Pith. sign in

REVIEW 5 minor 28 references

Higher-order covariance matrices let Gaussian operations act on non-Gaussian continuous-variable states by ordinary matrix multiplication of moderate size.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · grok-4.5

2026-07-14 03:22 UTC pith:72OSNRDT

load-bearing objection Clean, usable extension of the covariance-matrix toolkit to higher moments; solid math, modest novelty, ready for referees.

arxiv 2607.11759 v1 pith:72OSNRDT submitted 2026-07-13 quant-ph

Higher-order covariance matrices for non-Gaussian quantum states

Vojt\v{e}ch Kala , Petr Marek , Nicolas J. Cerf This is my paper
classification quant-ph
keywords higher-order covariance matrixnon-Gaussian statescontinuous-variable quantum informationnonlinear squeezingnon-Gaussian nullifiershomodyne detectionsymplectic transformationsWeyl-symmetric monomials
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

Standard covariance matrices fully describe Gaussian continuous-variable states and how they transform under linear optics, but they ignore the higher moments that carry non-Gaussian resources. This paper defines covariance matrices built from Weyl-symmetric monomials of the quadratures up to a chosen order N. Those matrices still transform under Gaussian operations by the familiar rule γ′ = M γ Mᵀ, yet they encode the nonlinear squeezing and non-Gaussian nullifiers that ordinary second-moment matrices miss. They can be reconstructed from homodyne data at only a handful of fixed phase angles, their size is independent of any Fock-space cutoff, and they grow only polynomially with the number of modes. The construction therefore supplies a practical intermediate description that is rich enough for non-Gaussian witnesses yet cheap enough for multi-mode calculations and loss simulations.

Core claim

A finite set of Weyl-symmetric quadrature monomials closed under the Gaussian group yields a higher-order covariance matrix that transforms exactly as an ordinary covariance matrix under symplectic maps, remains of moderate dimension independent of Fock truncation, and can be estimated from a fixed finite set of homodyne angles; the same object evaluates both nonlinear squeezing and non-Gaussian cluster nullifiers.

What carries the argument

The higher-order covariance matrix γ built from the vector r of Weyl-symmetric monomials up to order N (illustrated for N=2 by the five-dimensional r = (x, p, x^{2}, xp+px, p^{2})ᵀ). Its transformation law under any Gaussian map is still γ′ = M γ Mᵀ, so Gaussian free operations become ordinary matrix multiplications of size independent of Hilbert-space cutoff.

Load-bearing premise

The chosen finite list of monomials must stay closed under every Gaussian operation; if a cubic or higher gate is applied, new monomials appear and the matrix representation ceases to be finite.

What would settle it

Prepare a known cubic-phase or photon-subtracted state, reconstruct its five-dimensional higher-order covariance matrix from six fixed homodyne angles, apply a sequence of Gaussian unitaries via the paper’s M matrices, and check whether the resulting nonlinear-squeezing or nullifier values match independent Fock-space calculations within sampling error.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • Nonlinear squeezing can be optimized over all Gaussian free operations by multiplying only 5×5 matrices instead of truncated Fock operators.
  • Optical loss on a non-Gaussian state is obtained in closed form from a 14-dimensional two-mode higher-order covariance matrix without enlarging the Fock cutoff.
  • Non-Gaussian cluster-state nullifiers become ordinary quadratic forms on a matrix whose dimension grows only as O(n^{2}) with mode number n.
  • Experimental certification of non-Gaussianity no longer requires full tomography or prior knowledge of the state’s orientation in phase space.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same construction immediately suggests a hierarchy of witnesses ordered by the maximal monomial degree N, each still transformable by Gaussian matrices.
  • Because the dimension is polynomial in modes, the method is a natural candidate for real-time feedback control of multi-mode non-Gaussian resources.
  • Any continuous-variable protocol that treats Gaussian operations as free (GKP Clifford layers, Gaussian conversion of cubic states, etc.) can reuse the same higher-order matrices without redesign.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

