{"id":"892fb6c2-4841-42ab-bfd0-f1a962553f03","arxiv_id":"2412.02887","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The steady-state phase probabilities of a parametric oscillator equal the integrated, noise-smoothed quadrature distribution of the initial quantum state.","lead":"This paper shows that in a pumped parametric oscillator the probability of ending in one of two stable output phases is set by the initial quantum state's quadrature distribution, smoothed by a pump-dependent noise width. The result turns a macroscopic bistable measurement into a quantum state probe and is proposed for both optical and superconducting Josephson implementations.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Eq. (3) is presented as Gaussian smoothing for the Q representation, but for xi=1 the variance is (2-lambda)/(2(lambda-1)), which is negative for lambda>2; the Q-function reconstruction claim is therefore undefined unless the paper restricts to lambda<=2 or provides a Wigner-based protocol.","rationale":"The reader's weakest assumption focuses on the early-time linearization and dropping the g^2(a†a)a term. That is a genuine concern, but the simulations in Figs. 2 and 3 provide support within the low-photon regime, and the paper explicitly assumes g^2 << lambda-1. The lambda>2 issue is sharper: it is visible directly from Eq. (3) and requires no parameter estimation. A Gaussian smoothing with negative variance is unphysical; therefore the claim as written cannot hold for the Q representation above lambda=2. This does not invalidate the core physics for lambda<=2 or for Wigner/P representations, but it means the 'arbitrary initial states' scope and the Q-function tomography claim are over-broad. The verdict should remain CONDITIONAL, with the condition that the paper either state the lambda<=2 restriction for the Q result or extend the protocol to Wigner for lambda>2. My agreement with the reader is partial: the reader noted the lambda<=2 validity issue in the rationale but did not make it the weakest assumption.","tokens_in":10278,"tokens_out":12879,"duration_ms":131867,"concrete_test":"Simulate the full Eq. (1) master equation for a vacuum initial state at lambda=2.5 with a truncated Hilbert space, and extract p(alpha(1),b) for several b. Compare this to Eq. (3) evaluated with xi=1: because sigma^2<0, the integral is not a real Gaussian convolution, and with xi=0 (Wigner). If the full numerics match the Wigner-marginal formula but not the Q-marginal formula, the missing lambda<=2 restriction is confirmed. Run lambda=2.0 and lambda=1.5 as controls to verify the simulation pipeline.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central formula in Eq. (3), and the subsequent claim that rotating the pump phase reconstructs the initial Q-function, are stated without a validity range. For the Husimi Q representation (xi=1) the Gaussian variance in Eq. (3) is sigma^2=(2-lambda)/(2(lambda-1)). This is positive only for 1<lambda<2, vanishes at lambda=2, and is negative for every lambda>2. The paper never mentions this bound; Fig. 3 only demonstrates lambda=1.2 and lambda=2.0, and the text refers generically to arbitrary initial states and to the steady-state statistics being governed by parametric gain. If lambda>2, the Q-function is not obtained by Gaussian smoothing, so the proposed measurement of p(b) does not directly yield a Q-marginal. One could instead use the Wigner representation (xi=0), whose marginal is still a valid quadrature distribution, but that replacement and its deconvolution protocol are not described. Since the strongest claim is the map from p(b) to the initial Q-function, the missing lambda<=2 condition is load-bearing for the stated scope.