Pith. sign in

REVIEW 2 major objections 5 minor 68 references

Dynamically suppressing cavity dephasing induced by frequency fluctuations of a coupled nonlinear mode

T0 review · 2 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read A weak detuned drive on a transmon cancels the flux-noise dephasing it imposes on a coupled cavity, lifting the cavity dephasing time from about 0.1 ms to about 2 ms.

desk verdict SAFE is a clever and well-analyzed drive-only protocol for protecting cavities from flux noise, but the headline 20x dephasing improvement rests on a 5-μs exponential fit that does not rule out a longer-time residual. read the letter →

arxiv 2608.04494 v1 pith:K234443X submitted 2026-08-05 quant-ph

Yunwei Lu , Xinyuan You , Ziwen Huang , Sébastien Léger , Jens Koch , Yao Lu This is my paper · ORCID
classification quant-ph
keywords cavitydephasingfluxnoiseACStarkshiftdynamicalsweetspotsuperconductingflux-tunabletransmonbosonicerrorcorrectionphoton-lossbias
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper introduces Stark-Assisted Flux-noise Evasion (SAFE), a protocol that protects a superconducting cavity from dephasing inherited from a flux-tunable transmon used to control it. The mechanism is a weak, off-resonant microwave drive on the transmon: the drive's AC Stark shift moves the cavity transition frequency to a dynamical sweet spot where its first-order sensitivity to flux noise vanishes. In Monte Carlo simulations with realistic 1/f flux noise, SAFE extends the cavity pure-dephasing time from about 0.1 ms to about 2 ms, roughly a 20-fold improvement, while leaving cavity relaxation and photon-loss bias essentially unchanged. A sympathetic reader would care because inherited dephasing erodes the photon-loss bias that bosonic error-correction codes rely on, and SAFE removes it without extra hardware: just a drive on the element already present.

What carries the argument

The load-bearing object is the rotating-frame Hamiltonian $H_R = \Delta_{\rm bd} b^\dagger b + (K/2) b^{\dagger 2} b^2 + \omega_c c^\dagger c + \chi b^\dagger b c^\dagger c + \Omega_0(b+b^\dagger)$ together with its perturbative eigenvalues, which give the shifted cavity transition frequency $\tilde{\omega}_{c;n,m}$ and its flux susceptibility $D_{n,m} = \partial \tilde{\omega}_{c;n,m}/\partial\Phi$. The cancellation works because both the static Lamb shift (proportional to $g^2/\Delta_{bc}^2$) and the drive-induced Stark shift (proportional to $\Omega_0^2 \chi/\Delta_{\rm bd}^2$) fluctuate with the transmon frequency $\partial\omega_b/\partial\Phi$; choosing $\Delta_{\rm bd}=(4\Omega_0^2 K)^{1/3}$ makes the two contributions cancel to first order. The paper supplements this with an effective Lindblad master equation, derived by Schrieffer–Wolff transformation and two-level truncation, whose residual pure-dephasing rate $\Gamma_{c\phi}^{\rm ss} = (\partial\omega_b/\partial\Phi)^2 A^2/(4|K/2\pi|) + (\gamma_{b\downarrow}/16)(\Delta_{\rm bd}/K)^2 + \gamma_{b\uparrow}$ quantifies the floor at the sweet spot and sets the trade-off between the number of protected Fock levels and the suppression factor.

What would settle it

Measure the cavity coherence time with the SAFE drive at $\Delta_{\rm bd}=(4\Omega_0^2 K)^{1/3}$ on a 3D cavity–FTT device with $1/f$ flux-noise amplitude near $10^{-5}\Phi_0$, and extend the Ramsey measurement well beyond the 5 $\mu$s fitting window, to hundreds of microseconds or longer. If the decay remains exponential with $T_2\approx 2$ ms, the extrapolation holds; if a Gaussian or quadratic low-frequency component appears and the apparent $T_\phi$ shortens, the claimed suppression factor is not sustained.

