Pith. sign in

REVIEW 2 major objections 4 minor 58 references

This paper shows that the Robinet state—the best estimate of a quantum system given only time-binned measurement records—can be reconstructed using an explicitly completely positive Kraus expansion that is accurate to arbitrarily high order

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 · deepseek-v4-flash

2026-08-01 15:33 UTC pith:BD3DQVXJ

load-bearing objection Zero-mode dilation yields a solid, well-tested CP discretization of the Robinet state; the companion purity theorem has a real but fixable proof gap. the 2 major comments →

arxiv 2607.18392 v2 pith:BD3DQVXJ submitted 2026-07-20 quant-ph

Calculus of Robinet: completely positive reconstruction of time-averaged diffusive quantum trajectories

Hector Hutin , Antoine Tilloy This is my paper
classification quant-ph
keywords Robinet statecontinuous quantum measurementKraus operatorscompletely positive mapsstochastic master equationtransmission line dilationhigh-order discretizationtime-averaged trajectories
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.

This paper aims to make the Robinet state—the optimal quantum-state estimate conditioned only on time-binned, digitized measurement records—numerically computable to arbitrarily high accuracy in the bin size Δt while preserving the mathematically essential property of complete positivity. The authors achieve this by dilating the stochastic master equation into a unitary system-plus-transmission-line model, expanding the joint wavefunction in a Dyson series, Gram–Schmidt-orthogonalizing the resulting line states, and isolating the 'zero mode' whose quadrature is the binned signal. Measuring that mode yields the Robinet map as an explicit finite Kraus sum, accurate to order 2N+1; the average map is made exactly trace-preserving by an S^{-1/2} normalization. They verify the predicted scaling numerically up to order 10. The same machinery gives a sharp structural result: past the order at which a kinked, mode-entangled line state appears, no finite number of integrals of the signal against additional functions can keep the reconstructed state pure.

Core claim

At the core of the paper is the construction, for any half-integer N, of the map K_I(ρ)=Σ_μ P_μ(x)ρP_μ(x)†+O(Δt^{2N+1}), with P_μ(x)=Σ_k ⟨x|k⟩P_{k,μ} obtained by projecting a Dyson expansion of the system–line state onto a Gram–Schmidt-orthogonalized basis of the line. Because the sum is a genuine Kraus form, complete positivity holds by construction even after truncation, and the normalization S^{-1/2} restores an exactly trace-preserving average. This makes the Robinet state directly computable without an external solver, and the paper validates convergence on a random 7-level system, observing the expected orders through 2N=10. In the same framework, the purity-preservation property previ

What carries the argument

The central mechanism is the zero-mode decomposition H_line = H_0 ⊗ H_rest of the transmission line, where H_0 is spanned by the constant temporal mode g^(1)_1(t) = 1/√Δt. Measuring a quadrature of this mode is exactly the time-averaged homodyne signal that defines the Robinet state. The paper's construction expands the unitary evolution in a Dyson series, uses BCH normal-ordering and vacuum contractions to control the singular commutation relations [â(t),â†(u)] = δ(t−u), orthogonalizes the multi-photon line states by Gram–Schmidt, splits them into zero mode plus 'rest', and finally forms Kraus operators P_μ(x) by projecting onto the quadrature states ⟨x|k⟩. The workhorse identity is the til

Load-bearing premise

The whole coefficient hierarchy rests on the standard physics-level treatment of singular field commutators [â(t),â†(u)]=δ(t−u), the BCH normal-ordering at the dt scale with [â,â†]=1/dt, and the vacuum contraction that produces the tilted Liouvillian; if those manipulations hide a systematic error at some order, all Kraus coefficients inherit it while the order-10 numerics might still look convergent.

What would settle it

Independently run the explicit order-10 Kraus sum on a high-dimensional random system, checking (i) the Choi matrix of K_I is positive semidefinite for many random x and (ii) the geometric-mean error against a small-time-step SME reference scales as Δt^{10}; a single negative Choi eigenvalue or a fitted exponent below 10 would disprove the scheme's structural claims. Separately, to test the purity barrier, include integrals of the signal against all polynomial functionals up to degree, say, 6 and check whether the purity of an initially pure state decays as Δt^4 for a generic system; if it doe

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

If this is right

  • A single drop-in replacement for first-order SME discretization: reconstructing experimental time-binned records at low order is already principled, and stepping up to order 10 matches the exact quadrature/cascaded reference without an external solver.
  • The probability distribution of binned signals can be sampled directly at any order (as a Gaussian times a polynomial), enabling physical, law-accurate sampling of Robinet trajectories and pure-state unravelings with a discrete index plus continuous record.
  • Purity is lost at order 4 for a single monitored channel whenever [L,[L,G]] does not vanish, and no finite set of extra signal integrals can beat order 3; the obstruction is an infinite-rank kink state, not a matter of cleverly chosen functionals.
  • For two non-commuting jump operators, purity is lost already at order 2, and no additional measurement statistics can restore it; the same formalism identifies exactly when commuting pairs such as heterodyne detection evade this loss.