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper analyzes a degenerate optical parametric oscillator (OPO) initialized in an arbitrary quantum state and claims that the steady-state probabilities of the two bistable phases are determined by the early-time linearized dynamics. The central formula, Eq. (3), expresses the probability p(α(1), b) as the integral over the right half of the initial X-quadrature marginal convolved with a Gaussian whose variance depends on the gain λ and the phase-space representation ξ. The authors validate Eq. (3) with stochastic simulations for vacuum, squeezed, Fock, cat, and arbitrary states, propose that rotating the pump phase reconstructs the initial Husimi Q-function, and extend the model to Josephson parametric oscillators (JPOs).","tokens_in":10557,"tokens_out":5174,"duration_ms":55059,"significance":"If Eq. (3) is correct and its validity range is properly stated, the paper provides a simple analytical map from an initial quantum state's X-marginal to a macroscopic bistable outcome probability, with potential applications in quantum state reconstruction and in controlling driven-dissipative systems. The analytical derivation from Eq. (2) to Eq. (3) is straightforward, and the simulation agreement shown in Figs. 2 and 3 supports the result in the tested parameter regimes. The JPO proposal connects the theory to an existing superconducting-circuit platform, which strengthens the practical relevance. However, the Q-function reconstruction claim is presently stated without a necessary stability bound in λ, and the linearization assumption lacks a quantitative validity condition, so the central claim is not yet established in the full generality claimed.","major_comments":[{"comment":"For the Husimi Q representation (ξ=1), the variance in Eq. (3) is σ²=(2−λ)/(2(λ−1)), which is positive only for 1<λ<2, vanishes at λ=2, and is negative for λ>2. The text states the Q-function dynamics generically and refers to λ=2 as the 'noiseless regime', but it never restricts the Q-function claim to λ≤2. For λ>2, Eq. (2) itself has an imaginary noise coefficient for the Q representation, so the stochastic differential equation is not a real SDE and the Gaussian-smoothing interpretation of Eq. (3) is undefined. The paper must either explicitly restrict the Q-function reconstruction claim to 1<λ≤2 (treating λ=2 as a limiting case) or provide a Wigner-based version (ξ=0) for λ>2 and state that the Q-function reconstruction is not valid there.","section":"Results, Eq. (3) and Fig. 3"},{"comment":"The derivation of Eq. (2) drops the nonlinear term g²(a†a)a under the assumption g²≪λ−1, with the argument that steady-state probabilities are decided by early-time linearized dynamics. The manuscript does not state a quantitative condition on the initial state that makes this approximation valid. If the initial X-marginal has significant support near the separatrix (X≈0) or if the initial fluctuation amplitude is comparable to the saturation photon number, the nonlinearity can act before the sign decision is complete and bias the outcome away from Eq. (3). Please provide a concrete validity criterion (for example, a bound on the initial-state width relative to the saturation amplitude, or an estimate of the nonlinear correction during the decision time) and test Eq. (3) for initial states that approach that boundary, such as large-amplitude displaced states or states with substantial weight near X=0.","section":"Results, before Eq. (2)"},{"comment":"The statement that 'rotating the pump phase and repeating this procedure across various phases reconstructs the entire initial Q-function' requires inverting the Gaussian convolution in Eq. (3). For λ close to 1, the smoothing variance σ² can be large, making the deconvolution ill-conditioned; the paper does not describe the deconvolution procedure, its achievable fidelity, or its noise sensitivity. Figure 3(d) only shows reconstruction of the smoothed marginal at λ=2.0 and λ=1.2, and for λ=1.2 the derivative p'(b) still contains the Gaussian filter. Please specify the reconstruction protocol (including the deconvolution step and its resolution limits) or revise the claim to state that the measured quantity is a smoothed marginal rather than the exact initial Q-function.","section":"Results, Q-function reconstruction claim"}],"minor_comments":[{"comment":"The caption says 'Marginal Q-function (integrated along Y quadrature)' but does not state that this is the X-marginal Q(X)=∫dY Q(X+iY); please make that explicit.","section":"Fig. 2 caption"},{"comment":"The label p'(b) is not defined in the text or caption; clarify whether it is a numerical derivative of p(b) and how the Gaussian deconvolution (if any) was applied to obtain the reconstructed quadrature.","section":"Fig. 3(d)"},{"comment":"The JPO Langevin equation uses +ig²(a†a)a while the OPO equation (1) has −g²(a†a)a; if this sign difference is intentional (e.g., a different Kerr convention), state it explicitly; otherwise