Watch

Extended reading notes

Core claim

SAFE's central discovery is that a continuous off-resonant drive applied to the nonlinear element can undo the dephasing that the same element's flux noise imposes on a dispersively coupled cavity. Because the dispersive shift makes the drive detuning photon-number dependent ($\Delta_{\rm bd}+m\chi$ for cavity state $|m\rangle$), different Fock states acquire different AC Stark shifts, and the shifted cavity transition frequency $\tilde{\omega}_{c;n,m} = (n-m)(\omega_c - g^2/\Delta_{bc}) - \Omega_0^2/(\Delta_{\rm bd}+n\chi) + \Omega_0^2/(\Delta_{\rm bd}+m\chi)$ acquires a flux dependence that can cancel the Lamb-shift contribution $(g^2/\Delta_{bc}^2)\,\delta\omega_b(t)$. Tuning the drive detuning to $\Delta_{\rm bd} = (4\Omega_0^2 K)^{1/3}$ (large-detuning regime) renders the transition frequency first-order insensitive to flux fluctuations, $D_{n,m}=0$, simultaneously for many Fock-state pairs; a separate condition $\Delta_{\rm bd} = \pm|\Delta_{bc}/g|\,\Omega_0$ protects a specific transition in the small-detuning regime. With representative 3D-cavity/FTT parameters and $1/f$ flux noise of amplitude $10^{-5}\Phi_0$, the paper's simulations show the cavity pure-dephasing time increasing from roughly 0.1 ms to 2 ms, and show that the same sweet spot suppresses logical dephasing in binomial and dual-rail encodings by about an order of magnitude without changing the erasure or photon-loss rate.

Load-bearing premise

The reported 20-fold suppression assumes that the residual dephasing seen in a 5-microsecond simulation window is the full story, meaning the coherence there is set by the Markovian channels of the effective master equation and continues to decay exponentially to the extracted 2 ms, with no slower unmodeled low-frequency or non-Markovian noise tail taking over at longer times.

Editorial extensions

If this is right

  • A single drive setting $\Delta_{\rm bd}=(4\Omega_0^2 K)^{1/3}$ simultaneously protects many Fock-state transition frequencies, so encodings like binomial codes that use several Fock-state coherences are protected without re-tuning per transition.
  • The cavity relaxation time and the photon-loss/erasure rate are essentially unchanged by the drive, so the photon-loss bias that cat, dual-rail, and binomial codes rely on is preserved while the dephasing channel is suppressed.
  • The residual pure-dephasing floor $\Gamma_{c\phi}^{\rm ss}$ grows with $\gamma_{b\downarrow}(\Delta_{\rm bd}/K)^2$, so reducing transmon relaxation or increasing anharmonicity $|K|$ directly improves the cavity dephasing time at the sweet spot.
  • A single driven transmon can act as a shared noise shield for multiple cavity modes, as demonstrated for a two-dual-rail-qubit Bell state, where logical dephasing is suppressed by about an order of magnitude and the decoherence-induced infidelity drops from 0.18 to 0.011.
  • Drive-induced side effects are subdominant in the analysis: drive-amplitude noise alone would limit $T_\phi$ to about 25 ms for one representative AWG platform, well above the 2 ms demonstrated.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The cancellation mechanism is not obviously restricted to transmon–cavity systems; a plausible extension is that any flux-tunable nonlinear coupler with a photon-number-dependent dispersive shift can be Stark-protected, making SAFE a general tool for shielding bosonic modes behind noise-prone couplers.
  • Because the sweet spot is created dynamically, a device could idle at a flux-bias point that is otherwise useful for parametric interactions rather than returning to a static sweet spot, a freedom the paper only gestures toward.
  • If the extrapolated 2 ms survives experimental scrutiny, SAFE could be combined with other protection techniques, such as two-tone drives or dynamical decoupling, to push residual dephasing toward the drive-amplitude-noise floor of tens of milliseconds.
  • A sharp test of the theory is the predicted non-monotonic dependence of the suppression factor on noise amplitude, with a maximum of roughly $\eta\approx 22$ near $A\approx 6\times 10^{-6}\Phi_0$; measuring $\eta(A)$ on a single device would discriminate the model from a generic dephasing-suppression story.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 5 minor