Where Pith is reading between the lines

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

  • An adaptive integrator could be built directly from the paper's data: the joint system–zero-mode density at successive orders is computable before sampling the record, so the difference across orders gives a bias-free error estimate and suggests automatic step-size control.
  • The same Gram–Schmidt/zero-mode machinery likely transfers to discretizing continuous matrix product states: choosing non-constant bin shapes (wavelets, finite elements) changes which line states are 'zero modes' and might bypass the infinite-mode obstruction at the price of a different measurement interpretation.
  • The kink-state argument suggests a general diagnostic: for any truncation of the line to finitely many modes, the accuracy floor is set by the lowest-order Kraus operator whose line state has an infinite-rank correlation kernel; systems where that operator vanishes are permissive, while generic systems are not.

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

2 major / 4 minor

Summary. The paper develops a completely positive (CP) discretization scheme for reconstructing time-averaged diffusive quantum trajectories (Robinet states). The construction dilates the stochastic master equation into a unitary system-plus-transmission-line dynamics, isolates a “zero mode” of the line whose homodyne measurement yields the Robinet state, expands the Dyson series, Gram–Schmidt orthogonalizes the resulting line states, and obtains explicit Kraus operators P_μ(x) such that the averaged map is CPTP to any order up to Δt^{10}. The scheme is benchmarked on a random 7-level system, with convergence verified against a cascaded-quantum-systems reference solver. The paper also claims a no-go result: the purity-preserving extension of Wonglakhon–Chantasri–Wiseman cannot reach order Δt^4 with finitely many integrated-signal statistics, because the required line state g'_6 is kinked and has an infinite one-body mode rank; an analogous claim is made for two non-commuting jump operators.

Significance. If the central construction is correct, the paper provides a practical, high-order, CPTP numerical integrator for the Robinet state that does not require an external solver, together with a transparent zero-mode picture of why purity is lost under time binning. The paper is explicit and reproducible: the symbolic derivations are automated, the code is archived, and the numerical verification is careful, using a random 7-level system and geometric-mean errors over 100 trajectories. The physical insight that the WCW purity result is tied to the zero-mode/first-mode truncation is appealing and likely correct in broad terms. However, one of the advertised theoretical byproducts—the no-go against Δt^4 purity with finitely many integrated statistics—is supported by an incomplete argument, as detailed below. This does not undermine the numerical scheme, but it does affect the strength of a headline claim and therefore needs correction.