it appears to be a typo.","section":"Eq. (4)"},{"comment":"Calling λ=2 the 'noiseless regime' is potentially confusing: the physical system is not noiseless; rather, the noise coefficient in the Q-representation SDE vanishes. Please clarify this wording.","section":"Fig. 3 and main text"},{"comment":"The phrase 'high sensitivity to initial conditions' may be misread as classical chaos; the paper actually addresses the dependence of bistable-outcome probabilities on the initial quantum state. A sentence distinguishing this from exponential divergence of trajectories would improve readability.","section":"Introduction"},{"comment":"The statement that data and codes are 'available from the corresponding authors upon reasonable request' is weaker than a public repository; please provide a repository link for reproducibility.","section":"Data and code availability"},{"comment":"Eq. (2) is the foundation for the central result, but its derivation is relegated to SI Section S1, which was not available for inspection; please include at least a concise derivation sketch in the main text or supplement the main text with the key steps.","section":"Eq. (2) derivation"}],"recommendation":"major_revision","confidential_remarks":"The main text leans heavily on the SI for the derivation of Eq. (2), which I could not inspect; my major comments therefore focus on claims that are checkable from the main text alone. The self-citation [30] appears to be directly relevant prior work and is not problematic. The paper's central formula is appealing and likely correct in a restricted regime, but the missing λ≤2 bound for the Q-function and the missing validity condition for the linearization are load-bearing and need to be addressed before publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The real new step here is the arbitrary-state bias-probability formula, Eq. (3), and the proposed derivative-based reconstruction of the initial Q-function from a macroscopic steady-state probability. The derivation from the linearized Langevin equation is clean, and the simulations for vacuum, squeezed, Fock, cat, and arbitrary states match. The paper does not overclaim its novelty: it is a direct extension of Roques-Carmes et al. [30] to non-vacuum initial states, and it says so.\n\nWhere it does well: the logical structure is transparent, there are no fitted parameters, and the circularity burden is nil — Eq. (3) is solved from the SDE, not fitted. The proposed superconducting platform is concrete and builds on a demonstrated device. The generalization beyond OPOs to JPOs is plausible and supported by simulation.\n\nThe soft spots are real but fixable. The stress-test note about the Q-representation variance is correct: for xi=1, Eq. (3)'s Gaussian variance is (2-lambda)/(2(lambda-1)), which is negative for lambda>2. The paper never states the lambda<=2 restriction for the Q-function reconstruction; Fig. 3 only shows lambda=1.2 and 2.0, and the text says only that the noise coefficient \"vanishes at lambda = 2.\" That is load-bearing for the claim that rotating the pump phase reconstructs the initial Q-function. Either restrict the claim or switch to the Wigner representation and state the deconvolution protocol. This can be fixed with one paragraph.\n\nThe early-time linearization worry is also real, but less severe. The g^2 << lambda-1 condition is stated, but no quantitative bound is given for when the nonlinear term can be neglected while the sign of the amplified quadrature is decided. Initial states with significant support near the separatrix could bias the outcome before the linear Gaussian smoothing applies. The simulations stay within the safe regime; the generality statement in the Discussion is a bit broader than what is proven.\n\nMinor: the derivation of Eq. (2) is in the SI, which I did not inspect, and no code or data are shipped. The paper says data are available upon reasonable request, which is fine but not reproducible in the strong sense.