0 major / 5 minor

Summary. The manuscript introduces higher-order covariance matrices constructed from Weyl-symmetric quadrature monomials of total degree at most N. These matrices transform under Gaussian (symplectic) operations exactly as ordinary covariance matrices do (γ′ = M γ Mᵀ, with an affine shift for the mean vector), remain of dimension independent of any Fock-space cutoff, and scale only polynomially with mode number. Explicit transformation matrices are given for single-mode rotations, squeezers and displacements (N = 2) and for the two-mode beam splitter; pure loss is obtained in closed form. The construction is applied to the evaluation of cubic nonlinear squeezing and of nullifiers for non-Gaussian cluster states, both of which can be read off from a five-dimensional matrix estimated from a fixed, small set of homodyne angles. A physicality condition of the form γ + (i/2)⟨Ω⟩ ≥ 0 is derived from positivity of the second-moment matrix of the centered operators.

Significance. If the claims hold, the work supplies a practical intermediate description between full density-matrix numerics and ordinary second-moment Gaussian methods. Because the matrices are small, closed under the Gaussian group, and directly estimable from few-phase-lock homodyne data, they enable efficient optimization of non-Gaussian witnesses over free Gaussian operations and efficient simulation of loss without Fock truncation. The explicit N = 2 matrices and the loss channel formula are immediately usable; the polynomial scaling with mode number is a genuine advantage for multi-mode non-Gaussian cluster-state analysis. The limitation that cubic gates take the chosen set outside itself is stated openly and does not undermine the stated scope (Gaussian free operations on non-Gaussian resources).

minor comments (5)
  1. In the abstract and introduction the phrase “higher-order covariance matrices” is used interchangeably with “covariance matrices built from higher-order quadrature monomials.” A single clarifying sentence early in Sec. I would prevent any confusion with ordinary higher moments of the usual (x,p) covariance matrix.
  2. Eq. (14) for Ω is written with dots for the lower triangle; filling the antisymmetric entries (or stating that they follow by antisymmetry) would make the physicality check fully self-contained.
  3. Appendix C lists the conversion formulae for moments up to order 4, yet the main text claims that six phase locks suffice for the full N = 2 matrix. A short table mapping each entry of γ to the required angles would make the experimental protocol completely explicit.
  4. Figure 1 caption should state the numerical values of r and χ used for the state |ψ_{3}⟩ so that the plot is reproducible without hunting through the text.
  5. A few typographical inconsistencies appear (e.g., “varˆρ” versus “var”, missing spaces around operators). A light copy-edit pass would remove them.

Circularity Check

0 steps flagged

No significant circularity: higher-order covariance matrices and their Gaussian transformation law are defined and derived from first principles.

full rationale

The paper introduces the set SN of Weyl-symmetric quadrature monomials (closed under phase rotations by construction), forms the vector r, and defines the higher-order mean and covariance matrix γ exactly as in the ordinary second-moment case (Eqs. 3–5). The central transformation law γ′ = M γ Mᵀ is then obtained by direct substitution of the Heisenberg action r′ = M r + v into the definition of γ, with the algebraic steps written out in Eq. (9); the same holds for the mean vector. Physicality (Eq. 10) follows at once from positivity of the second-moment matrix of the centered operators (Appendix A). Explicit finite matrices M for rotation, squeezing, displacement and the beam-splitter are obtained by expanding the elementary symplectic maps on the chosen monomials (Eqs. 15–20). Estimation from a fixed finite set of homodyne angles is constructive (Appendix C). The numerical illustration uses independently chosen parameters (r = −0.3, χ = 0.1) and compares against a separately derived Gaussian-limit formula; nothing is fitted and then re-predicted. Self-citations appear only for the application contexts (nonlinear squeezing, nullifiers) and do not underwrite the definition or the transformation law. The derivation is therefore self-contained and free of the listed circularity patterns.

