Pith. sign in

REVIEW 2 major objections 4 minor 75 references

A reaction-coordinate embedding lets Bayesian thermometry work on continuously monitored non-Markovian quantum systems, with analytic Fisher information and 1/√τ scaling.

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 21:46 UTC pith:UBR2CE5U

load-bearing objection A genuinely useful extension of continuous-monitoring Bayesian estimation to non-Markovian dynamics via reaction coordinates; the core math holds, but the measured steady state is only spot-checked and the conclusions overreach a bit. the 2 major comments →

arxiv 2607.15978 v1 pith:UBR2CE5U submitted 2026-07-17 quant-ph

Parameter Estimation in a Continuously Monitored Non-Markovian Quantum System

Erik L. Andr\'e , Pharnam Bakhshinezhad , Patrick P. Potts , Luis A. Correa , Mohammad Mehboudi This is my paper
classification quant-ph MSC 81S2281P5081P15 PACS 03.65.Ta03.65.Yz03.67.-a05.30.-d
keywords continuous monitoringnon-Markovian dynamicsreaction coordinate mappingBayesian estimationFisher informationquantum thermometryGaussian measurementsKalman filter
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.

The paper tackles the open problem of estimating unknown parameters from a continuously monitored quantum system when the system's dynamics are non-Markovian (CP-indivisible). It argues that the obstacle can be bypassed by embedding the system in a larger 'system-plus-reaction-coordinate' unit whose dynamics are Markovian. The paper then develops a Bayesian estimation framework for this augmented unit, derives an analytic expression for the Fisher information of the continuous measurement record, and shows that estimation precision improves as the inverse square root of measurement time. The method is demonstrated by using homodyne detection on a Brownian oscillator to estimate the temperature of a strongly coupled bosonic bath, with numerical Monte Carlo verification.

Core claim

The central claim is that the reaction-coordinate (RC) mapping, a standard technique for Markovian embedding of structured environments, can restore complete positive divisibility in a suitably augmented system–RC unit even when the bare system dynamics are strongly non-Markovian. Because this augmented unit evolves under a GKLS master equation, the usual sequential Bayesian update of conditional Gaussian states applies. The paper derives the conditional dynamics via a quantum Kalman filter (Eqs. (8)–(9)) and shows that, in the long-time limit, the total Fisher information of the measurement record is given by F_total(θ) = τ λ Tr[LᵀV⁻¹L Q Σ_ss Qᵀ], where Σ_ss solves an augmented Lyapunov equ

What carries the argument

The central object is the reaction-coordinate mapping, which replaces the original bath by a collective mode (the reaction coordinate) coupled to a residual bath. When enough reaction coordinates are extracted, the residual bath becomes effectively unstructured and weakly coupled, so the system–RC unit obeys a global GKLS master equation. The conditional dynamics of this unit under a Gaussian measurement are then described by a quantum Kalman filter, with the measurement outcome likelihood given by a Gaussian whose covariance is set by the detector environment and measurement strength. The Fisher information is computed by tracking the sensitivity of the conditional mean through an augmented

Load-bearing premise

The load-bearing premise is that the system–reaction-coordinate unit truly undergoes Markovian (CP-divisible) dynamics under the Born–Markov–secular approximation in the eigenbasis of H_S–RC; if this effective Markovianity fails, the Kalman filter and Fisher-information formula no longer apply.

What would settle it

Compute the exact quantum-Langevin conditional steady-state covariance of the system for a parameter set outside the single benchmarked point (e.g., stronger system–bath coupling g or lower cutoff Λ) and compare it with the steady state predicted by the RC master equation and used in the Kalman gain. A mismatch at the level of the system's conditional covariance—rather than just the unconditional variances—would invalidate the Fisher-information formula.

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

If this is right

  • Any differentiable parameter of a linear Gaussian continuously monitored system with an RC-embeddable non-Markovian bath can be estimated with error ∝ 1/√τ, matching the standard quantum-limited scaling despite strong non-Markovianity.
  • The method provides an analytic formula for the Fisher information (Eq. (27)) that can be used to optimize Gaussian measurements (e.g., homodyne vs. heterodyne) and detector-environment states for a given estimation task.
  • The Bayesian framework directly applies to thermometry, enabling accurate temperature estimates from a continuous homodyne record even when the probe is strongly coupled to a structured bath.
  • The approach can be iterated: extracting multiple reaction coordinates extends the Markovian embedding to more complex spectral densities, broadening the class of non-Markovian systems amenable to continuous-monitoring metrology.
  • The derived Cramér–Rao bounds provide benchmarks for practical estimators and quantify the information-theoretic limits of continuously monitored non-Markovian sensors.