major comments (2)
  1. [I.C and Appendix C 1] The claim that no finite number of integrated-signal statistics can preserve purity at order Δt^4 is not established by the proof given. Appendix C 1 shows that the line state g'_6 (Eq. (61)) has a one-body correlation kernel with no zero eigenvalue (Eqs. (C4)–(C8)), hence that the line cannot be represented with finitely many modes. But conditional purity of the system after a finite-mode measurement is not controlled by the internal mode rank of the line state; it is controlled by the linear dependence of the residual Kraus operators on the system. At order 4 the problematic contributions are P2(x) and P3(x) in Eqs. (100)–(101), both proportional to A=[L,[L,G]]. If, for a given finite set of measured modes, the operator produced by the measured part were proportional to A for every outcome, the conditional state would remain pure even though the tail has infinitely many modes. Appendix
  2. [VII.D and Appendix C 2] The same logical gap appears in the two-jump-operator case. The paper claims that when L1 and L2 do not commute, purity is lost at order 2 and no additional statistics can restore it. The supporting Appendix C 2 shows that the antisymmetric state |ψ⊥> (Eq. (147)) has a one-body correlation kernel with no zero eigenvalue (Eqs. (C11)–(C13)), so it populates infinitely many modes. As in the single-jump case, this only establishes that the line state is genuinely multimode. It does not exclude the possibility that all unmeasured tail Kraus operators are proportional to the operators obtained by measuring the zero modes plus any finite additional set. In that case the conditional state would remain pure despite the infinite tail. The distinction matters because the tail operator is M⊥ (Eq. (160)), and the paper does not prove that M⊥ is linearly independent of the operators generated by any f
minor comments (4)
  1. [Sec. II A, Eqs. (28)–(31)] The BCH manipulations at the dt scale with [a,a†]=1/dt are formal. A brief remark connecting these steps to the standard Ito/quantum-stochastic-calculus justification would improve rigor and accessibility. The numerical benchmarks mitigate the practical risk, but the formal presentation is faster than is ideal.
  2. [Figs. 1 and 2] The geometric-mean error used in Fig. 2b is defined in the text but not in the captions. Adding a one-sentence definition in the caption would be helpful, especially since the exact error convention affects the apparent scatter.
  3. [Appendix F] The printed expansion is extremely long. It may be preferable to move full symbolic expressions to the archived code or a supplementary file and retain only representative terms or recurrence relations in the appendix, to improve readability.
  4. [Sec. VI] The Cholesky-based S^{-1/2} normalization is described briefly. A sentence clarifying the domain of validity (S strictly positive) and the non-uniqueness of the polar decomposition in degenerate cases would be useful, although these issues do not arise in the reported numerical experiments.

Circularity Check

0 steps flagged

No significant circularity: the central CP construction is derived independently, with self-citations used mainly as consistency checks.

full rationale

The Robinet map K_I is initially quoted from [24], whose author list includes Tilloy, but the paper does not rely on that citation for its central construction. In Sec. II C (Eqs. 39–46), K_I is re-derived inside the system-plus-line dilation from the tilted Liouvillian, and in Sec. IV the Kraus operators P_μ(x) are obtained by projecting the Gram–Schmidt-orthogonalized Dyson expansion of the joint state, with no term imposed by hand to reproduce [24]'s expansion. The match with [24] is explicitly used as a consistency check ('we recover exactly the expansion of the Robinet state... proposed in [24]'), not as the source of the high-order completely positive scheme. The numerical validation is external: the reference solution comes from the cascaded-quantum-systems formalism (Appendix B), so the claimed order-10 accuracy is not a fitted-input prediction. The purity obstruction argument in Appendix C is an independent spectral/rank statement about line states; even if the step from infinite-rank correlation kernels to the non-existence of a finite additional measurement set has a proof gap, that is a correctness concern rather than a circularity. The only self-citation of note is [24], and it is not load-bearing because the map is re-derived and separately benchmarked.

Axiom & Free-Parameter Ledger

0 free parameters · 8 axioms · 2 invented entities

No free parameters are fitted: the order 2N and bin size Δt are algorithm inputs, and all Kraus coefficients are fixed by the Dyson expansion and Gram-Schmidt orthogonalization; η is a model input from experiments, and the random test matrices are benchmark choices, not part of the construction. The axioms are the standard input-output/quantum-stochastic-calculus toolkit of the field plus the spectral criterion for mode decompositions; none are ad hoc to this paper. The two listed 'entities' are mathematical objects introduced by the construction (a chosen mode and a kinked line state), not new physical postulates, and both carry independent falsifiable handles in the form of numerical benchmarks and the proven [L,[L,G]] condition.