Axiom & Free-Parameter Ledger

1 free parameters · 3 axioms · 1 invented entities

The construction rests on standard continuous-variable quantum mechanics plus the single modeling choice that a finite, rotation-closed set of monomials is sufficient for the intended witnesses. No free parameters are fitted to data; the example parameters are illustrative only.

free parameters (1)
  • example squeezing r and cubic strength χ
    Chosen by hand (r = −0.3, χ = 0.1) solely to produce a concrete numerical matrix; the formalism itself does not depend on them.
axioms (3)
  • standard math Canonical commutation relations [x, p] = i and the existence of a well-defined Weyl-symmetric ordering for monomials.
    Used throughout to define the operators that enter the higher-order vector r and the commutator matrix Ω.
  • domain assumption Gaussian unitaries act by linear symplectic transformations on the quadrature vector (x, p).
    Standard continuous-variable quantum optics; invoked to guarantee that the induced map on higher monomials is linear and therefore representable by a matrix M.
  • ad hoc to paper The finite set SN of monomials of total degree ≤ N is closed under phase rotations (and, by extension, under the full Gaussian group for the chosen applications).
    Stated as a requirement immediately after Eq. (3); without it the transformation law γ′ = M γ Mᵀ would not close inside a finite matrix.
invented entities (1)
  • higher-order covariance matrix γ built from the vector of Weyl-symmetric monomials r independent evidence
    purpose: To package all moments up to order N into a single object that transforms covariantly under Gaussian operations and can be estimated from finite homodyne data.
    The central definitional contribution of the paper; independent evidence is the explicit reconstruction formulae and the numerical match with Fock-space simulation.

pith-pipeline@v1.1.0-grok45 · 16989 in / 2314 out tokens · 21855 ms · 2026-07-14T03:22:28.470951+00:00 · methodology

0 comments
read the original abstract

Covariance matrices lie at the heart of the powerful and well-established symplectic framework for describing continuous-variable Gaussian quantum states. However, since this framework only relies on first- and second-order moments, it is not sufficient for the analysis of non-Gaussian states because their higher-order moments are essential to capture some of their key properties. Here, we define higher-order covariance matrices -- more precisely, covariance matrices built from higher-order quadrature monomials -- which provide a simple way to evaluate the effect of Gaussian transformations on non-Gaussian states and can be used, for example, to address nonlinear squeezing or non-Gaussian nullifiers. Higher-order covariance matrices can be estimated from homodyne measurement data using only a limited number of quadrature angles, which involves matrices of moderate dimension compared with a full simulation in the Fock basis. The dimension of a higher-order covariance matrix does not depend on the span of the quantum states in Fock basis and, furthermore, scales only polynomially with the number of modes.

Figures

Figures reproduced from arXiv: 2607.11759 by Nicolas J. Cerf, Petr Marek, Vojt\v{e}ch Kala.

Figure 1
Figure 1. Figure 1: FIG. 1. Variance of the operator [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

28 extracted references · 9 canonical work pages

  1. [1]

    J. Aasi, J. Abadie, B. Abbott, et al. Enhanced sensi- tivity of the ligo gravitational wave detector by using squeezed states of light.Nat. Phot., 7:613–619, 2013. doi:10.1038/nphoton.2013.177

    doi:10.1038/nphoton.2013.177 2013

  2. [2]

    Adnan A. E. Hajomer, Ivan Derkach, Radim Filip, Ulrik L. Andersen, Vladyslav C. Usenko, and Tobias Gehring. Continuous-variable quantum passive optical network.Light Sci. Appl., 13(1), Oct 2024. ISSN 2047-7538. doi:10.1038/s41377-024-01633-9. URLhttp: //dx.doi.org/10.1038/s41377-024-01633-9

    doi:10.1038/s41377-024-01633-9 2024

  3. [3]