Summary. The paper introduces SAFE (Stark-Assisted Flux-noise Evasion), a protocol that applies a weak off-resonant microwave drive to a flux-tunable transmon dispersively coupled to a cavity, creating a dynamical sweet spot at which the dressed cavity transition frequency is first-order insensitive to 1/f flux noise. In the large-detuning regime the authors derive the simple condition Δbd = (4Ω0^2 K)^{1/3}, compute the residual Markovian dephasing channels, and support the analytical results with exact Floquet calculations and Monte Carlo simulations of the full Hamiltonian. They report an approximately 20-fold improvement in cavity pure-dephasing time (from about 0.1 ms to about 2 ms) and show that the same drive simultaneously protects multiple Fock-state coherences and multiple cavity modes, with applications to binomial and dual-rail encodings.

Significance. If the quantitative claims hold, SAFE is a practically attractive and hardware-efficient way to preserve the photon-loss bias of bosonic error-correcting codes while operating the nonlinear element away from its static flux sweet spot. The central sweet-spot condition is derived rather than fit, and the paper's key qualitative predictions are checked against independent Floquet numerics and Monte Carlo trajectory simulations. The extension to simultaneous protection of several Fock-state pairs and several cavity modes is a useful advance over single-transition dynamical sweet-spot schemes. The main quantitative claim, however, rests on an extrapolation from a very short coherence window, and the manuscript would be substantially strengthened by a longer-time numerical check or an analytical bound on the second-order flux-noise channel.

major comments (2)
  1. [Sec. III C, Fig. 3(d), App. G] The headline value Tφ≈2 ms is supported numerically only by a single-exponential fit to the first 5 µs of the ensemble-averaged coherence. At the claimed residual rate Γ≈500–600 s−1, the decay over 5 µs is only about 0.3% of the initial coherence, and the robustness check in App. G extends the window only to 10 µs, still roughly 200 times shorter than the extracted Tφ. The argument that a residual second-order sensitivity would produce a zero-slope quadratic onset only shows that such a component is small over 5–10 µs; it does not establish that its contribution at t≈1 ms is negligible relative to the Markovian channels in Eq. (38). Please extend the trajectory-averaged simulation to a time at which the decay is a measurable few percent, or provide an analytical estimate of the second-order longitudinal dephasing rate (for example, using ∂²ω̃c/∂Φ² and A² ln(ω_uv/ω_ir)) demonstrating that it is subdominant at millisecond times. As written, the 20-fold suppression factor is an extrapolation.
  2. [Sec. III B, Eq. (38)] Equation (38) is presented as the complete residual pure-dephasing rate at the dynamical sweet spot, but it contains no term representing dephasing from the low-frequency tail of the 1/f spectrum acting through the second- and higher-order flux sensitivity of the dressed cavity frequency. Because SAFE cancels only the first-order susceptibility D_{n,m}=∂ω̃c/∂Φ, the quadratic channel cannot be assumed negligible without a quantitative estimate. The paper should state the time range over which Eq. (38) is expected to hold and provide a bound on the second-order contribution, which is needed to justify the extrapolation from the short-time numerical window to Tφ≈2 ms.
minor comments (5)
  1. [Sec. III B] The cross-reference "Sec. II B 0 a" appears to be a typo; it should refer to Sec. II B, item a.
  2. [Sec. III C and App. G] The drive amplitude Ω0 used for the Monte Carlo and time-domain results in Fig. 3 is not stated; only Δbd/2π=−19 MHz is quoted. Please report Ω0 and, if it differs from the value implied by Eq. (22), the exact detuning used in the time-domain simulation, so that the results can be reproduced and compared with Eq. (38).
  3. [Appendix H, Eq. (H4)] The displayed inequality "τ≫ Ω0/|Δbd|² (22) = sqrt(1/(4KΔbd))" is confusingly formatted because the reference number "(22)" is embedded in the formula. Please rewrite it with the reference placed elsewhere.
  4. [Abstract and Sec. I] The phrase "1/fflux noise" is missing a space; it should read "1/f flux noise".
  5. [Appendix E] The stated rescaled coupling g/2π=0.10 GHz is slightly inconsistent with g=g′/(2φ_ZPF) using φ_ZPF≈0.24 and g′/2π=0.050 GHz, which gives about 0.104 GHz. Please clarify the rounding or the exact value of φ_ZPF used.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the SAFE sweet-spot condition and residual dephasing rate are derived analytically and then verified by full-Hamiltonian Monte Carlo simulation.

