for its explicit numerical verification: the statistical equivalence in Fig. 3, the proportionality in Fig. 6, the scaling collapses in Fig. 7, and the tabulated concentration parameters in Table I provide reproducible checks of the main formulas. The subject is timely and the claimed universality is a falsifiable prediction that could be tested in finite oscillator chains.","major_comments":[{"comment":"The paper claims universality of the full distribution Irrespective of entanglement probes, but the direct numerical verification of the tail formula (113) is shown only for the entanglement entropy $S$ (Fig. 5). For $S_2$ and $S_3$, the numerics verify only the variance proportionality (Fig. 6) and the variance scaling (Fig. 7), not the asymmetric tail distribution. The concentration parameters for $S_2$ in Table I are close to those for $S$, but the tail distribution for $S_2$ and $S_3$ is not displayed. Given the abstract's claim of probe-independence, a direct numerical test of the P± tails for at least one Rényi entropy would materially strengthen the claim.","section":"Sec. VI and Table I"},{"comment":"The incommensurability condition as written omits $x_0$: it states $x_1=\\dots=x_{N-1}=0$, but the sum in (64) includes $k=0$, so the condition should read $x_0=x_1=\\dots=x_{N-1}=0$. Please correct this typo, which could confuse a reader checking the condition.","section":"Sec. III B, Eq. (64)"},{"comment":"Eq. (111) as printed, $b_+/b_+=c_+/c_+=b_-/b_-=c_-/c_-$, is trivially equal to 1 and cannot be the intended statement; presumably the tilded and untilded constants are being identified. Please rewrite the equation to express the intended relation between the constants in (109) and those in (112).","section":"Sec. IV C, Eq. (111)"},{"comment":"The rescaling statement following Eq. (142) appears dimensionally inconsistent: rescaling $O$ by $\\sqrt{ab}$ and $L,L_A$ by $\\sqrt{a/b}$ does not map (141) into $1/L+L_A^3/L^2$. A correct rescaling would be, for example, $L\\to bL$, $L_A\\to (ab^2)^{1/3}L_A$, and $O\\to O/\\sqrt{a}$ (up to relabeling). Please correct the stated factors.","section":"Sec. V D, Eq. (142)"},{"comment":"There are several typos that should be fixed: 'incommensurality' in Sec. III B; 'eigenvlaues' in Sec. II D; 'Heinsenberg' in Appendix A; and 'suppressed notion' in Appendix E. In Fig. 5(b), the explicit fitting functions in the inset are not described in the caption or text; please list the fitted forms used for the sub-Gaussian and sub-Gamma curves.","section":"Throughout"}],"minor_comments":[],"recommendation":"major_revision","confidential_remarks":"The paper is a direct continuation of the authors' own previous work [32], and a substantial part of the analytical machinery (the modified log-Sobolev approach, the variance formula, and the use of concentration bounds) is imported from that paper. The new element is the bosonic model and the claim of boson-fermion universality. The refereeing should weigh whether the new contribution, as currently presented, is sufficiently independent and rigorous: the main gaps are the unproved incommensurability assumption and the heuristic upgrade of inequalities to equalities, both of which are central to the universality statement. If these are fixed or explicitly flagged as conjectural, the paper could be a solid contribution; in its present form, it overstates the rigor of the derivation."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's my take on arXiv:2411.14687. The genuinely new thing is the first analytic treatment of long-time entanglement fluctuation statistics for a harmonic oscillator chain, and the claim that the asymmetric sub-Gaussian/sub-Gamma tail distribution and the 1/L to L_A^3/L^2 variance crossover match the free-fermion result from Ref. [32]. The bosonic calculation is not a trivial transcription: the symplectic eigenvalue structure, the block-Toeplitz C matrix, and the variance derivation are all worked out separately for bosons. The numerical verification is extensive — variance data across L up to 10^8, clean data collapse — so the central scaling law is well supported. I buy the qualitative result for generic parameters.