    Non-clifford gate on optical qubits by nonlinear feedforward.Phys

    Shunya Konno, Warit Asavanant, Kosuke Fukui, Atsushi Sakaguchi, Fumiya Hanamura, Petr Marek, Radim Filip, Jun-ichi Yoshikawa, and Akira Furu- sawa. Non-clifford gate on optical qubits by nonlinear feedforward.Phys. Rev. Res., 3:043026, Oct 2021. doi:10.1103/PhysRevResearch.3.043026. URLhttps: //link.aps.org/doi/10.1103/PhysRevResearch.3. 043026. 7

    doi:10.1103/physrevresearch.3.043026 2021

  4. [4]

    Gaussian conversion protocols for cubic phase state generation.PRX Quantum, 2:010327, Feb

    Yu Zheng, Oliver Hahn, Pascal Stadler, Patric Holm- vall, Fernando Quijandr´ ıa, Alessandro Ferraro, and Giu- lia Ferrini. Gaussian conversion protocols for cubic phase state generation.PRX Quantum, 2:010327, Feb

  5. [5]

    URLhttps: //link.aps.org/doi/10.1103/PRXQuantum.2.010327

    doi:10.1103/PRXQuantum.2.010327. URLhttps: //link.aps.org/doi/10.1103/PRXQuantum.2.010327

    work page doi:10.1103/prxquantum.2.010327

  6. [6]

    Cerf, Timothy C

    Christian Weedbrook, Stefano Pirandola, Ra´ ul Garc´ ıa- Patr´ on, Nicolas J. Cerf, Timothy C. Ralph, Jeffrey H. Shapiro, and Seth Lloyd. Gaussian quantum informa- tion.Rev. Mod. Phys., 84:621–669, May 2012. doi: 10.1103/RevModPhys.84.621. URLhttps://link.aps. org/doi/10.1103/RevModPhys.84.621

    doi:10.1103/revmodphys.84.621 2012

  7. [7]

    Gaussian states and operations – a quick reference.arXiv:2102.05748 [quant-ph], 2021

    Jonatan Bohr Brask. Gaussian states and operations – a quick reference.arXiv:2102.05748 [quant-ph], 2021

    Pith/arXiv arXiv 2021

  8. [8]

    Positive wigner functions render classical simulation of quantum computation ef- ficient.Phys

    Andrea Mari and Jens Eisert. Positive wigner functions render classical simulation of quantum computation ef- ficient.Phys. Rev. Lett., 109:230503, Dec 2012. doi: 10.1103/PhysRevLett.109.230503. URLhttps://link. aps.org/doi/10.1103/PhysRevLett.109.230503

    doi:10.1103/physrevlett.109.230503 2012

  9. [9]

    Re- sources for bosonic quantum computational advan- tage.Phys

    Ulysse Chabaud and Mattia Walschaers. Re- sources for bosonic quantum computational advan- tage.Phys. Rev. Lett., 130:090602, Mar 2023. doi: 10.1103/PhysRevLett.130.090602. URLhttps://link. aps.org/doi/10.1103/PhysRevLett.130.090602

    work page doi:10.1103/physrevlett.130.090602 2023

  10. [10]

    Braunstein

    Seth Lloyd and Samuel L. Braunstein. Quan- tum computation over continuous variables. Phys. Rev. Lett., 82:1784–1787, Feb 1999. doi: 10.1103/PhysRevLett.82.1784. URLhttps: //link.aps.org/doi/10.1103/PhysRevLett.82.1784

    doi:10.1103/physrevlett.82.1784 1999

  11. [11]

    Encoding a qubit in an oscillator.Phys

    Daniel Gottesman, Alexei Kitaev, and John Preskill. Encoding a qubit in an oscillator.Phys. Rev. A, 64: 012310, Jun 2001. doi:10.1103/PhysRevA.64.012310. URLhttps://link.aps.org/doi/10.1103/PhysRevA. 64.012310

    doi:10.1103/physreva.64.012310 2001

  12. [12]