Where Pith is reading between the lines

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

  • The same machinery likely generalizes to parameter estimation in non-Markovian Gaussian systems beyond thermometry, such as estimating coupling strengths or bath spectral parameters, provided the RC construction remains valid.
  • A testable extension is to apply the framework to non-Gaussian measurements or non-linear dynamics; the RC embedding may still hold, but the Gaussian Kalman filter would need replacement by a more general filtering equation.
  • The analytic Fisher information could be used to design adaptive measurement strategies that choose the measurement strength λ or the detector state in real time to maximize information gain, a direction the paper mentions but does not pursue.
  • The calibration of the RC master equation against exact quantum-Langevin dynamics for a single parameter set suggests the method should be validated across a wider range of parameters, especially where the Markovian approximation for the S–RC unit is stressed.

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 proposes a reaction-coordinate (RC) embedding to extend Bayesian parameter estimation from continuous Gaussian measurements to non-Markovian (CP-indivisible) dynamics. The central idea is that by augmenting the original system with one or more collective bath modes, the system–RC unit becomes Markovian and admits a GKLS master equation, so that a quantum Kalman filter (Eqs. (8)–(9)) can be applied. The authors derive an analytic asymptotic Fisher information F_total(θ) = τ λ Tr[LᵀV⁻¹L QΣ_ss Qᵀ] (End Matter, Eq. (27)), where Σ_ss solves the augmented Lyapunov equation (26). They demonstrate the method on thermometry of a bosonic bath with a Lorentzian spectral density, using Monte Carlo simulations of the EMSLE and comparing with Bayesian Cramér–Rao bounds. The unconditional dissipative dynamics is benchmarked against the exact quantum-Langevin solution in Fig. 3.

Significance. If the central claim is correct, the paper makes a valuable contribution: it extends continuous-monitoring quantum metrology to a class of non-Markovian dynamics that are otherwise difficult to treat, and it provides an analytic, parameter-free expression for the asymptotic Fisher information. The combination of RC mapping with Gaussian quantum filtering is conceptually clean, the End-Matter derivation is coherent, and the code for the simulations is publicly available. The analytic derivation of the FI relies on standard martingale properties of conditional scores, which is a strength. The main caveat is that the GKLS embedding—the load-bearing assumption—is only validated against the exact solution for the unmonitored system, not for the measured conditional dynamics that enters the FI formula.

major comments (2)
  1. [End Matter, Eq. (27); Fig. 3] The total FI is computed from the steady-state solution Σ_ss of the augmented Lyapunov equation (26), which uses the GKLS drift and diffusion matrices from Appendix C2. The only external benchmark, Fig. 3, compares the unconditional position and momentum variances with the exact quantum-Langevin solution. The conditional steady state σ_ss solving Eq. (18) and the Kalman gain K(θ) entering F_total are never independently validated. Since the measurement strength in Fig. 2 is λ=Ω_S, the measurement-induced dissipation is not a small perturbation and could modify system–residual-bath correlations and memory times, potentially invalidating the Born–Markov–secular approximation in Appendix C2. This is a load-bearing gap: the central claim (Eq. (27)) depends on the accuracy of the GKLS matrices under continuous monitoring. Please provide a benchmark of the monitored steady state or a quantitat
  2. [Appendix A2, Eq. (A46)] The residual-bath representation requires the divergent counter-term δΩ_R² = 2γΛ/π. The numerics set Λ = 1000Ω_S, but no cutoff-sensitivity analysis is reported. The Fisher information (27) depends on the steady-state covariance of the monitored S–RC unit, which may inherit a Λ dependence through the GKLS drift/diffusion matrices. Without evidence that the FI and the measurement-modified steady state are stable as Λ→∞, the claim that the RC embedding bypasses CP-indivisibility in a cutoff-independent way is not fully established for the measured dynamics.
minor comments (4)
  1. [End Matter, Eq. (19)] The vanishing of the score cross-terms is asserted without derivation. A one-line argument based on the martingale property of conditional scores, or a reference, would make the step easier to check.
  2. [Fig. 2 caption] The sentence 'The bound holds only for unbiased estimators' refers to the TBCRB, but as written it could be misread as applying to the BCRB. Please clarify which bound is meant.
  3. [Eq. (C41)] The matrix notation in the normal-mode transformation is dense; a sentence specifying the dimensions of the matrices and the ordering of the quadrature vector would improve readability.
  4. [Eq. (15)] The prior contains a free parameter α, but its role and the value used in Fig. 2 are not stated. Please indicate how α is chosen.