axioms (8)
  • standard math Quantum stochastic calculus with operator-valued distributions: [â(t),â†(u)] = δ(t−u); BCH normal-ordering at the dt scale with [â,â†] = 1/dt; discarding terms of order higher than dt.
    Invoked in Sec. II A-B (Eqs. 27-29) to obtain the vacuum action U|ψ⟩⊗|vac⟩ and the tilted Liouvillian; the authors explicitly stay at 'the standard physics level of rigor'.
  • standard math Fock space over L²([0,Δt]) as the line Hilbert space, with n-photon inner products on the simplex S_Δt^n (Eq. 23).
    Sec. II A; the basis for the n-photon functional Dyson expansion (49).
  • standard math Validity of the time-ordered Dyson expansion (48) and the power counting: the state F^{(n)}_{k0..kn}|Ψ⟩ has order n/2 + Σk_i in Δt.
    Sec. III A; the entire truncation hierarchy and all final Kraus operators depend on this counting.
  • domain assumption The system+line Hamiltonian H_int = iLâ†(t) − iL†â(t) (Eq. 25) reproduces the SME (1)-(4) including measurement-record statistics.
    Sec. II B-C; the standard input-output dilation of continuous measurement; used to re-derive the Lindblad equation (Eq. 32) and the tilted Liouvillian (Eqs. 39-41).
  • domain assumption Homodyne readout is the projective measurement of the field quadrature with projector Π_{I_f} = δ(I_f − (Â[f]+†[f])) (Eq. 42), and the Robinet state is the posterior state of this measurement.
    Sec. II C; matches [24]'s stochastic-calculus definition; the characteristic-function equivalence (Eqs. 36-41) is the bridge between dilation and the closed form (8).
  • domain assumption The cascaded quantum system with coupling γ(t) = −1/√t makes the appended cavity acquire the zero-mode state exactly, up to Fock truncation.
    App. B; the independent reference solver for all benchmarks; its accuracy is asserted following [46].
  • standard math The correlation-kernel eigenmode expansion (App. C, Eqs. C1-C3) gives the minimal mode decomposition; a state needs finitely many modes iff its kernel has finitely many non-zero eigenvalues.
    App. C, citing [53]; both impossibility results (g'(2)_6 and |ψ_⊥⟩) rest entirely on this spectral criterion.
  • standard math Rejection sampling of polynomial×Gaussian densities using Gamma-distribution envelopes is statistically exact.
    App. D; supports the trajectory-sampling claims in Sec. V C-D.
invented entities (2)
  • Zero mode of the transmission line (mode function g^{(1)}_1 = 1/√Δt) independent evidence
    purpose: Isolating this mode lets a single quadrature measurement yield the Robinet map K_I in Kraus form (Eqs. 87-91).
    A mathematical decomposition (a chosen orthonormal line mode), not a new physical entity. Independent handle: the scheme built on it is benchmarked against the independent cascaded-systems solver and reproduces [24]'s expansion as a consistency check.
  • The kinked two-photon line state g'(2)_6 = 2(−Δt + 3|t1−t2|)/Δt² independent evidence
    purpose: It is the obstruction term (M'_6 = Δt²/12 [L,[L,G]]) preventing finite-mode purification at order Δt⁴.
    Derived, not postulated: it emerges from Gram-Schmidt of the Dyson expansion (Eqs. 60-61). Its infinite-mode property is proven in App. C1 via a correlation kernel with no null eigenvector, and its vanishing condition ([L,[L,G]]=0) is checkable; the WCW comparison provides external consistency.

pith-pipeline@v1.3.0-alltime-deepseek · 6040 in / 6624 out tokens · 243651 ms · 2026-08-01T15:33:54.245097+00:00 · methodology

0 comments
read the original abstract

Truly continuous quantum trajectories, obtained from homodyne or heterodyne readouts, can only ever be reconstructed approximately. The continuous measurement signal, needed for exact reconstruction, is averaged over bins of finite time $\Delta t$ during any analog to digital conversion step. The best reconstruction possible, knowing only this discrete record, was introduced recently and dubbed the Robinet state. In this article, we show how the Robinet state can be computed with a numerical discretization scheme that is completely positive, accurate to arbitrarily high order in $\Delta t$, and that does not rely on any other external solver. Our derivation relies on a dilation of the stochastic master equation into a system + transmission line setup, constructed in such a way that measuring what we call the "zero mode" of the line yields the Robinet state. We test the method on a challenging example with random Hamiltonian and jump operator, and verify its accuracy up to order $10$. Apart from its numerical interest, our approach provides a wealth of physical insights, extending in particular recent results on purity obtained by Wonglakhon, Chantasri, and Wiseman, that would be difficult to obtain in any other way.

Figures

Figures reproduced from arXiv: 2607.18392 by Antoine Tilloy, Hector Hutin.

