{"id":"fc654175-aa7f-490e-9cc5-8a395fa3c88b","arxiv_id":"2607.17023","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Quadrature-power (sign-free) measurements after phase-sensitive amplification are tomographically complete for parity-symmetric states and certify Wigner negativity and stellar rank from the same data.","lead":"A new measurement trick — amplifying a light state with an optical parametric amplifier and reading only its output power — can reconstruct the full quantum state of any 'parity-symmetric' light state and certify its non-classical features while discarding the quadrature sign. The method is meant to tolerate detector loss and bandwidth limits that hurt ordinary homodyne tomography, though the loss tolerance is shown by theory and emulation, not yet by an actual experiment.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Loss tolerance relies entirely on post-amplifier detector-loss model; pre-OPA or internal OPA loss is not corrected by Eq. (13), so the headline advantage is untested.","rationale":"The reader identified the post-amplifier-only loss model as the weakest assumption, and I agree: it is structurally distinct from the mathematics of sign-free quadrature completeness and is the premise on which the paper's headline advantage depends. The tomographic completeness for parity-symmetric states is on firmer ground—for such states P(x) is even, so |x|-data are equivalent to full quadrature data—and the parameter-free elementwise estimator of Eq. (13) gives independent support (Table II). The SDP's γ-tuning is a real but secondary concern because the elementwise route partially corroborates the tomography claim; the loss model, however, has no such independent support in the paper. The proposed test directly probes whether the loss-tolerance claim survives when loss occurs before the amplifier, which is the condition that would make the central claim false. Since the paper is honest about the limited experimental validation and the concern is addressable, the existing CONDITIONAL verdict is appropriate; my read does not move it.","tokens_in":32585,"tokens_out":8338,"duration_ms":81949,"concrete_test":"Simulate single-photon, cat, and GKP states and generate sign-free quadrature data from a realistic loss model: insert a beam splitter with transmittance η_in before the OPA (equivalently, apply a loss channel to ρ before the Bogoliubov transform), then include finite gain r and post-amplifier detection loss. Run the SDP (Eq. 8) and elementwise estimator (Eq. 13) reconstructions. If the reconstructed state converges to the lossy state rather than the original ρ for η_in < 1, or if the fidelity drops by more than the statistical error bars, the loss-tolerance claim is invalid outside the post-amplifier-loss-only model. Repeat for η_in = 0.9, 0.75, 0.5 and report fidelities as a function of gain.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central 'loss-tolerant' claim rests on the loss model of Appendix A: all detector inefficiency is placed after the OPA as a beam splitter of transmittance η (Eq. A8), giving a POVM whose added vacuum noise scales as σ² = e^{-2r}Δ² and vanishes as r → ∞ (Eqs. A9–A10). This is the only stated mechanism by which quadrature-power data approach the ideal sign-free homodyne POVM of Eq. (5). The assumption is load-bearing because the paper's advertised advantage over BHD is loss tolerance. If a comparable fraction of loss occurs before or inside the amplifier—input coupling loss, OPA internal loss, excess amplifier noise, or pump depletion—the measured statistics are the sign-free quadratures of a lossy state, not of ρ. Equation (13), with the estimator R^η of Eq. (12), corrects for detector efficiency after state preparation but cannot undo loss applied before the OPA. The paper provides no actual OPA power measurements on non-Gaussian states (Sec. IV explicitly defers this), so the loss model is validated only by emulation from homodyne data. If this model fails under realistic pre-amplifier loss, the 'loss-tolerant characterization' claim fails; the parity-symmetric completeness result would survive but only in the lossless-measurement idealization.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes an all-optical measurement scheme based on a high-gain phase-sensitive optical parametric amplifier followed by quadrature-power detection, which realizes a 'sign-free quadrature' POVM (|x⟩⟨x| + |−x⟩⟨−x|). The main theoretical claim is that sign-free quadrature data are sufficient for complete tomography of parity-symmetric states (Fock, cat, binomial, GKP), via an elementwise estimator identity (Eq. 13) and a semidefinite-program reconstruction (Eq. 8). The paper also develops direct estimators for linear functionals such as fidelities, and uses them to certify stellar rank and Wigner negativity. Validation is performed on both simulated data and experimental homodyne data, in which the sign-free dataset is obtained by taking absolute values of measured quadratures.","tokens_in":32742,"tokens_out":15356,"duration_ms":146017,"significance":"If the central claims hold, the scheme would relax detector-efficiency and bandwidth constraints in continuous-variable