Circularity Check

0 steps flagged

No significant circularity: the FI formula is derived analytically from the Gaussian-measurement model and the RC-GKLS approximation is benchmarked externally; no fitted input is relabeled as a prediction.

full rationale

The derivation chain is self-contained rather than circular. Starting from the Hamiltonian (2) and Lorentzian spectral density (11), the RC mapping is an exact canonical transformation (Appendix A). The GKLS drift and diffusion matrices (C51)-(C70) are obtained by an explicit Born-Markov-secular reduction, with parameters fixed by the Hamiltonian and residual spectral density; none are fitted to the estimation target. The Fisher information (27) is derived from the Gaussian likelihood (8) via standard identities (20)-(26), and the steady-state covariance is the fixed point of the Lyapunov equation (26), not an input. The numerical demonstration in Fig. 2 fixes all simulation parameters and compares Monte Carlo EMSLE against the analytic BCRB/TBCRB; it does not adjust the analytic curve to match the simulations. The RC master equation is independently checked against the exact quantum Langevin solution (Fig. 3). Self-citations (e.g., Refs. [16,28,56,64,65]) support contextual or standard Bayesian-thermometry formulas and are not load-bearing; the Gaussian FI identity is additionally grounded in external Refs. [61,62]. The main vulnerability—the conditional steady state entering the Kalman gain is not separately benchmarked against an exact monitored solution—is an unvalidated approximation in the measurement regime, not a circular reduction: Eq. (27) would be incorrect if the GKLS embedding fails, but its content is not equivalent to an input by construction.

Axiom & Free-Parameter Ledger

7 free parameters · 6 axioms · 0 invented entities

The central claim rests on three pillars: (i) the exact RC mapping for Lorentzian baths with Ohmic residual spectral density (Appendix A); (ii) the global GKLS master equation for the S-RC unit under Born-Markov-secular approximations (Appendix C2); (iii) Gaussianity of states and measurements preserving the Kalman-filter structure (Appendix B). Pillar (ii) is the least certain: its premises are validated only at one parameter set (Fig. 3), and the counter-term renormalization involves a divergent cutoff-dependent shift. No new physical entities are postulated; the reaction coordinate is an established construction (Refs. [36-39]). The model parameters γ, g, Ω_R, Λ, λ are hand-chosen demonstration values, not fitted to data.

free parameters (7)
  • Bath memory rate γ = γ = Ω_S = 1
    Hand-chosen demonstration parameter controlling the Lorentzian bath's memory time; not fitted to data. Sets the secular-approximation quality.
  • System-bath coupling strength g = g = √10
    Hand-chosen strong-coupling value (g/γ ≈ 3.16) placing the probe in the non-Markovian regime the paper targets.
  • Reaction-coordinate frequency Ω_R = Ω_R = √10 (renormalized)
    Hand-chosen peak frequency of the structured bath in Eq. (11).
  • Residual-bath cutoff Λ = Λ = 1000 Ω_S
    Chosen large enough to approximate the Ohmic limit; the renormalization shift δΩ_R² = 2γΛ/π diverges with Λ, so cutoff-independence of the claims is an assumption.
  • Measurement strength λ = λ = Ω_S
    Hand-chosen continuous-measurement rate, equal to the detector-environment damping rate in the model; enters the Kalman gain and the FI prefactor.
  • Bayesian prior bounds and shape = T_min = 0.1, T_max = 2, α = 0
    Choice of prior for the thermometry demonstration; affects the finite-time EMSLE and the prior-information term Q(T), not the FI itself.
  • Number of extracted reaction coordinates = one RC
    One RC suffices for a single Lorentzian peak; not guaranteed for more structured baths, so generality of the method depends on this count being finite and small.
axioms (6)
  • domain assumption Born-Markov-secular approximation in the eigenbasis of H_S-RC yields a valid global GKLS master equation (Appendix C2, Eqs. (C51)-(C55)).
    Load-bearing premise: all conditional dynamics, likelihoods, and the FI are computed from this master equation. Validated numerically against the exact Langevin solution only for one parameter set (Fig. 3) and only for unconditional dynamics.
  • domain assumption The residual bath spectral density is Ohmic with exponential cutoff, J_R(ω) = γω e^{-|ω|/Λ}, and the original Lorentzian bath is exactly reproduced in the Λ → ∞ limit after absorbing a divergent counter-term (Appendix A2).
    Exactness of the single-RC mapping for the Lorentzian bath, but the cutoff-dependent renormalization (bare RC frequency diverges with Λ) is a nontrivial technical premise; a footnote delegates the cancellation to Ref. [68].
  • domain assumption Gaussian measurements on the system quadrature and a Gaussian detector environment preserve Gaussianity of the conditional state, so the quantum Kalman filter (Eqs. (8)-(10)) and the divisibility of the augmented dynamics (Eq. (1)) hold.
    Restricts the framework to linear Gaussian systems with Gaussian measurements—stated in the abstract and main text as the paper's scope.
  • domain assumption Steady-state and leading-order-in-δt approximations in the FI derivation: V ≈ σ_env + σ_M, variance treated as θ-independent to leading order, and f(t) ≈ QΣ_ss Qᵀ for τ ≫ 1 (End Matter).
    The analytic FI and the 1/√τ scaling hold only asymptotically; short-time behavior is explicitly left to Refs. [59, 60].
  • standard math Caldeira-Leggett quantum Brownian motion model with position-position coupling and counter-term (Eqs. (2)-(4)).
    Background model, cited to Refs. [47, 48]; standard input, not invented here.
  • domain assumption For large residual-bath cutoff, the coherent potential-renormalization effect is approximately canceled by the Lamb shift (footnote 4).
    The GKLS generator uses physical (renormalized) frequencies; the validity of this cancellation is asserted with a numerical check 'in the regime considered' and a citation to Ref. [68].