\n\nWho it is for: people in quantum nonlinear optics and circuit QED who want a simple analytical handle on how initial quantum statistics leak into macroscopic bistable outcomes. It deserves a serious referee; the missing lambda<=2 bound and the linearization validity range should be addressed in revision. I would engage with it.","headline":"A clean, useful extension of the vacuum bias result to arbitrary quantum states, but the Q-function reconstruction claim needs a stated lambda<=2 validity bound and a quantitative linearization condition.","tokens_in":11066,"tokens_out":3051,"would_cite":true,"duration_ms":30342,"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":"Quantum states imprint themselves on a classical phase choice","keywords":["quantum sensitivity","parametric oscillation","quantum initial conditions","bistable steady states","Husimi Q-function reconstruction","Josephson parametric oscillator","driven-dissipative systems","nonlinear dynamics"],"falsifier":"Prepare a non-Gaussian initial state, such as a cat state with appreciable amplitude near $X=0$, pump the oscillator just above threshold, and measure $p(b)$ for several bias values; Eq. (3) predicts precisely the smoothed cumulative marginal $p_{X_0} * g_\\sigma$. The prediction would be falsified by observing a bias-probability curve that depends on the nonlinearity strength $g^2$ or that is sharper than the smoothed marginal, since either would show nonlinear saturation intervening before the linear filtering completed. A systematic scan of initial states with increasing separatrix weight would locate the boundary of the early-linearization regime.","tokens_in":10100,"feed_emoji":"⚛️","tokens_out":11153,"duration_ms":104793,"temperature":0.7,"pith_summary":"An optical parametric oscillator driven above threshold must choose between two steady phases, so its final condition is a macroscopic, apparently random decision. This paper claims that the probabilities of that decision are a filtered image of the quantum state that started in the cavity: the early parametric gain amplifies the initial $X$-quadrature distribution, cavity loss smooths it with a Gaussian of known width, and the sign at the moment of amplification decides the phase. The derived formula is Eq. (3), where the probability of the $\\alpha^{(1)}$ outcome is the integral over a shifted half-line of the initial marginal convolved with that Gaussian. Because sweeping the coherent bias shifts the decision boundary, the measured bias-probability curve is the cumulative of the smoothed marginal; differentiating it and rotating the pump phase reconstructs the initial Husimi $Q$-function. The authors argue the same mechanism governs Josephson parametric oscillators, making the effect testable in superconducting circuits.","feed_headline":"Quantum states imprint themselves on a classical phase choice","feed_subtitle":"Sweep the bias, count which phase wins, and the derivative of that curve reconstructs the initial quantum state.","key_machinery":"The key machinery is the early-time linearization of the Heisenberg-Langevin equation, Eq. (1), into the decoupled stochastic quadrature equations (2): $\\dot X = (\\lambda-1)X + \\sqrt{2}b + \\sqrt{1+\\xi(1-\\lambda)}\\,\\eta_1(t)$ and $\\dot Y = -(\\lambda+1)Y + \\sqrt{1+\\xi(1+\\lambda)}\\,\\eta_2(t)$, with $\\eta_1,\\eta_2$ independent standard-normal noises. Integrating the $X$ equation turns the initial marginal into a Gaussian-smoothed one whose width $\\sigma^2 = (1+\\xi(1-\\lambda))/(2(\\lambda-1))$ is the Green's function of the linearized amplifier; the sign of the smoothed trajectory at the decision time selects the steady state. The special value $\\lambda=2$ in the Husimi $Q$ representation makes the noise term vanish, so the final phase is deterministic given the initial $X$ sign; for other parameters the Gaussian filter widens and washes out fine features of the initial distribution. Everything downstream in the paper -- the bias-probability curve, the reconstruction protocol, the JPO extension -- is this linear-filter picture transported to a measurable classical probability.","core_discovery":"On its own terms, the paper establishes a quantitative 'quantum sensitivity': a closed relationship between an arbitrary initial quantum state and the steady-state probabilities of a degenerate biased OPO. In the regime $g^2 \\ll \\lambda - 1$, the steady-state probabilities are set during early-time linearized dynamics, before the nonlinear saturation term $g^2(a^\\dagger a)a$ becomes important. The result is Eq. (3), $p(\\alpha^{(1)}, b) = \\int_{-\\sqrt{2}b/(\\lambda-1)}^{\\infty} (p_{X_0} * g_\\sigma)(x)\\,dx$, with $p_{X_0}$ the $X$-marginal of the initial phase-space