Figure 1
Figure 1. Figure 1: FIG. 1. (a) Histogram of the integrated measurement record [PITH_FULL_IMAGE:figures/full_fig_p012_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2. (a) Expectation value [PITH_FULL_IMAGE:figures/full_fig_p013_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3. Histogram of integrated measurement record (blue) [PITH_FULL_IMAGE:figures/full_fig_p014_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4. Schematic of the cascaded system used to simulate the measurement back-action. The system [PITH_FULL_IMAGE:figures/full_fig_p022_4.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

58 extracted references · 1 canonical work pages · 1 internal anchor

  1. [1]

    S. J. Weber, A. Chantasri, J. Dressel, A. N. Jordan, K. W. Murch, and I. Siddiqi, Mapping the optimal route between two quantum states, Nature511, 570 (2014)

    2014

  2. [2]

    Negretti and K

    A. Negretti and K. Moelmer, Estimation of classical pa- rameters via continuous probing of complementary quan- tum observables, New Journal of Physics15, 125002 (2013)

    2013

  3. [3]

    P. Six, P. Campagne-Ibarcq, L. Bretheau, B. Huard, and P. Rouchon, Parameter estimation from measurements along quantum trajectories, in2015 54th IEEE Confer- ence on Decision and Control (CDC)(IEEE, 2015) pp. 7742–7748

    2015

  4. [4]

    Philippe,Measurement Back Action and Feedback in Superconducting Circuits, Ph.D

    C.-I. Philippe,Measurement Back Action and Feedback in Superconducting Circuits, Ph.D. thesis, Ecole Normale Superieure (ENS) (2015)

    2015

  5. [5]

    Monitoring the energy of a cavity by observing the emission of a repeatedly excited qubit

    H. Hutin, A. Essig, R. Assouly, P. Rouchon, A. Bienfait, and B. Huard, Monitoring the energy of a cavity by ob- serving the emission of a repeatedly excited qubit, arXiv preprint arXiv:2402.05046 10.48550/Arxiv.2402.05046 (2024)

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2402.05046 2024

  6. [6]

    E. B. Davies,Quantum Theory of Open Systems(Aca- demic Press, London, 1976)

    1976

  7. [7]

    C. W. Gardiner,Handbook of Stochastic Methods: For Physics, Chemistry and the Natural Sciences(Springer- Verlag, Berlin, 1985)

    1985

  8. [8]

    H. J. Carmichael,An Open Systems Approach to Quan- tum Optics(Springer, Berlin, 1993)

    1993

  9. [9]

    V. P. Belavkin, Measurement, filtering and control in quantum open dynamical systems, Reports on Mathe- matical Physics43, A405 (1999). 19

    1999

  10. [10]

    Breuer and F

    H.-P. Breuer and F. Petruccione,The Theory of Open Quantum Systems(Oxford University Press, New York, 2002)

    2002

  11. [11]

    Jacobs and D

    K. Jacobs and D. A. Steck, A straightforward introduc- tion to continuous quantum measurement, Contempo- rary Physics47, 279 (2006)

    2006

  12. [12]

    H. M. Wiseman and G. J. Milburn,Quantum Measure- ment and Control(Cambridge University Press, Cam- bridge, 2009)

    2009

  13. [13]

    Jacobs,Stochastic Processes for Physicists: Under- standing Noisy Systems(Cambridge University Press, Cambridge, 2010)

    K. Jacobs,Stochastic Processes for Physicists: Under- standing Noisy Systems(Cambridge University Press, Cambridge, 2010)

    2010

  14. [14]

    Attal and Y

    S. Attal and Y. Pautrat, From repeated to continu- ous quantum interactions, Annales Henri Poincar´ e7, 59 (2006)

    2006

  15. [15]

    Pellegrini, Existence, uniqueness and approximation of a stochastic Schr¨ o dinger equation: The diffusive case, The Annals of Probability36, 2332 (2008)

    C. Pellegrini, Existence, uniqueness and approximation of a stochastic Schr¨ o dinger equation: The diffusive case, The Annals of Probability36, 2332 (2008)

    2008

  16. [16]

    Pellegrini, Poisson and diffusion approximation of stochastic master equations with control, Annales Henri Poincar´ e10, 995 (2009)

    C. Pellegrini, Poisson and diffusion approximation of stochastic master equations with control, Annales Henri Poincar´ e10, 995 (2009)

    2009

  17. [17]

    Pellegrini, Markov chains approximation of jump- diffusion stochastic master equations, Annales de l’Institut Henri Poincar´ e, Probabilit´ es et Statistiques46, 924 (2010)

    C. Pellegrini, Markov chains approximation of jump- diffusion stochastic master equations, Annales de l’Institut Henri Poincar´ e, Probabilit´ es et Statistiques46, 924 (2010)

    2010

  18. [18]

    Ciccarello, S

    F. Ciccarello, S. Lorenzo, V. Giovannetti, and G. M. Palma, Quantum collision models: Open system dynam- ics from repeated interactions, Physics Reports954, 1 (2022)

    2022

  19. [19]