pith-pipeline@v1.3.0-alltime-deepseek · 32278 in / 33924 out tokens · 346645 ms · 2026-08-01T21:46:09.734722+00:00 · methodology

0 comments
read the original abstract

Continuous monitoring is a powerful tool for analyzing quantum systems and is increasingly used as a non-demolition technique in quantum metrology. In this context, noisy data acquired through continuous monitoring can be used to precisely pinpoint unknown parameters of the system. However, extending the theoretical framework that connects these data to the underlying parameters beyond Markovian dynamics is notoriously difficult, primarily because such dynamics cannot be expressed as completely positive divisible maps. To overcome this, we propose a method based on the reaction coordinate mapping to extend these parameter-estimation techniques beyond the Markovian regime. Our approach specifically targets linear systems with non-Markovian dynamics undergoing Gaussian continuous measurements, such as homodyne detection. Within this framework, we analyze Bayesian estimation and provide an analytical expression for the Fisher information, alongside the asymptotic scaling of the estimation precision for arbitrary parameters. Finally, we demonstrate the efficacy of our method through the example of thermometry of a bosonic bath.

Figures

Figures reproduced from arXiv: 2607.15978 by Erik L. Andr\'e, Luis A. Correa, Mohammad Mehboudi, Patrick P. Potts, Pharnam Bakhshinezhad.

Figure 1
Figure 1. Figure 1: (a) We consider a continuously monitored system [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Top: A sample Monte Carlo simulation of the ho [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Comparison of the position and momentum variances evaluated using the exact quantum Langevin equation, the [PITH_FULL_IMAGE:figures/full_fig_p022_3.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

75 extracted references · 1 linked inside Pith

  1. [1]

    Tóth and I

    G. Tóth and I. Apellaniz, Quantum metrology from a quantum information science perspective, Journal of Physics A: Mathematical and Theoretical47, 424006 (2014)

    2014

  2. [2]

    M. G. A. Paris, Quantum estimation for quantum tech- nology, International Journal of Quantum Information 07, 125–137 (2009)

    2009

  3. [3]

    Giovannetti, S

    V. Giovannetti, S. Lloyd, and L. Maccone, Quantum metrology, Phys. Rev. Lett.96, 010401 (2006)

    2006

  4. [4]

    Giovannetti, S

    V. Giovannetti, S. Lloyd, and L. Maccone, Advances in quantum metrology, Nature Photonics5, 222–229 (2011)

    2011

  5. [5]

    B. M. Escher, R. L. de Matos Filho, and L. Davidovich, General framework for estimating the ultimate preci- sion limit in noisy quantum-enhanced metrology, Nature Physics7, 406–411 (2011)

    2011

  6. [6]

    Demkowicz-Dobrzański, J

    R. Demkowicz-Dobrzański, J. Kołodyński, and M. Guţă, The elusive Heisenberg limit in quantum-enhanced metrology, Nature Communications3, 1063 (2012)

    2012

  7. [7]