quantum state characterization, which is an important practical goal. The estimator identity in Appendix E is clean and correct: for even order differences, the homodyne pattern function is even in x, so the sign-free distribution gives the same expectation value, and this yields a parameter-free elementwise reconstruction. The trace-distance cross-checks (0.029, 0.022, 0.015) between the elementwise and SDP reconstructions provide nontrivial internal consistency. However, the SDP fidelity benchmark is weakened by the target-dependent choice of the regularization weight γ, and the loss-tolerance derivation in Appendix A contains a serious mathematical gap. The paper is therefore a valuable contribution in need of substantial revision.","major_comments":[{"comment":"The derivation of the lossy POVM is not correct. In passing to the third line of Eq. (A9), the term (G/2)Δ²v² in the exponent is dropped from the vacuum expectation without justification. For a beam splitter of transmittance η followed by an ideal square-law detector, the measured intensity is G(√η x + √(1−η)v)², so the vacuum variable v enters both linearly and quadratically. The resulting POVM kernel is not the Gaussian exp[−G(x²−x′²)²/Δ²] of Eq. (A9); it is a noncentral chi-square distribution. The conclusion (A10) that the POVM approaches the ideal sign-free POVM in the high-gain limit may still hold, but it is not established by the given calculation, and the stated convergence rate is not correct. Since Eq. (A10) is the formal basis of the loss-tolerance claim, this derivation must be redone properly.","section":"Appendix A, Eq. (A9)"},{"comment":"The estimator R^η_{n,n+d} is said to be defined 'for all ... η≤1', but the integral converges only for η > 1/2. For η < 1/2 the exponent −(1−1/(2η))k² is positive and the integral diverges; at η = 1/2 the exponential is unity and the polynomial |k|k^d L_n^{(d)}(k²) does not decay. Thus the identity (13) cannot hold for η ≤ 1/2. The paper must either restrict the domain explicitly, or, if the high-gain OPA is intended to realize an effective efficiency of 1, set η = 1 in Eq. (13) and provide a quantitative bound on the finite-gain deviation from the ideal POVM.","section":"Eq. (12)"},{"comment":"The SDP regularization weight γ is described as an 'ad hoc parameter' chosen to maximize the fidelity of the reconstruction. When the target state is known (as in the simulations and in the experimental comparison against MaxLik), this choice makes the reported fidelities in Table I partly tautological: γ is tuned to the answer and the fidelity is then quoted as a performance metric. This does not demonstrate predictive power for unknown states. Please replace this by a principled selection rule (e.g., cross-validation on a training set, or a default schedule) and report the sensitivity of the fidelities to γ. The parameter-free elementwise estimator of Table II is a meaningful cross-check and should be foregrounded.","section":"Sec. III B, Eq. (8), Table I"},{"comment":"The loss-tolerance claim is only established under the assumption that all loss is modeled as a beam splitter placed after the OPA. Losses before or inside the OPA (input coupling loss, internal loss, pump depletion) are not corrected by Eq. (13), and their effect is not quantified. Since the abstract and introduction advertise 'loss-tolerant characterization', the scope of the claim should be stated explicitly. A numerical study with a fraction of loss placed before the OPA would clarify the practical validity of the scheme under realistic conditions.","section":"Sec. II, Appendix A"}],"minor_comments":[{"comment":"The reference for the photon-subtracted squeezed state appears as '[?]' in the text; it should be Ref. [63].","section":"Sec. III B, Ref. [63]"},{"comment":"The caption labels the stellar-rank witness as obtained via 'direct fidelity estimation', but the text in Sec. III C says the stellar rank is certified from the SDP-reconstructed density matrix. Please align the terminology.","section":"Fig. 4 caption"},{"comment":"After taking the high-gain limit, the proportionality constant of the POVM should be stated explicitly; the factor 1/(2G|x|) in Eq. (A7) should be resolved.","section":"Eq. (A10)"},{"comment":"The paper validates the scheme with experimental homodyne data and emulated sign-free data. It would be helpful to state explicitly in the abstract or introduction that no direct OPA power measurement on non-Gaussian states is reported, to avoid overclaiming experimental validation.","section":"Abstract / Sec. IV"},{"comment":"There are several typographical issues: 'using using' in Fig. 4 caption, 'T op row' in Figs. 2 and 3 captions, and inconsistent use of 'Wigner negativity witness' phrasing. These should be corrected.","section":"General"}],"recommendation":"major_revision","confidential_remarks":"The core estimator identity (Appendix E) is correct and valuable, and the elementwise reconstruction provides a credible parameter-free cross-check. The main blockers are the erroneous lossy-POVM derivation in Appendix A and the circular choice of γ in the SDP benchmark. Both are fixable in revision. If the loss derivation cannot be corrected, the loss-tolerance claim should be downgraded to a heuristic high-gain limit. I recommend major revision rather than rejection because the central mathematical contribution is sound and the experimental/emulation results, while not decisive, support the feasibility of the approach."