    C. W. Gardiner and M. J. Collett, Input and output in damped quantum systems: Quantum stochastic differen- tial equations and the master equation, Physical Review A31, 3761 (1985)

    1985

  20. [20]

    A. A. Clerk, M. H. Devoret, S. M. Girvin, F. Mar- quardt, and R. J. Schoelkopf, Introduction to quantum noise, measurement, and amplification, Reviews of Mod- ern Physics82, 1155 (2010)

    2010

  21. [21]

    B. Abdo, K. Sliwa, F. Schackert, N. Bergeal, M. Ha- tridge, L. Frunzio, A. D. Stone, and M. Devoret, Full Co- herent Frequency Conversion between Two Propagating Microwave Modes, Physical Review Letters110, 173902 (2013)

    2013

  22. [22]

    Flurin,The Josephson Mixer : A Swiss Army Knife for Microwave Quantum Optics, Ph.D

    E. Flurin,The Josephson Mixer : A Swiss Army Knife for Microwave Quantum Optics, Ph.D. thesis, Ecole nor- male sup´ erieure - ENS PARIS (2014)

    2014

  23. [23]

    Hutin,Measurement and Neural Network-Assisted Control of Quantum Microwave Modes with Supercon- ducting Circuits, Theses, Ecole normale sup´ erieure de lyon - ENS LYON (2024)

    H. Hutin,Measurement and Neural Network-Assisted Control of Quantum Microwave Modes with Supercon- ducting Circuits, Theses, Ecole normale sup´ erieure de lyon - ENS LYON (2024)

    2024

  24. [24]

    Guilmin, P

    P. Guilmin, P. Rouchon, and A. Tilloy, Time- averaged continuous quantum measurement (2025), arXiv:2505.20382 [quant-ph]

    Pith/arXiv arXiv 2025

  25. [25]

    J. F. Rouchon, Pierre ands Ralph, Efficient quantum fil- tering for quantum feedback control, Physical Review A 91, 012118 (2015)

    2015

  26. [26]

    Robin, P

    R. Robin, P. Rouchon, and L.-A. Sellem, Unconditionally stable time discretization of Lindblad master equations in infinite dimension using quantum channels (2025), arXiv:2503.01712, arXiv:2503.01712 [math]

    Pith/arXiv arXiv 2025

  27. [27]

    Y. Cao, M. He, and X. Li, Dynamically Optimal Unrav- eling Schemes for Simulating Lindblad Equations (2025), arXiv:2509.19887 [quant-ph]

    Pith/arXiv arXiv 2025

  28. [28]

    Appelo and Y

    D. Appelo and Y. Cheng, Kraus is King: High-order Completely Positive and Trace Preserving (CPTP) Low Rank Method for the Lindblad Master Equation (2025), arXiv:2409.08898, arXiv:2409.08898 [math]

    Pith/arXiv arXiv 2025

  29. [29]

    Wonglakhon, A

    N. Wonglakhon, A. Chantasri, and H. M. Wiseman, Quantum trajectories for time-binned data and their closeness to fully conditioned quantum trajectories (2026), arXiv:2601.10937 [quant-ph]

    arXiv 2026

  30. [31]

    von Waldenfels, Solution of the Hudson– Parthasarathy equation by ordered products, Stochastics 89, 818 (2017)

    W. von Waldenfels, Solution of the Hudson– Parthasarathy equation by ordered products, Stochastics 89, 818 (2017)

    2017

  31. [32]

    Haegeman, J

    J. Haegeman, J. I. Cirac, T. J. Osborne, and F. Ver- straete, Calculus of continuous matrix product states, Physical Review B88, 085118 (2013)

    2013

  32. [33]

    Barchielli, Input and output channels in quantum systems and quantum stochastic differential equations, inQuantum Probability and Applications III, edited by L

    A. Barchielli, Input and output channels in quantum systems and quantum stochastic differential equations, inQuantum Probability and Applications III, edited by L. Accardi and W. von Waldenfels (Springer, Berlin, Hei- delberg, 1988) pp. 37–51

    1988

  33. [34]

    Zoller and C

    P. Zoller and C. W. Gardiner, Quantum Noise in Quantum Optics: The Stochastic Schr¨ odinger Equation (1997), arXiv:quant-ph/9702030

    Pith/arXiv arXiv 1997

  34. [35]