    Y. L. Len, T. Gefen, A. Retzker, and J. Kołodyński, Quantum metrology with imperfect measurements, Na- ture Communications13, 6971 (2022)

    2022

  8. [8]

    Alipour, M

    S. Alipour, M. Mehboudi, and A. T. Rezakhani, Quan- tum metrology in open systems: Dissipative Cramér-Rao bound, Phys. Rev. Lett.112, 120405 (2014)

    2014

  9. [9]

    G. T. Landi, D. Poletti, and G. Schaller, Nonequilibrium boundary-driven quantum systems: Models, methods, and properties, Rev. Mod. Phys.94, 045006 (2022)

    2022

  10. [10]

    Demkowicz-Dobrzański, M

    R. Demkowicz-Dobrzański, M. Jarzyna, and J. Kołodyński, Quantum limits in optical interfer- ometry, inProgress in Optics(Elsevier, 2015) p. 345–435

    2015

  11. [11]

    Jacobs and D

    K. Jacobs and D. A. Steck, A straightforward introduc- tion to continuous quantum measurement, Contempo- rary Physics47, 279–303 (2006)

    2006

  12. [12]

    Gammelmark and K

    S. Gammelmark and K. Mølmer, Bayesian parameter in- ference from continuously monitored quantum systems, Phys. Rev. A87, 032115 (2013)

    2013

  13. [13]

    Cabot, F

    A. Cabot, F. Carollo, and I. Lesanovsky, Continuous sensing and parameter estimation with the boundary time crystal, Phys. Rev. Lett.132, 050801 (2024)

    2024

  14. [14]

    Albarelli, M

    F. Albarelli, M. A. C. Rossi, D. Tamascelli, and M. G. Genoni, Restoring Heisenberg scaling in noisy quantum metrology by monitoring the environment, Quantum2, 110 (2018)

    2018

  15. [15]

    M. A. C. Rossi, F. Albarelli, D. Tamascelli, and M. G. Genoni, Noisy quantum metrology enhanced by continu- ous nondemolition measurement, Phys. Rev. Lett.125, 200505 (2020)

    2020

  16. [16]

    Mehboudi, F

    M. Mehboudi, F. Meier, M. Huber, and H. J. D. Miller, Optimal limits of continuously monitored thermometers and their Hamiltonian structure, Phys. Rev. A111, L020401 (2025)

    2025

  17. [17]

    Vijay, C

    R. Vijay, C. Macklin, D. H. Slichter, S. J. Weber, K. W. Murch, R. Naik, A. N. Korotkov, and I. Siddiqi, Stabi- lizing Rabi oscillations in a superconducting qubit using quantum feedback, Nature490, 77–80 (2012)

    2012

  18. [18]

    J. Kong, R. Jiménez-Martínez, C. Troullinou, V. G. Lucivero, G. Tóth, and M. W. Mitchell, Measurement- induced, spatially-extended entanglement in a hot, strongly-interacting atomic system, Nature Communica- tions11, 2415 (2020)

    2020

  19. [19]

    Zhang, Y.-x

    J. Zhang, Y.-x. Liu, R.-B. Wu, K. Jacobs, and F. Nori, Quantum feedback: Theory, experiments, and applica- tions, Physics Reports679, 1–60 (2017)

    2017

  20. [20]

    W. Khan, P. P. Potts, S. Lehmann, C. Thelander, K. A. Dick, P. Samuelsson, and V. F. Maisi, Efficient and continuous microwave photoconversion in hybrid cavity- semiconductor nanowire double quantum dot diodes, Na- ture Communications12, 5130 (2021)

    2021

  21. [21]

    Gammelmark and K

    S. Gammelmark and K. Mølmer, Fisher information and the quantum Cramér-Rao sensitivity limit of continuous measurements, Phys. Rev. Lett.112, 170401 (2014)

    2014

  22. [22]

    A. H. Kiilerich and K. Mølmer, Estimation of atomic in- teraction parameters by photon counting, Phys. Rev. A 89, 052110 (2014)

    2014

  23. [23]

    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

  24. [24]

    Kiely and G

    A. Kiely and G. Landi, Extracting filtered signal statis- tics of continuously measured quantum systems, Phys. Rev. A113, 032442 (2026)

    2026

  25. [25]

    Radaelli, J

    M. Radaelli, J. A. Smiga, G. T. Landi, and F. C. Binder, Parameter estimation for quantum jump unrav- eling, Quantum10, 1993 (2026)

    1993

  26. [26]