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"You should read this one. The core idea is simple and useful: for parity-symmetric states, sign-free quadrature data (measuring |x| instead of x) is tomographically complete, and the paper backs that with a correct estimator identity (Eq. 13) built on D'Ariano–Paris–Sacchi homodyne estimators. The elementwise reconstruction is parameter-free, and the trace distances against the SDP reconstruction (0.029/0.022/0.015) are a real cross-check. That is the strongest evidence in the paper, and it holds up. The stellar-rank and Wigner-negativity certifications are also a nice add-on, reusing the same data without needing photon-number-resolving detectors. The paper is honest: it uses emulated homodyne data, explicitly defers the actual OPA experiment on non-Gaussian states, and admits the stellar-rank witness fails for mixed GKP states. Credit where due—this is a well-structured, useful contribution to CV quantum information.\n\nThe soft spots are real but not fatal. First, the headline SDP fidelities in Table I are tuned benchmarks: gamma is chosen per state to maximize fidelity against the known target. That makes the near-unity numbers less impressive, and since the same SDP reconstruction feeds the stellar-rank and Wigner-negativity witnesses, those certifications inherit the fitted benchmark issue. Second, the \"rigorous confidence intervals\" are essentially a promise. M in Eq. 15 is never computed, and no interval is ever reported. The bound is an existence statement, not an operational tool in this paper. Third, the loss-tolerance claim rests on a specific loss model: all detector inefficiency placed as a beam splitter after the OPA (Eq. A8), with added vacuum noise vanishing as r → ∞. That is fine for detection loss after the amplifier, but input coupling loss, internal OPA loss, and pump depletion are outside the model. The parity-symmetric completeness result does not depend on this model, so the mathematical core survives; the advertised practical advantage over BHD needs qualification or a real experiment.\n\nMinor: fidelities and trace distances lack error bars, and the histogram/truncation hyperparameters are just reported without sensitivity checks.\n\nWho is this for? Experimentalists working on CV state verification, especially for bosonic error correction. It deserves serious peer review—the central claim is well-supported by the parameter-free estimator cross-check, and the issues are addressable in revision. I would send it to referees.","headline":"A clean and honest tomography package for parity-symmetric CV states from |x| data, with a loss-tolerance claim that outruns the experiment.","tokens_in":33449,"tokens_out":2116,"would_cite":true,"duration_ms":22834,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A high-gain optical parametric amplifier followed by power measurement realizes the sign-free quadrature POVM, and sign-free quadrature data are sufficient to fully reconstruct and certify parity-symmetric quantum states.","keywords":["sign-free quadrature measurement","optical parametric amplifier","continuous-variable quantum tomography","parity-symmetric states","stellar rank","Wigner negativity","loss-tolerant detection","semidefinite programming"],"falsifier":"Measure a known parity-symmetric state (e.g., a single photon) with an OPA at finite gain, deliberately adding a controlled pre-amplifier loss (e.g., an attenuator before the OPA). If the reconstruction fidelity degrades according to the added pre-amplifier loss rather than remaining flat as gain increases, the claim that the scheme is loss-tolerant in the stated sense is falsified. Equivalently, resolve the two-peaked POVM at fixed η and increasing gain: the width should shrink as e^{-2r}; a plateau would indicate an unmodeled noise source.","tokens_in":32282,"feed_emoji":"⚛️","tokens_out":3819,"duration_ms":32908,"temperature":0.7,"pith_summary":"The paper establishes that dropping the sign of the measured quadrature—recording only the squared value after phase-sensitive amplification—does not cost full tomographic information for any state with definite photon-number parity. It proves loss-tolerance: detector inefficiency adds vacuum noise whose width shrinks as e^{-2r} with amplifier gain, so the measurement approaches the ideal |x⟩⟨x|+|−x⟩⟨−x| POVM. From such data, a convex semidefinite program reconstructs the density matrix, and a closed-form estimator gives each same-parity matrix element with Hoeffding-level confidence intervals. The same data feed stellar-rank and Wigner-negativity witnesses without full tomography. If true, this gives a hardware-simple, loss-tolerant route to verifying non-Gaussian states used in bosonic quantum error correction.","feed_headline":"Power-only measurements fully reconstruct parity-symmetric quantum states","feed_subtitle":"Amplify, square, measure: loss-tolerant tomography