    H. M. Wiseman and G. J. Milburn, Quantum theory of field-quadrature measurements, Physical Review A47, 642 (1993)

    1993

  35. [36]

    Barchielli, Stochastic differential equations and a pos- teriori states in quantum mechanics, International Jour- nal of Theoretical Physics32, 2221 (1993)

    A. Barchielli, Stochastic differential equations and a pos- teriori states in quantum mechanics, International Jour- nal of Theoretical Physics32, 2221 (1993)

    1993

  36. [37]

    Tilloy, Exact signal correlators in continuous quantum measurements, Physical Review A98, 010104 (2018)

    A. Tilloy, Exact signal correlators in continuous quantum measurements, Physical Review A98, 010104 (2018)

    2018

  37. [38]

    Guilmin, P

    P. Guilmin, P. Rouchon, and A. Tilloy, Correlation functions for realistic continuous quantum measurement, IF AC-PapersOnLine 22nd IF AC World Congress,56, 5164 (2023)

    2023

  38. [39]

    G. T. Landi, M. J. Kewming, M. T. Mitchison, and P. P. Potts, Current fluctuations in open quantum sys- tems: Bridging the gap between quantum continuous measurements and full counting statistics, PRX Quan- tum5, 020201 (2024)

    2024

  39. [40]

    Hutin and A

    H. Hutin and A. Tilloy, HHeda/robinet calculus: v1.0.0 (2026)

    2026

  40. [41]

    Verstraete and J

    F. Verstraete and J. I. Cirac, Continuous matrix product states for quantum fields, Physical Review Letters104, 190405 (2010)

    2010

  41. [42]

    Sperling, A

    J. Sperling, A. Perez-Leija, K. Busch, and C. Sil- berhorn, Mode-independent quantum entanglement for light, Physical Review A100, 62129 (2019)

    2019

  42. [43]

    C. E. Lopetegui, M. Isoard, N. Treps, and M. Walschaers, Detection of mode-intrinsic quantum entanglement, Op- tica Quantum3, 312 (2025)

    2025

  43. [44]

    Meurer, C

    A. Meurer, C. P. Smith, M. Paprocki, O. ˇCert ´ ık, S. B. Kirpichev, M. Rocklin, Am. Kumar, S. Ivanov, J. K. Moore, S. Singh, T. Rathnayake, S. Vig, B. E. Granger, R. P. Muller, F. Bonazzi, H. Gupta, S. Vats, F. Johans- son, F. Pedregosa, M. J. Curry, A. R. Terrel, ˇS. Rouˇ cka, A. Saboo, I. Fernando, S. Kulal, R. Cimrman, and A. Scopatz, SymPy: Symbolic ...

    2017

  44. [45]

    Cao and J

    Y. Cao and J. Lu, Structure-preserving numerical schemes for Lindblad equations, Journal of Scientific Computing102, 27 (2025)

    2025

  45. [46]

    A. H. Kiilerich and K. Mølmer, Input-Output Theory with Quantum Pulses, Physical Review Letters123, 20 123604 (2019)

    2019

  46. [47]

    Guilmin, A

    P. Guilmin, A. Bocquet, ´E. Genois, D. Weiss, and R. Gautier, Dynamiqs: An open-source Python library for GPU-accelerated and differentiable simulation of quantum systems (2025), preprint

    2025

  47. [48]

    J. R. Johansson, P. D. Nation, and F. Nori, QuTiP: An open-source Python framework for the dynamics of open quantum systems, Computer Physics Communications 183, 1760 (2012)

    2012

  48. [49]

    J. R. Johansson, P. D. Nation, and F. Nori, QuTiP 2: A Python framework for the dynamics of open quantum systems, Computer Physics Communications184, 1234 (2013)

    2013

  49. [50]

    Shankar, K

    A. Shankar, K. V. Acoleyen, and J. Haegeman, Finite- element matrix product states for continuum models in one dimension (2026), arXiv:2606.14873 [quant-ph]

    Pith/arXiv arXiv 2026

  50. [51]

    Kaplan and A

    M. Kaplan and A. Tilloy, Wavelet matrix product states for quantum fields (2026), arXiv:2606.23823 [quant-ph]

    Pith/arXiv arXiv 2026

  51. [52]

    H. M. Wiseman, SU(2) distribution functions and mea- surement of the fluorescence of a two-level atom, Quan- tum and Semiclassical Optics: Journal of the European Optical Society Part B7, 569 (1995)

    1995

  52. [53]

    Khanahmadi, M

    M. Khanahmadi, M. M. Lund, K. Mølmer, and G. Jo- hansson, Multimode character of quantum states released from a superconducting cavity, Physical Review Research 5, 43071 (2023). Appendix A: Robinet state for heterodyne readout The heterodyne measurement can be built by introducing an independent auxiliary transmission line with field operator ˆb(t), with ...