    A. H. Kiilerich and K. Mølmer, Bayesian parameter es- timation by continuous homodyne detection, Phys. Rev. A94, 032103 (2016)

    2016

  27. [27]

    Molenda, L

    M. Molenda, L. A. Clark, M. Płodzień, and J. Kołodyński, Unlocking photodetection for quan- tum sensing with Bayesian likelihood-free methods and deep learning (2026), arXiv:2602.19792

    arXiv 2026

  28. [28]

    Boeyens, B

    J. Boeyens, B. Annby-Andersson, P. Bakhshinezhad, G. Haack, M. Perarnau-Llobet, S. Nimmrichter, P. P. Potts, and M. Mehboudi, Probe thermometry with con- tinuous measurements, New Journal of Physics25, 123009 (2023)

    2023

  29. [29]

    Amorós-Binefa and J

    J. Amorós-Binefa and J. Kołodyński, Noisy atomic mag- netometry with Kalman filtering and measurement-based feedback, PRX Quantum6, 030331 (2025)

    2025

  30. [30]

    Amorós-Binefa, M

    J. Amorós-Binefa, M. W. Mitchell, and J. Kołodyński, Tracking time-varying signals with quantum-enhanced atomic magnetometers (2025), arXiv:2503.14793

    arXiv 2025

  31. [31]

    H. M. Wiseman and J. M. Gambetta, Pure-state quan- tum trajectories for general non-Markovian systems do not exist, Phys. Rev. Lett.101, 140401 (2008)

    2008

  32. [32]

    Breuer, E.-M

    H.-P. Breuer, E.-M. Laine, J. Piilo, and B. Vacchini, Colloquium: Non-Markovian dynamics in open quantum systems, Rev. Mod. Phys.88, 021002 (2016)

    2016

  33. [33]

    S. Milz, M. S. Kim, F. A. Pollock, and K. Modi, Com- pletely positive divisibility does not mean Markovianity, Phys. Rev. Lett.123, 040401 (2019)

    2019

  34. [34]

    Bylicka, M

    B. Bylicka, M. Johansson, and A. Acín, Constructive method for detecting the information backflow of non- Markovian dynamics, Phys. Rev. Lett.118, 120501 (2017)

    2017

  35. [35]

    A. Garg, J. N. Onuchic, and V. Ambegaokar, Effect of friction on electron transfer in biomolecules, Journal of Chemical Physics83, 4491 (1985)

    1985

  36. [36]

    Iles-Smith, N

    J. Iles-Smith, N. Lambert, and A. Nazir, Environmental dynamics, correlations, and the emergence of noncanon- 7 ical equilibrium states in open quantum systems, Phys. Rev. A90, 032114 (2014)

    2014

  37. [37]

    Nazir and G

    A. Nazir and G. Schaller, The reaction coordinate map- ping in quantum thermodynamics, inThermodynam- ics in the Quantum Regime, Fundamental Theories of Physics, Vol. 195, edited by F. Binder, L. A. Correa, C. Gogolin, J. Anders, and G. Adesso (Springer, Cham,

  38. [38]

    Strasberg, G

    P. Strasberg, G. Schaller, N. Lambert, and T. Brandes, Nonequilibrium thermodynamics in the strong coupling and non-Markovian regime based on a reaction coordi- nate mapping, New Journal of Physics18, 073007 (2016)

    2016

  39. [39]

    L. A. Correa, B. Xu, B. Morris, and G. Adesso, Push- ing the limits of the reaction-coordinate mapping, The Journal of Chemical Physics151, 094107 (2019)

    2019

  40. [40]

    Mahadeviya, S

    K. Mahadeviya, S. V. Moreira, S. P. Mandal, M. Pan- dit, J. Prior, and M. T. Mitchison, Current fluctuations in nonequilibrium open quantum systems beyond weak coupling: a reaction coordinate approach, New Journal of Physics28, 044511 (2026)

    2026

  41. [41]

    Martinazzo, B

    R. Martinazzo, B. Vacchini, K. H. Hughes, and I. Burghardt, Communication: Universal Markovian re- duction of Brownian particle dynamics, Journal of Chem- ical Physics134, 011101 (2011)

    2011

  42. [42]

    Lindblad, On the generators of quantum dynamical semigroups, Communications in Mathematical Physics 48, 119 (1976)

    G. Lindblad, On the generators of quantum dynamical semigroups, Communications in Mathematical Physics 48, 119 (1976)

    1976

  43. [43]