full rationale

The central derivation chain is self-contained rather than self-referential. The shifted cavity transition frequency is obtained from the rotating-frame Hamiltonian H_R (Eq. 16) by perturbation theory, giving Eq. (17); differentiating with respect to flux yields the susceptibility D_n,m (Eqs. 19–21). Setting D_n,m = 0 in the large-detuning regime algebraically produces the sweet-spot detuning Δ_bd = (4Θ_0^2 K)^{1/3} (Eq. 22), and the same procedure yields the small-detuning condition (Eq. 28). The residual cavity pure-dephasing rate at the sweet spot, Eq. (38), is obtained by substituting Eq. (22) into the independently derived rate expression Eq. (37), not by fitting. The numerical confirmation is also independent: Fig. 2 computes the susceptibility from Floquet quasienergies obtained with QuTiP using the full Hamiltonian, and the time-domain dephasing simulation in Sec. III C and App. G evolves the full Hamiltonian H_full with stochastically generated 1/f flux trajectories. The extracted T_φ ≈ 2 ms is an output of those trajectories, not an input to any part of the derivation. The only imported empirical constant is √|ℓn(ω_ir t)| ≈ 4, taken from external experimental work (Ref. [52]), and the flux-noise amplitude A is an external device parameter; neither is a renaming of this paper's own outputs. The authors cite their earlier dynamical-sweet-spot work (Ref. [43], with overlapping authors) for the general principle of continuous-drive sweet spots, but the new cavity-specific protection is derived here and checked against independent simulation, so that self-citation is not load-bearing. App. G's short-window exponential fit (5 μs, doubled to 10 μs) is a possible extrapolation risk for the claimed 2 ms number, but that is a robustness or correctness concern, not circularity: no fitted parameter is renamed as a prediction. Therefore no circular step can be exhibited from the paper's own equations, and the appropriate finding is no significant circularity.

Assumptions & free parameters 0 free parameters · 6 assumptions · 0 invented entities

The paper's claims rest on the standard dispersive circuit-QED model, the 1/f flux-noise assumption, and the weak-drive perturbative expansion; none of these are new entities, and no parameters are fitted to the target result. The main unquantified risk is the short-window extraction of the residual dephasing time.

assumptions (6)
  • domain assumption Transmon regime and dispersive approximation: EC<<EJ, g/|Δbc|<<1, g/|Δbc+K|<<1.
    Used to reduce the full circuit Hamiltonian to the dispersive Hamiltonian Eq. (11) and the noise Hamiltonian Eq. (10).
  • domain assumption Weak off-resonant drive: Ω0≪|Δbd| and |Δbd|≪|K| in the large-detuning SAFE regime.
    Justifies second-order perturbative treatment of the drive in Eq. (17) and the two-level truncation in Sec. III A.
  • domain assumption 1/f flux noise model with S(ω)=A^2/|ω/2π| and an infrared cutoff.
    The entire dephasing analysis is built on this noise spectrum; the Gaussian dephasing contribution uses sqrt(|ln(ω_ir t)|)≈4 from Ref. [52].
  • domain assumption FTT initialized in its ground state, with k_B T≪ℏω_b so that thermal excitation is negligible.
    Justifies neglecting excited-state population and the corresponding dispersive-shift fluctuation channel; used in Sec. III A.
  • domain assumption Markovian approximation for photon-shot noise and drive-induced excitation in Eq. (37).
    The photon-shot noise dephasing rate is set equal to the FTT excitation rate in the strong-dispersive limit χ≫γ_b↓, following Refs. [50,51].
  • domain assumption The two dual-rail cavities are mutually far detuned so FTT-mediated interactions and cross-Kerr terms are negligible.
    Used in the dual-rail model in Sec. IV B to justify treating each cavity mode independently.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Dynamically suppressing cavity dephasing induced by frequency fluctuations of a coupled nonlinear mode." pith.science (2026). https://pith.science/paper/K234443X