    2023

  53. [54]

    ThenG=K 2, soGandKshare the same eigenmodes

    State g′(2) 6 E In the case of the state g′(2) 6 E , the correlation function is given by G g′(2) 6 E(t1, t2) := D g′(2) 6 ˆa†(t1)ˆa(t2) g′(2) 6 E , τ:=|t 1 −t 2|,(C4) G g′(2) 6 E(t1, t2) = 1 ∆t4 4∆t3 −6∆t 2(t1 +t 2) + 12∆t(t1t2 −τ 2) + 12τ3 .(C5) LetKbe the integral operator with kernelK(t, s) =g ′(2) 6 (t, s). ThenG=K 2, soGandKshare the same eigenmodes...

  54. [55]

    State|ψ ⊥⟩ We provide here the rigorous proof that|ψ ⊥⟩is multimode. We look at the one-body correlation kernel of line 1: Starting from |ψ⊥⟩= 1√ 2 g1,2 2 E − g2,1 2 E ,(C9) one gets ˆa1(t)|ψ ⊥⟩= 1 ∆t Z ∆t 0 dt1 θ(t−t 1)−θ(t 1 −t) ˆa† 2(t1)|vac⟩.(C10) Hence, fort, u∈[0,∆t], the one-body correlation kernel of line 1 is G1 |ψ⊥⟩(t, u) :=⟨ψ⊥|ˆa† 1(t)ˆa1(u)|ψ ...

  55. [56]

    Ifkis even,w k =c k R ∞ −∞ R(xk)e−x2/θdx× θ(ck): this vanishes wheneverc k is negative, so such terms are simply dropped

    Choosekwith probability 1 N wk, wherew k := R ∞ −∞ R(ckxk)e−x2/θdx. Ifkis even,w k =c k R ∞ −∞ R(xk)e−x2/θdx× θ(ck): this vanishes wheneverc k is negative, so such terms are simply dropped. Ifkis odd, the two half-lines contribute equally by symmetry, andw k =|c k| R ∞ −∞ R(xk)e−x2/θdxregardless of the sign ofc k

  56. [57]

    Get a sampleXaccording to the distributionR(sign(c k)xk)e−x2/θ. Ifkis even (and greater than 0 whose case is trivial), this is done by getting a sampleYfollowing a Gamma distribution with shape (k+ 1)/2 and scaleθ, and settingX= √ YorX=− √ Ywith probability 1/2. Ifkis odd, this is done by getting a sampleYfrom a Gamma distribution with shape (k+ 1)/2 and ...

  57. [58]

    Lindblad Here are the operators for the Lindblad equation at order 2N= 8: M0 = ∆t8G8 40320 + ∆t7G7 5040 + ∆t6G6 720 + ∆t5G5 120 + ∆t4G4 24 + ∆t3G3 6 + ∆t2G2 2 + ∆tG+ 1 26 M1 = ∆t 15 2 GLG6 40320 + ∆t 15 2 G2LG5 40320 + ∆t 15 2 G3LG4 40320 + ∆t 15 2 G4LG3 40320 + ∆t 15 2 G5LG2 40320 + ∆t 15 2 G6LG 40320 + ∆t 15 2 G7L 40320 + ∆t 15 2 LG7 40320 + ∆t 13 2 GLG...

    2016

  58. [59]

    Robinet Here are the operators for the integrated measurement at order 2N= 6: P0,0 = ∆t6G6 720 + ∆t5G5 120 + ∆t4G4 24 + ∆t3G3 6 + ∆t2G2 2 + ∆tG+ 1 P1,0 = ∆t 11 2 GLG4 720 + ∆t 11 2 G2LG3 720 + ∆t 11 2 G3LG2 720 + ∆t 11 2 G4LG 720 + ∆t 11 2 G5L 720 + ∆t 11 2 LG5 720 + ∆t 9 2 GLG3 120 + ∆t 9 2 G2LG2 120 + ∆t 9 2 G3LG 120 + ∆t 9 2 G4L 120 + ∆t 9 2 LG4 120 + ...

    2016