    Gorini, A

    V. Gorini, A. Kossakowski, and E. C. G. Sudarshan, Completely positive dynamical semigroups of n-level sys- tems, Journal of Mathematical Physics17, 821 (1976)

    1976

  44. [44]

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

    2009

  45. [45]

    Jacobs, Continuous measurement, inQuantum Mea- surement Theory and its Applications(Cambridge Uni- versity Press, 2014) p

    K. Jacobs, Continuous measurement, inQuantum Mea- surement Theory and its Applications(Cambridge Uni- versity Press, 2014) p. 90–159

    2014

  46. [46]

    Annby-Andersson, F

    B. Annby-Andersson, F. Bakhshinezhad, D. Bhat- tacharyya, G. De Sousa, C. Jarzynski, P. Samuelsson, and P. P. Potts, Quantum Fokker-Planck master equa- tion for continuous feedback control, Phys. Rev. Lett. 129, 050401 (2022)

    2022

  47. [47]

    Weiss,Quantum Dissipative Systems, 4th ed

    U. Weiss,Quantum Dissipative Systems, 4th ed. (World Scientific, 2012)

    2012

  48. [48]

    Caldeira and A

    A. Caldeira and A. Leggett, Quantum tunnelling in a dissipative system, Annals of Physics149, 374 (1983)

    1983

  49. [49]

    Serafini,Quantum continuous variables: a primer of theoretical methods(CRC Press, 2017)

    A. Serafini,Quantum continuous variables: a primer of theoretical methods(CRC Press, 2017)

    2017

  50. [50]

    Weedbrook, S

    C. Weedbrook, S. Pirandola, R. García-Patrón, N. J. Cerf, T. C. Ralph, J. H. Shapiro, and S. Lloyd, Gaussian quantum information, Rev. Mod. Phys.84, 621 (2012)

    2012

  51. [51]

    De Sousa, P

    G. De Sousa, P. Bakhshinezhad, B. Annby-Andersson, P. Samuelsson, P. P. Potts, and C. Jarzynski, Continuous feedback protocols for cooling and trapping a quantum harmonic oscillator, Phys. Rev. E111, 014152 (2025)

    2025

  52. [52]

    J. B. Brask, Gaussian states and operations – a quick reference (2022), arXiv:2102.05748

    Pith/arXiv arXiv 2022

  53. [53]

    Geremia, J

    J. Geremia, J. K. Stockton, A. C. Doherty, and H. Mabuchi, Quantum Kalman filtering and the Heisen- berg limit in atomic magnetometry, Phys. Rev. Lett.91, 250801 (2003)

    2003

  54. [54]

    P. P. Hofer, M. Perarnau-Llobet, L. D. M. Miranda, G. Haack, R. Silva, J. B. Brask, and N. Brunner, Marko- vian master equations for quantum thermal machines: local versus global approach, New Journal of Physics19, 123037 (2017)

    2017

  55. [55]

    J. O. González, L. A. Correa, G. Nocerino, J. P. Palao, D. Alonso, and G. Adesso, Testing the validity of the ‘local’ and ‘global’ GKLS master equations on an exactly solvable model, Open Systems & Information Dynamics 24, 1740010 (2017)

    2017

  56. [56]

    Rubio, J

    J. Rubio, J. Anders, and L. A. Correa, Global quantum thermometry, Phys. Rev. Lett.127, 190402 (2021)

    2021

  57. [57]

    Rubio, Quantum scale estimation, Quantum Sci

    J. Rubio, Quantum scale estimation, Quantum Sci. Tech- nol.8, 015009 (2022)

    2022

  58. [58]

    Boeyens, S

    J. Boeyens, S. Seah, and S. Nimmrichter, Uninformed Bayesian quantum thermometry, Phys. Rev. A104, 052214 (2021)

    2021

  59. [59]

    M. G. Genoni, Cramér-Rao bound for time-continuous measurements in linear Gaussian quantum systems, Phys. Rev. A95, 012116 (2017)

    2017

  60. [60]

    Yokomizo, A

    K. Yokomizo, A. A. Clerk, and Y. Ashida, Fundamental limitsofcontinuousGaussianquantummetrology(2026), arXiv:2601.13554

    arXiv 2026

  61. [61]

    Malagò and G

    L. Malagò and G. Pistone, Information geometry of the Gaussian distribution in view of stochastic optimization, inProceedings of the 2015 ACM Conference on Foun- dations of Genetic Algorithms XIII, FOGA ’15 (ACM,

    2015

  62. [62]