distribution and $g_\\sigma$ a zero-mean Gaussian of variance $\\sigma^2 = (1+\\xi(1-\\lambda))/(2(\\lambda-1))$. The authors describe the content as: the steady-state distribution $p(\\alpha^{(1)})$ is the initial $X$-quadrature marginal, smoothed by a Gaussian of variance $\\sigma^2$, then integrated over the right of the decision boundary. The parameter $\\xi$ selects the quasiprobability representation, with $\\xi=1$ for the Husimi $Q$-function used in the simulations, $\\xi=0$ for Wigner, and $\\xi=-1$ for Glauber-Sudarshan $P$. The paper further shows numerically that the same bias-probability formula holds for vacuum, squeezed, Fock, cat, and arbitrary states, and that a Josephson parametric oscillator obeys the same Heisenberg-Langevin structure, so the same mapping applies there.","pith_inferences":["Editorial extension: with a calibrated Gaussian width, deconvolution of $p'(b)$ could recover the unsmoothed marginal, effectively performing quantum state tomography of non-Gaussian states without heterodyne detection, limited by how close to threshold the oscillator can run.","Editorial extension: the treatment suggests the classical bias acts as a movable decision boundary while the quantum state supplies the shape of the probability distribution; thus phase-choice statistics isolate the quantum contribution to the initial condition, which could be used to certify non-Gaussianity in a single macroscopic observable.","Editorial extension: operating near threshold ($\\lambda$ close to 1) maximizes the sensitivity of the outcome to small differences in the initial state, pointing toward quantum-enhanced sensing of quadrature displacements, provided the early-linearization assumption remains valid.","Editorial extension: applying the same logic to multimode systems such as Kerr combs would require a multimode generalization of Eq. (3), but if that exists, the steady-state phase pattern could encode intermodal quantum correlations and enable tomography of multimode entangled states."],"forward_implications":["The derivative of the measured bias-probability curve gives the Gaussian-smoothed $X$-marginal of the initial quantum state, so a single sweep of the bias is a partial state-characterization measurement.","Rotating the pump phase before repeating the sweep reconstructs the full Husimi $Q$-function, which fully describes the initial state, from classical phase-choice statistics alone.","The theory predicts that each initial state's $X$-marginal shape, not just its center of mass, controls the phase-choice curve, so Fock and squeezed states produce visibly different $p(b)$ at fixed bias.","Because Josephson parametric oscillators obey the same equations, the prediction is testable in superconducting circuits where non-Gaussian states can be prepared and swapped into a parametric cavity.","The authors argue that the same early-linearization reasoning should generalize to other multistable driven-dissipative systems in the low-quantum-noise regime."],"supporting_citations":[{"why":"Predecessor result showing a coherent bias on the quantum vacuum controls the macroscopic bistable probabilities of an OPO; the present paper generalizes this to arbitrary quantum initial states.","marker":"[30]"},{"why":"Standard quantum-optics reference supplying the phase-space stochastic methods used to derive the decoupled quadrature equations (2).","marker":"[31]"},{"why":"Companion proposal for observing the dynamics of quantum states inside nonlinear optical cavities, cited for reconstructing the initial Q-function from measured slices.","marker":"[32]"},{"why":"Derives the Heisenberg-Langevin equation for a flux-modulated Josephson parametric cavity used as Eq. (4).","marker":"[40]"},{"why":"Extends the parametric circuit treatment to circuit quantum electrodynamics, supporting the JPO model.","marker":"[41]"},{"why":"Introduces and characterizes Josephson parametric oscillators as bistable parametric systems, the superconducting analogue this paper analyzes.","marker":"[42]"},{"why":"Supplies the demonstrated chip-scale parametric cavity and state-swap protocol used in the proposed superconducting experiment.","marker":"[45]"},{"why":"Surveys Josephson parametric amplifiers, supporting the availability of the quantum-limited amplification and readout assumed in the experiment.","marker":"[39]"}],"fun_headline_variants":["Quantum