@misc{pith2026260804494,
  author       = {Pith},
  title        = {Pith review of: Dynamically suppressing cavity dephasing induced by frequency fluctuations of a coupled nonlinear mode},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/K234443X}},
  note         = {Machine review of arXiv:2608.04494}
}
abstract

High-coherence superconducting cavities offer a promising platform for quantum information, with long coherence times and negligible intrinsic dephasing. However, cavity control generally relies on nonlinear Josephson elements whose frequency fluctuations are inherited by the cavity as dephasing, potentially limiting control fidelities and eroding the noise bias used in error-correction protocols. Here, we introduce Stark-Assisted Flux-noise Evasion (SAFE), a hardware-efficient protocol that protects the cavity from inherited dephasing using only a weak off-resonant microwave drive applied to the nonlinear element. As a concrete setup, we analyze a 3D superconducting cavity dispersively coupled to a flux-tunable transmon (FTT) subject to $1/f$ flux noise. Analytical predictions are confirmed by Monte Carlo simulations with realistic parameters, which show that SAFE can extend the cavity dephasing time by more than an order of magnitude while keeping residual drive-induced decoherence subdominant.

Figures

Figures reproduced from arXiv: 2608.04494 by the authors.

Figure 1
Figure 1. Illustration of suppressing cavity dephasing by Stark-assisted flux-noise evasion (SAFE). [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Dynamical sweet spots generated by SAFE. Two [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Numerical evaluation of cavity pure-dephasing suppression and time-domain simulation of SAFE. [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Improvement of the coherence of bosonic encodings using SAFE (the drive amplitude is Ω [PITH_FULL_IMAGE:figures/full_fig_p009_4.png]
Figure 5
Figure 5. Figure 5: Schematic of the system of two dual-rail qubits [PITH_FULL_IMAGE:figures/full_fig_p010_5.png]
Figure 6
Figure 6. Figure 6: Dynamical sweet spot in a SNAIL–cavity system. [PITH_FULL_IMAGE:figures/full_fig_p016_6.png]
Figure 7
Figure 7. Figure 7: Power spectral density of δΦ(t) for 100 representa￾tive 1/f noise trajectories generated with an infrared cutoff ωmin/2π = 10 kHz and an ultraviolet cutoff ωmax = 1 GHz. b. Extraction of the dephasing time. The states dis￾cussed in the main text as exhibiting enhanced …
Figure 8
Figure 8. Figure 8: State-preparation infidelity I(τ ) as a function of ramp time τ for several detunings ∆bd, comparing Gaussian ramps with DRAG ramps. The nonmonotonic dependence reflects the trade-off between coherent leakage from unwanted transitions, which is suppressed for longer τ …

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

68 extracted references · 63 canonical work pages

  1. [1]

    Floquet Theorem Consider a Hilbert spaceHof dimensionNand a time- dependent HamiltonianH(t) with periodτ, satisfying H(t+τ) =H(t). Floquet’s theorem states that the so- lutions to the Schr¨ odinger equation can be written as |ϕn(t)⟩=e −iεnt|ψn(t)⟩,(A1) where{|ψ n(t)⟩}is aτ-periodic orthonormal basis and {εn}is the set of quasienergies

    work page

  2. [2]

    H for further discussion

    When the drive is turned on adiabatically, the states|g,0⟩and|g,1⟩are mapped onto the target Floquet states|ϕ g,0(t)⟩and|ϕ g,1(t)⟩, respectively; see App. H for further discussion. The coherence timeT 2 is extracted from the decay of ⟨X(t)⟩= Tr[¯ρ(t)X(t)],(41) X(t) =|ϕ g,1(t)⟩⟨ϕg,0(t)|+ h.c.,(42) where ¯ρ(t) is the density matrix expressed in a frame rota...