\n\nThe main soft spot is the one the stress-test flags: the reduction from the time series O(t) to a uniform average over the torus rests entirely on the incommensurability of the N normal-mode frequencies, stated in Eq. (64) and then simply assumed ('We shall assume this property...'). That is load-bearing, and the paper never proves it for the harmonic chain's dispersion, nor does it discuss what happens at or near resonances (e.g., K=0 or special ratios of ω and K). The abstract's 'irrespective of microscopic parameters' is too strong as written; the claim should be qualified to generic parameters. This is a correctable gap, not a fatal flaw, but it needs to be addressed.\n\nThe other heuristics — upgrading the modified log-Sobolev inequality to an approximate equality with numerically fixed b±, c±, and the uncorrelatedness assumption in Appendix E — are honest but uncontrolled. They make the derivative sharp-tail part phenomenological. That's consistent with the physics goal, but 'strictly the same' in the abstract oversells it. Also, no code or data are shipped.\n\nFor me, the paper deserves a serious referee. The result is plausible, the numerics are strong, and the boson-fermion universality, if correct, is a nice broadening of the prior fermionic theory. I would send it to review, and I'd tell the referees to focus their main energy on the incommensurability assumption and on getting the parameter-exception statement right. I'd probably cite it if I was working on entanglement dynamics in free systems, though with a caveat about the gap.","headline":"Solid bosonic extension with convincing numerics, but the ergodic reduction leans on an unproved incommensurability assumption and the 'universality' claim needs a generic-parameter qualifier.","tokens_in":29473,"tokens_out":4263,"would_cite":true,"duration_ms":39022,"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 argues that after a quench, the long-time entanglement entropy and Rényi entropies of a coupled harmonic oscillator chain fluctuate with exactly the same universal distribution and variance scaling as free fermions.","keywords":["entanglement entropy","Rényi entropy","quench dynamics","harmonic oscillator chain","mesoscopic fluctuations","concentration of measure","boson-fermion universality","entanglement fluctuations"],"falsifier":"Take a chain length where an integer relation or a near-integer relation exists among the frequencies $\\omega_k=\\omega\\sqrt{1+(4K/\\omega^2)\\sin^2(k/2)}$, for instance by tuning $K/\\omega^2$ so that several $s_k$ coincide rationally, and simulate $S(t)$ over times long compared with the revival scale. If the empirical upper and lower deviation probabilities deviate from Eq. (113), for example the upper tail becomes Gaussian or sub-exponential or the variance crossover shifts, then the assumed ergodicity on $\\mathbb{T}^N$ fails and the claimed universality is not as general as stated.","tokens_in":28537,"feed_emoji":"⚛️","tokens_out":9346,"duration_ms":85395,"temperature":0.7,"pith_summary":"This paper sets out to show that the long-time entanglement fluctuations of a quenched chain of coupled harmonic oscillators, a canonical bosonic many-body model, are statistically identical to those of the free-fermion models studied earlier by the authors. The claimed fluctuation law is asymmetric and universal: upward deviations of the entanglement entropy or any Rényi entropy decay as $e^{-\\epsilon^2/(2b_+)}$ (sub-Gaussian), downward deviations as $e^{-\\epsilon^2/(2(b_-+c\\epsilon))}$ (sub-Gamma), and the variance obeys the rescaled law $\\mathrm{Var}(O)=1/L+L_A^3/L^2$ with a crossover at $L_A\\sim L^{1/3}$. If true, the result establishes boson-fermion universality: particle statistics drops out of the fluctuation statistics, which instead is governed by the product structure of random phases and by concentration-of-measure phenomena. The paper derives the law analytically from the Gaussianity of the evolving state and verifies it numerically over a wide range of system sizes.","feed_headline":"Entanglement fluctuations are identical for bosons and fermions","feed_subtitle":"Quenched chains and free fermions share same asymmetric tails and universal variance law, hiding particle statistics.","key_machinery":"The engine of the argument is the map from time evolution to uniform sampling of an $N$-dimensional torus, together with the modified logarithmic Sobolev inequality for product probability measures, a concentration inequality for functions of many independent random variables. The covariance matrix $C(t)=C_0+C_1(t)$ is block-Toeplitz, with $2\\times 2$ blocks depending only on the oscillator displacement, and because the normal-mode frequencies are assumed incommensurate, the trajectory $\\varphi(t)=\\omega t$ is dense on $\\mathbb{T}^N$. This makes the long-time statistics of $O(t)$ equal to the statistics of $\\tilde O(\\varphi)=\\frac14\\,\\mathrm{Tr}_A\\,h(\\tilde C(\\varphi))$ with $\\varphi$ uniform; since $\\tilde O$ is highly nonlinear in $\\varphi$, the logarithmic Sobolev inequality bounds the moment-generating function and yields the asymmetric sub-Gaussian/sub-Gamma tails. The variance follows from the identity $\\mathrm{Var}(O)=\\frac12\\langle|\\partial_\\varphi O|^2\\rangle$, whose leading terms split into a subsystem-edge contribution $\\sim 1/L$ and a bulk contribution $\\sim L_A^3/L^2$.","core_discovery":"The central discovery is that, after a global quench, the persistent temporal fluctuations of entanglement in a finite harmonic oscillator chain are statistically the same as sample-to-sample fluctuations of disorder ensembles in mesoscopic physics, and strictly the same for bosons as for fermions. Concretely, for the entanglement entropy $S$ and Rényi entropies $S_n$, the probability $P(|O-\\langle O\\rangle|\\ge \\epsilon)$ has a sub-Gaussian upper tail $e^{-\\epsilon^2/(2b_+)}$ and a sub-Gamma lower tail $e^{-\\epsilon^2/(2(b_-+c\\epsilon))}$, independent of the probe and of the chain's microscopic parameters after rescaling; the variance satisfies $\\mathrm{Var}(O)\\sim 1/L$ for small subsystems and $\\sim L_A^3/L^2$ for larger ones, crossing at $L_A\\sim L^{1/3}$. The authors attribute the universality to two shared structures: the entanglement probe is a nonlinear function of a block-Toeplitz covariance matrix whose time dependence enters only through $N=L/2+1$ phases, and those phases uniformly sample an $N$-dimensional torus, so the product probability measure triggers concentration of measure.","pith_inferences":["Editorial extension: one can test the authors' universality in higher-dimensional harmonic lattices or long-range coupled chains; the derivation only uses Toeplitz structure and an incommensurate spectrum, so the same tails should appear, though the paper does not compute them.","Editorial extension: because the fluctuation amplitude vanishes as $1/\\sqrt{L}$, the law sets a quantitative reproducibility floor for entanglement-based quantum simulators: at fixed subsystem fraction the run-to-run spread shrinks only as a power of system size.","Editorial extension: the negative answer to whether particle statistics can be read off from temporal entanglement noise suggests that detecting statistics requires probes sensitive to the sign or spectral location of the covariance spectrum, such as entanglement negativity or occupation-number full counting, rather than entropy-type functionals.","Editorial extension: a concrete experimental protocol implied but not developed by the paper is to record $S(t)$ over many revival periods in a trapped-ion or optical-lattice chain with deliberately incommensurate normal-mode frequencies and compare the empirical tail ratio to Eq. (113); the paper's own numerics go up to $L=10^8$ for the scaling check, not to a physical device."],"forward_implications":["Free-boson and free-fermion systems in the same quench setting cannot be distinguished by the statistics of entropy or Rényi fluctuations: both give the same asymmetric tail law.","The variance scaling law $\\mathrm{Var}(O)=1/L+L_A^3/L^2$ is universal after rescaling, so measurements of fluctuation size can determine the subsystem and total sizes but not the microscopic coupling or the probe.","The fluctuation regime is a distinct long-time stage of entanglement dynamics, after linear growth and revivals, and in finite systems it persists as reproducible quasi-periodic oscillations rather than decaying to a thermal value.","The same concentration-of-measure argument applies to any Gaussian initial state of the chain, not only the vacuum, so the universality is expected to survive other Gaussian preparations.","The paper's conjecture extends the universality to anyonic and supersymmetric systems, since the two minimal ingredients, a product probability measure and a nonlinear probe, are not statistics-specific."],"supporting_citations":[{"why":"Establishes the long-time entanglement fluctuation