state readout via steady-state phase probabilities","Initial quantum state maps onto classical decision curve","OPO bias sweep reconstructs quantum marginal","Quantum sensitivity links state to steady-state outcomes","Count phase wins to recover the initial quantum state"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the phase choice is fully determined during early linear growth, so the nonlinear saturation term can be ignored while the sign of the amplified quadrature is being decided; this requires the initial state to have negligible probability weight near the unstable separatrix and the saturation photon number to be large compared with the initial fluctuation scale.","fun_headline_variants_meta":{"raw":{"variants":["Quantum state readout via steady-state phase probabilities","Initial quantum state maps onto classical decision curve","OPO bias sweep reconstructs quantum marginal","Quantum sensitivity links state to steady-state outcomes","Count phase wins to recover the initial quantum state"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000245,"raw_usage":{"total_tokens":1594,"prompt_tokens":1061,"completion_tokens":533,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":677,"completion_tokens_details":{"reasoning_tokens":479}},"tokens_in":677,"tokens_out":533,"duration_ms":6340,"temperature":1.0,"reasoning_tokens":479,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T23:00:01.869384+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Prepare a non-Gaussian initial state, such as a cat state with appreciable amplitude near $X=0$, pump the oscillator just above threshold, and measure $p(b)$ for several bias values; Eq. (3) predicts precisely the smoothed cumulative marginal $p_{X_0} * g_\\sigma$. The prediction would be falsified by observing a bias-probability curve that depends on the nonlinearity strength $g^2$ or that is sharper than the smoothed marginal, since either would show nonlinear saturation intervening before the linear filtering completed. A systematic scan of initial states with increasing separatrix weight would locate the boundary of the early-linearization regime.","supporting_citations":[{"cited_title":"Biasing the quantum vacuum to control macro- scopic probability distributions,","cited_arxiv_id":null,"evidence_quote":"Predecessor result showing a coherent bias on the quantum vacuum controls the macroscopic bistable probabilities of an OPO; the present paper generalizes this to arbitrary quantum initial states."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Standard quantum-optics reference supplying the phase-space stochastic methods used to derive the decoupled quadrature equations (2)."},{"cited_title":"Observing the dynamics of quantum states generated inside nonlinear optical cavities","cited_arxiv_id":"2412.01772","evidence_quote":"Companion proposal for observing the dynamics of quantum states inside nonlinear optical cavities, cited for reconstructing the initial Q-function from measured slices."},{"cited_title":"Parametric resonance in tun- able superconducting cavities,","cited_arxiv_id":null,"evidence_quote":"Derives the Heisenberg-Langevin equation for a flux-modulated Josephson parametric cavity used as Eq. (4)."},{"cited_title":"Parametric effects in circuit quantum electrodynamics,","cited_arxiv_id":null,"evidence_quote":"Extends the parametric circuit treatment to circuit quantum electrodynamics, supporting the JPO model."},{"cited_title":"In- vestigation of nonlinear effects in Josephson parametric oscil- lators used in circuit quantum electrodynamics,","cited_arxiv_id":null,"evidence_quote":"Introduces and characterizes Josephson parametric oscillators as bistable parametric systems, the superconducting analogue this paper analyzes."},{"cited_title":"Efficient and low-backaction quantum measurement using a chip-scale detector,","cited_arxiv_id":null,"evidence_quote":"Supplies the demonstrated chip-scale parametric cavity and state-swap protocol used in the proposed superconducting experiment."},{"cited_title":"Superconducting parametric amplifiers: The state of the art in Josephson parametric amplifiers,","cited_arxiv_id":null,"evidence_quote":"Surveys Josephson parametric amplifiers, supporting the availability of the quantum-limited amplification and readout assumed in the experiment."}],"review_version":1}