    Genuine continuous quan- tumness.arXiv:2503.07574 [quant-ph], 2025

    Vojtˇ ech Kala, Jiˇ r´ ı Fadrn´ y, Michal Neset, Jan B´ ılek, Petr Marek, and Miroslav Jeˇ zek. Genuine continuous quan- tumness.arXiv:2503.07574 [quant-ph], 2025

    arXiv 2025

  13. [13]

    Witnesses of non-gaussian fea- tures as lower bounds of stellar rank.arXiv:2603.03185 [quant-ph], 2026

    Jan Provazn´ ık, ˇSimon Br¨ auer, Vojtˇ ech Kala, Jarom´ ır Fiur´ aˇ sek, and Petr Marek. Witnesses of non-gaussian fea- tures as lower bounds of stellar rank.arXiv:2603.03185 [quant-ph], 2026

    arXiv 2026

  14. [14]

    Practical framework for conditional non-gaussian quan- tum state preparation.PRX Quantum, 1:020305, Oct

    Mattia Walschaers, Valentina Parigi, and Nicolas Treps. Practical framework for conditional non-gaussian quan- tum state preparation.PRX Quantum, 1:020305, Oct

  15. [15]

    URLhttps: //link.aps.org/doi/10.1103/PRXQuantum.1.020305

    doi:10.1103/PRXQuantum.1.020305. URLhttps: //link.aps.org/doi/10.1103/PRXQuantum.1.020305

    work page doi:10.1103/prxquantum.1.020305

  16. [16]

    Johansson, P.D

    J.R. Johansson, P.D. Nation, and Franco Nori. Qutip: An open-source python framework for the dynamics of open quantum systems.Computer Physics Communi- cations, 183(8):1760–1772, August 2012. ISSN 0010-

    2012

  17. [17]

    URLhttp://dx

    doi:10.1016/j.cpc.2012.02.021. URLhttp://dx. doi.org/10.1016/j.cpc.2012.02.021

    doi:10.1016/j.cpc.2012.02.021 2012

  18. [18]

    Cubic non- linear squeezing and its decoherence.Opt

    Vojtˇ ech Kala, Radim Filip, and Petr Marek. Cubic non- linear squeezing and its decoherence.Opt. Express, 30 (17):31456–31471, Aug 2022. doi:10.1364/OE.464759. URLhttps://opg.optica.org/oe/abstract.cfm?URI= oe-30-17-31456

    doi:10.1364/oe.464759 2022

  19. [19]

    Nonlinear squeezing generation via multi- mode pdc and single photon measurement.Optics Ex- press, 33(6):14000, March 2025

    Vojtˇ ech Kala, Denis Kopylov, Petr Marek, and Polina Sharapova. Nonlinear squeezing generation via multi- mode pdc and single photon measurement.Optics Ex- press, 33(6):14000, March 2025. ISSN 1094-4087. doi: 10.1364/oe.550358. URLhttp://dx.doi.org/10.1364/ OE.550358

    work page doi:10.1364/oe.550358 2025

  20. [20]

    Breum, Mikkel V

    Vojtˇ ech Kala, Casper A. Breum, Mikkel V. Larsen, Ulrik L. Andersen, Jonas S. Neergaard- Nielsen, Radim Filip, and Petr Marek. Nulli- fiers of non-gaussian cluster states through ho- modyne measurement.arXiv:2505.21066 [quant- ph], 2025. doi:10.48550/ARXIV.2505.21066. URL https://arxiv.org/abs/2505.21066

    doi:10.48550/arxiv.2505.21066 2025

  21. [21]

    Hierarchy of steer- ing criteria based on moments for all bipartite quantum systems.Phys

    Ioannis Kogias, Paul Skrzypczyk, Daniel Cavalcanti, An- tonio Ac´ ın, and Gerardo Adesso. Hierarchy of steer- ing criteria based on moments for all bipartite quantum systems.Phys. Rev. Lett., 115:210401, Nov 2015. doi: 10.1103/PhysRevLett.115.210401. URLhttps://link. aps.org/doi/10.1103/PhysRevLett.115.210401

    work page doi:10.1103/physrevlett.115.210401 2015

  22. [22]