theory for free-fermion models and spin chains that this paper generalizes to bosons and uses as the comparison basis.","marker":"[32]"},{"why":"Supplies the modified logarithmic Sobolev inequality and the nonasymptotic concentration framework that produce the sub-Gaussian and sub-Gamma tail bounds.","marker":"[33]"},{"why":"Provides the Calabrese-Cardy description of early-time entanglement growth and saturation that defines the regime the paper contrasts with the long-time fluctuation stage.","marker":"[21]"},{"why":"Williamson's symplectic normal form is used to diagonalize the covariance matrix and to express entanglement entropy and Rényi entropies as matrix functionals of $C(t)$.","marker":"[49]"},{"why":"Gives the quasi-periodicity result for incommensurate frequency vectors, from which the persistent reproducible fluctuations follow.","marker":"[57]"},{"why":"Ergodic theorem on the torus converts the time series $O(t)$ into uniform sampling of the phase ensemble, the key statistical equivalence.","marker":"[58]"},{"why":"Define the mesoscopic sample-to-sample fluctuation paradigm that the emergent disorder ensemble is identified with.","marker":"[34, 35]"}],"fun_headline_variants":["Boson and fermion entanglement fluctuations share identical statistics","Entanglement noise is universal: bosons and fermions alike","Sub-Gaussian tails prove boson-fermion entanglement universality","Same fluctuation laws for boson and fermion entanglement","Particle statistics irrelevant for entanglement fluctuations"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the $N=L/2+1$ normal-mode frequencies of the quenched chain are incommensurate, so the phase trajectory densely covers the torus; the paper assumes this after Eq. (64) and does not prove it for the specific dispersion $\\omega_k=\\omega\\sqrt{1+(4K/\\omega^2)\\sin^2(k/2)}$ or bound the effect of near-resonances.","fun_headline_variants_meta":{"raw":{"variants":["Boson and fermion entanglement fluctuations share identical statistics","Entanglement noise is universal: bosons and fermions alike","Sub-Gaussian tails prove boson-fermion entanglement universality","Same fluctuation laws for boson and fermion entanglement","Particle statistics irrelevant for entanglement fluctuations"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000289,"raw_usage":{"total_tokens":1751,"prompt_tokens":1062,"completion_tokens":689,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":678,"completion_tokens_details":{"reasoning_tokens":611}},"tokens_in":678,"tokens_out":689,"duration_ms":6476,"temperature":1.0,"reasoning_tokens":611,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T15:00:56.075045+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a chain length where an integer relation or a near-integer relation exists among the frequencies $\\omega_k=\\omega\\sqrt{1+(4K/\\omega^2)\\sin^2(k/2)}$, for instance by tuning $K/\\omega^2$ so that several $s_k$ coincide rationally, and simulate $S(t)$ over times long compared with the revival scale. If the empirical upper and lower deviation probabilities deviate from Eq. (113), for example the upper tail becomes Gaussian or sub-exponential or the variance crossover shifts, then the assumed ergodicity on $\\mathbb{T}^N$ fails and the claimed universality is not as general as stated.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the modified logarithmic Sobolev inequality and the nonasymptotic concentration framework that produce the sub-Gaussian and sub-Gamma tail bounds."},{"cited_title":"Botero and B","cited_arxiv_id":null,"evidence_quote":"Williamson's symplectic normal form is used to diagonalize the covariance matrix and to express entanglement entropy and Rényi entropies as matrix functionals of $C(t)$."},{"cited_title":"Thus the persistent fluctuations displayed in long-time entangle- ment fluctuations, exemplified in Fig","cited_arxiv_id":null,"evidence_quote":"Gives the quasi-periodicity result for incommensurate frequency vectors, from which the persistent reproducible fluctuations follow."},{"cited_title":"Williamson, On the algebraic problem concerning the 23 normal forms of linear dynamical systems, Amer","cited_arxiv_id":null,"evidence_quote":"Ergodic theorem on the torus converts the time series $O(t)$ into uniform sampling of the phase ensemble, the key statistical equivalence."}],"review_version":1}