    work page

  3. [3]

    We then compute the Wigner functions of the initial state and the final states att= 300µs as shown in Fig

    Here we use dis- persively dressed states to define logical states |0L⟩= |g,0⟩+| g,4⟩√ 2 ,| 1L⟩=| g,2⟩, . We then compute the Wigner functions of the initial state and the final states att= 300µs as shown in Fig. 4(b). Without the drive, the Wigner function after 300µs is strongly distorted relative to the initial state, and the interference features asso...

    work page

  4. [4]

    This yields the Floquet eigenvalue equation H(t)−i∂ t |ψn(t)⟩=ε n|ψn(t)⟩.(A2) Since the quantities on both sides of Eq

    Construction of the Solution To obtain a differential equation relating the peri- odic states|ψ n(t)⟩and the quasienergiesε n to the HamiltonianH(t), we substitute the ansatz|Ψ(t)⟩= e−iεnt|ψn(t)⟩into the time-dependent Schr¨ odinger equa- tion. This yields the Floquet eigenvalue equation H(t)−i∂ t |ψn(t)⟩=ε n|ψn(t)⟩.(A2) Since the quantities on both sides...

    work page

  5. [5]

    (A16) 12 a

    Quasienergies and Floquet States of the Driven Dispersive Hamiltonian In this section, we use the procedure described above to approximate the quasienergies and Floquet states of the driven dispersive Hamiltonian given by H(t) = ¯ωbb†b+ K 2 b†2b2 + ¯ωcc†c +χb†bc†c+ 2Ω 0 cos(ωdt)(b+b †). (A16) 12 a. Quasienergy Operator for the Driven Dispersive Hamiltonia...

    work page

  6. [6]

    Kg4 ∆5 bc 2 + 3∆bc ∆bd −126 K2g6 ∆8 bc # + 42(n3−m 3)K2g6 ∆8 bc ) ∂ωb ∂Φ ≈

    Number of Protected Levels. To quantify the range of protected cavity levels, we compare the residual sensitivities. We define rn,m≡ deωc;n,m/dΦ deωc;n,m/dΦ|Ω0=0 .(B14) We say that the transition is protected within a tolerance ϵifr n,m < ϵ. At fourth order, the drive can cancel the linear flux susceptibility, so thatr n,m = 0 at the sweet spot. To estima...

    work page

  7. [7]

    γc↓ + g ∆bc 2 γb↓ # D[c]ρdisp +

    Multimode Sweet-Spot Alignment We now ask whether a single driven FTT can simulta- neously protect multiple cavity modes. In practice, for a fixed drive amplitude Ω0, one can sweep the drive detun- ing ∆bd and evaluate the flux susceptibility of each cavity transition. The minima of the susceptibilities for differ- ent cavity modes mightnot occur at exact...

    work page

  8. [8]

    Effective Drive Hamiltonian We rewriteH disp in terms of the number operators nb =b †bandn c =c †c. Using the relationb †2b2 = b†b(b†b−1) =n b(nb−1), we can express the unperturbed HamiltonianH 0 and the perturbationVas H0 = ∆bdnb + K 2 nb(nb−1) + ωcnc +χnbnc,(F5) V= Ω 0(b+b †).(F6) For a cleaner derivation, we define an operatorE(n b) that represents the...

    work page

Show all 68 references
  1. [9]

    γc↓ + g ∆bc 2 γb↓ # D[c′]ρd +γb↓D[b′]ρd +

    Transformation of the Dissipators and the Noise Hamiltonian In this section, we transform the dissipators and the noise Hamiltonian into the frame ofH d. The relevant operators areb,c,b †b, andc †c. The transformation of an operator can be approximated using the Baker– Campbel...