    R. J. Ober, The Fisher information matrix for linear sys- tems, Systems & Control Letters47, 221 (2002)

    2002

  63. [63]

    Navarro, S

    J. Navarro, S. Morelli, M. Sanz, and M. Mehboudi, On super additivity of Fisher information in fully Gaussian metrology (2025), arXiv:2512.20534

    arXiv 2025

  64. [64]

    M. R. Jørgensen, J. Kołodyński, M. Mehboudi, M. Perarnau-Llobet, and J. B. Brask, Bayesian quan- tumthermometrybasedonthermodynamiclength,Phys. Rev. A105, 042601 (2022)

    2022

  65. [65]

    Mehboudi, M

    M. Mehboudi, M. R. Jørgensen, S. Seah, J. B. Brask, J. Kołodyński, and M. Perarnau-Llobet, Fundamental limits in Bayesian thermometry and attainability via adaptive strategies, Phys. Rev. Lett.128, 130502 (2022)

    2022

  66. [66]

    A. J. Leggett, Quantum tunneling in the presence of an arbitrary linear dissipation mechanism, Phys. Rev. B30, 1208 (1984)

    1984

  67. [67]

    Lampo, S

    A. Lampo, S. H. Lim, M. A. García-March, and M. Lewenstein, Bose polaron as an instance of quantum Brownian motion, Quantum1, 30 (2017)

    2017

  68. [68]

    L. A. Correa and J. Glatthard, Potential renormalisation, Lamb shift and mean-force Gibbs state – to shift or not to shift?, Quantum9, 1868 (2025)

    2025

  69. [69]

    Nicacio, M

    F. Nicacio, M. Paternostro, and A. Ferraro, Determin- ing stationary-state quantum properties directly from system-environment interactions, Phys. Rev. A94, 052129 (2016). 8 END MATTER The Fisher Information of Gaussian Continuous Measurements and the Asymptotic Behavior In this End Matter, we provide the explicit derivation of the ultimate precision boun...

    2016

  70. [70]

    P 2 R + g2 δΩ2 S XR − δΩ2 S g xS 2# + 1 2 X α̸=R

    Derivation of the Reaction Coordinate Hamiltonian We begin by explicitly writing the original system-bath Hamiltonian H= 1 2 p2 S + (Ω2 S +δΩ 2 S)x2 S + 1 2 X j (p2 j +ω 2 j x2 j )−x S X j cjxj.(A1) Here, the system is a harmonic oscillator with bare frequencyΩS, which interacts with a bosonic bath (i.e., also formed by harmonic oscillators) through a pos...

  71. [71]

    Effective Spectral Density of the Reaction Coordinate In this appendix, we derive the exact relation between the original bath spectral density,J(ω), and the effective reaction coordinate spectral density,JR(ω). To achieve this, we compute the Fourier-space propagators for the system oscillator in both the original and transformed coordinate frames and re...

  72. [72]

    This yields a generalized quantum Langevin equation with a non-local memory kernel, from which both the transient dynamics and the steady state are obtained

    Exact Dynamics via the Generalized Langevin Equation We begin by exactly solving the dynamics of the system by tracing out the bath degrees of freedom. This yields a generalized quantum Langevin equation with a non-local memory kernel, from which both the transient dynamics and the steady state are obtained. This exact solution serves as a benchmark for t...

  73. [73]

    Global GKLS Master Equation for the System–RC Unit We now derive a global master equation for the combined system–reaction coordinate unit. Since the system and reaction coordinate interact strongly, we first diagonalize the augmented system Hamiltonian HS−RC = 1 2 (p2 S + Ω2 1x2 S) + 1 2 (P 2 R + Ω2 2X 2 R)−gx SXR,(C38) 19 where we have definedΩ2 1 := Ω2...

  74. [74]

    The GKLS Master Equation for the Bare System We now formulate the standard local GKLS master equation for the original system-bath Hamiltonian given in Eq. (A1). As discussed in the main text, this standard approach fails in the strong-coupling regime, which necessitates the use of the global RC master equation derived in Appendix C2. To solve the open-sy...

  75. [75]

    (C12) and (C15) and compare them against both the global RC master equation (Appendix C2) and the local bare master equation (Appendix C3)

    Comparison of Exact and RC-Mapped Dynamics We now numerically evaluate the exact covariances from Eqs. (C12) and (C15) and compare them against both the global RC master equation (Appendix C2) and the local bare master equation (Appendix C3). This comparison explicitly demonstrates the superiority of the reaction coordinate mapping in the strong-coupling ...