and non-Gaussianity certification for Fock, cat, and GKP states.","key_machinery":"The sign-free quadrature POVM Π_φ=|x_φ⟩⟨x_φ|+|−x_φ⟩⟨−x_φ|, realized by high-gain phase-sensitive OPA plus quadrature-power detection. The companion estimator R^η_{n,n+2k}(x,φ), built from generalized Laguerre polynomials, turns each measured (|x|,φ) pair into an unbiased sample of a same-parity density-matrix element; the even-d reflection symmetry R(φ,−x)=R(φ,x) is what makes sign-free data sufficient.","core_discovery":"The central discovery is that the absolute value of a quadrature—measured via a high-gain phase-sensitive OPA and power detection—is tomographically complete for the class of parity-symmetric states. The POVM element is Π_φ=|x_φ⟩⟨x_φ|+|−x_φ⟩⟨−x_φ|, and the paper shows that for any n and k the density-matrix element ρ_{n+2k,n} equals the expectation value of an explicit estimator R^η_{n,n+2k} over uniformly random phase φ and outcomes |x_φ|. Consequently the full density matrix of any parity-symmetric state (Fock, cat, binomial, GKP) can be reconstructed elementwise, or via an SDP with physicality constraints, and linear functionals such as fidelities and Wigner-negativity witnesses can be es","pith_inferences":["If pre-amplifier loss or excess amplifier noise is present, the 'loss-tolerant' headline must be rederived: the vacuum-noise width no longer vanishes in the high-gain limit unless the loss is modeled after the amplifier. A quantitative bound on pre-amplifier loss would sharpen the claim.","The estimator-based reconstruction suggests a direct multimode generalization: sign-free data for each mode, with K-mode parity symmetry, could certify entanglement and non-Gaussianity in cluster states without full multimode tomography.","The high-gain argument implies a practical calibration test: measure the residual width of the two-peaked POVM as a function of gain; finding a plateau would identify a non-OPA noise source that the model does not cover.","The stellar-rank witnesses were found too stringent for mixed experimental GKP states; non-linear multi-copy witnesses suggested by the authors would be a natural next step to certify those states."],"forward_implications":["Tomography of parity-symmetric CV states (single-photon, cat, binomial, GKP) requires only |x| statistics, not full homodyne sign resolution, while reaching near-unity fidelities (≥0.99 in simulations).","Detector inefficiency can be tolerated: the measurement approaches the ideal sign-free POVM in the high-gain limit, removing the need for high-efficiency photon-number-resolving detectors.","Direct estimation of linear features—fidelity with target states, stellar rank, Wigner negativity—is possible with rigorous Hoeffding confidence intervals, bypassing full tomography.","Bandwidth is set by the OPA gain, not by electronic homodyne bandwidth, extending the accessible measurement bandwidth.","The same scheme certifies non-Gaussianity (stellar rank and Wigner negativity) for Fock, cat, and GKP states from experimental datasets."],"fun_headline_variants":["Absolute quadrature power alone reconstructs parity-symmetric states","OPA power measurements unlock loss-tolerant quantum state tomography","Certify Fock, cat, and GKP states with just amplified power readings","Parity-symmetric states fully recoverable from absolute quadrature data","High-gain OPA: one power meter replaces lossy homodyne tomography"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"All detection loss is modeled as a beam splitter placed after the amplifier, with an undepleted classical pump; if a comparable loss occurs before or inside the amplifier, the added vacuum noise does not shrink to zero at high gain and the loss-tolerant claim breaks down.","fun_headline_variants_meta":{"raw":{"variants":["Absolute quadrature power alone reconstructs parity-symmetric states","OPA power measurements unlock loss-tolerant quantum state tomography","Certify Fock, cat, and GKP states with just amplified power readings","Parity-symmetric states fully recoverable from absolute quadrature data","High-gain OPA: one power meter replaces lossy homodyne tomography"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000198,"raw_usage":{"total_tokens":1216,"prompt_tokens":768,"completion_tokens":448,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":512,"completion_tokens_details":{"reasoning_tokens":356}},"tokens_in":512,"tokens_out":448,"duration_ms":4595,"temperature":1.0,"reasoning_tokens":356,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T19:16:23.931000+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure a known parity-symmetric state (e.g., a single photon) with an OPA at finite gain, deliberately adding a controlled pre-amplifier loss (e.g., an attenuator before the OPA). If the reconstruction fidelity degrades according to the added pre-amplifier loss rather than remaining flat as gain increases, the claim that the scheme is loss-tolerant in the stated sense is falsified. Equivalently, resolve the two-peaked POVM at fixed η and increasing gain: the width should shrink as e^{-2r}; a plateau would indicate an unmodeled noise source.","supporting_citations":[],"review_version":1}