    Hierarchy of nonlinear entanglement dynamics for continuous vari- ables.Phys

    Da Zhang, David Barral, Yin Cai, Yanpeng Zhang, Min Xiao, and Kamel Bencheikh. Hierarchy of nonlinear entanglement dynamics for continuous vari- ables.Phys. Rev. Lett., 127:150502, Oct 2021. doi: 10.1103/PhysRevLett.127.150502. URLhttps://link. aps.org/doi/10.1103/PhysRevLett.127.150502

    work page doi:10.1103/physrevlett.127.150502 2021

  23. [23]

    Covariance matrix entanglement criterion for an arbitrary set of operators.New Journal of Physics, 22(7): 073055, jul 2020

    Vinay Tripathi, Chandrashekar Radhakrishnan, and Tim Byrnes. Covariance matrix entanglement criterion for an arbitrary set of operators.New Journal of Physics, 22(7): 073055, jul 2020. doi:10.1088/1367-2630/ab9ce7. URL https://dx.doi.org/10.1088/1367-2630/ab9ce7

    work page doi:10.1088/1367-2630/ab9ce7 2020

  24. [24]

    Braunstein and Peter van Loock

    Samuel L. Braunstein and Peter van Loock. Quantum information with continuous vari- ables.Rev. Mod. Phys., 77:513–577, Jun 2005. doi:10.1103/RevModPhys.77.513. URLhttps: //link.aps.org/doi/10.1103/RevModPhys.77.513

    doi:10.1103/revmodphys.77.513 2005

  25. [25]

    Tam- ing numerical errors in simulations of continuous vari- able non-gaussian state preparation.Scientific Re- ports, 12(1), October 2022

    Jan Provazn´ ık, Radim Filip, and Petr Marek. Tam- ing numerical errors in simulations of continuous vari- able non-gaussian state preparation.Scientific Re- ports, 12(1), October 2022. ISSN 2045-2322. doi: 10.1038/s41598-022-19506-9. URLhttp://dx.doi.org/ 10.1038/s41598-022-19506-9

    work page doi:10.1038/s41598-022-19506-9 2022

  26. [26]

    Generation of quan- tum states with nonlinear squeezing by kerr nonlinear- ity.Opt

    ˇSimon Br¨ auer and Petr Marek. Generation of quan- tum states with nonlinear squeezing by kerr nonlinear- ity.Opt. Express, 29(14):22648–22658, Jul 2021. doi: 10.1364/OE.427637. URLhttp://www.opticsexpress. org/abstract.cfm?URI=oe-29-14-22648

    doi:10.1364/oe.427637 2021

  27. [27]

    Nonlin- ear squeezing for measurement-based non-gaussian op- erations in time domain.Phys

    Shunya Konno, Atsushi Sakaguchi, Warit Asavanant, Hisashi Ogawa, Masaya Kobayashi, Petr Marek, Radim Filip, Jun-ichi Yoshikawa, and Akira Furusawa. Nonlin- ear squeezing for measurement-based non-gaussian op- erations in time domain.Phys. Rev. Applied, 15:024024, Feb 2021. doi:10.1103/PhysRevApplied.15.024024. URL https://link.aps.org/doi/10.1103/PhysRevA...

    doi:10.1103/physrevapplied.15.024024 2021

  28. [28]

    Estimation of squeezing in a nonlinear quadrature of a mechanical oscillator.New Journal of Physics, 21 (11):113050, nov 2019

    Darren W Moore, Andrey A Rakhubovsky, and Radim Filip. Estimation of squeezing in a nonlinear quadrature of a mechanical oscillator.New Journal of Physics, 21 (11):113050, nov 2019. doi:10.1088/1367-2630/ab5690. URLhttps://doi.org/10.1088/1367-2630/ab5690. A. Physicality condition onγ It is important to ensure, that the higher order covari- ance matrix co...

    work page doi:10.1088/1367-2630/ab5690 2019