  2. [10]

    1− Ω0 ∆bd 2 1− 2χnc ∆bd ! σ− # ρd +γb↓D

    Effective Lindblad Master Equation in a Truncated Subspace To proceed, we truncate the FTT to a two-level system, nb∈{0,1}. This approximation is physically justified by three conditions. First, to minimize decoherence chan- nels acting on the logical subspace, the FTT is init...

  3. [11]

    Milul, B

    O. Milul, B. Guttel, U. Goldblatt,et al., PRX Quantum 4, 030336 (2023)

    2023

  4. [12]

    A. E. Oriani, F. Zhao, T. Roy,et al., arXiv:2403.00286 (2024)

    2024 arXiv

  5. [13]

    Romanenko, R

    A. Romanenko, R. Pilipenko, S. Zorzetti,et al., Phys. Rev. Appl.13, 034032 (2020)

    2020

  6. [14]

    Reagor, W

    M. Reagor, W. Pfaff, C. Axline,et al., Phys. Rev. B94, 014506 (2016)

    2016

  7. [15]

    T. Kim, T. Roy, X. You,et al., arXiv:2506.03286 (2025)

    2025 arXiv

  8. [16]

    F. C. Wellstood, C. Urbina, J. Clarke,et al., Appl. Phys. Lett.85, 5296 (2004)

    2004

  9. [17]

    J. V. Gates, M. A. Washington, M. Gurvitch,et al., J. Appl. Phys.55, 1419 (1984)

    1984

  10. [18]

    Krasnok, P

    A. Krasnok, P. Dhakal, A. Fedorov,et al., Appl. Phys. Rev.11, 011302 (2024)

    2024

  11. [19]

    N. Ofek, A. Petrenko, R. Heeres,et al., Nature536, 441 (2016)

    2016

  12. [20]

    V. V. Sivak, A. Eickbusch, B. Royer,et al., Nature616, 50 (2023)

    2023

  13. [21]

    Z. Ni, S. Li, X. Deng,et al., Nature616, 56 (2023)

    2023

  14. [22]

    L. Hu, Y. Ma, W. Cai,et al., Nat. Phys.15, 503 (2019)

    2019

  15. [23]

    Campagne-Ibarcq, A

    P. Campagne-Ibarcq, A. Eickbusch, S. Touzard,et al., Nature584, 368 (2020)

    2020

  16. [24]

    J. M. Gertler, B. Baker, J. Li,et al., Nature590, 243 (2021)

    2021

  17. [25]

    Leghtas, G

    Z. Leghtas, G. Kirchmair, B. Vlastakis,et al., Phys. Rev. Lett.111, 120501 (2013)

    2013

  18. [26]

    Leghtas, S

    Z. Leghtas, S. Touzard, I. M. Pop,et al., Science347, 853 (2015)

    2015

  19. [27]

    W. Cai, Y. Ma, W. Wang,et al., Fundam. Res.1, 50 (2021)

    2021

  20. [28]

    J. D. Teoh, P. Winkel, H. K. Babla,et al., Proc. Natl. Acad. Sci. U.S.A.120, e2221736120 (2023)

    2023

  21. [29]

    Putterman, K

    H. Putterman, K. Noh, C. T. Hann,et al., Nature638, 927 (2025)

    2025

  22. [30]

    W.-L. Ma, S. Puri, R. J. Schoelkopf,et al., Sci. Bull.66, 1789 (2021)

    2021

  23. [31]

    X. You, Y. Lu, T. Kim,et al., Phys. Rev. Appl.22, 044072 (2024)

    2024

  24. [32]

    Y. Lu, A. Maiti, J. W. O. Garmon,et al., Nat. Commun. 14, 5767 (2023)

    2023

  25. [33]

    Huang, T

    J. Huang, T. J. DiNapoli, G. Rockwood,et al., arXiv:2503.10623 (2025)

    2025 arXiv

  26. [34]

    B. J. Chapman, S. J. de Graaf, S. H. Xue,et al., PRX Quantum4, 020355 (2023)

    2023

  27. [35]

    Valadares, N.-N

    F. Valadares, N.-N. Huang, K. T. N. Chu,et al., Nat. Commun.15, 5816 (2024)

    2024

  28. [36]

    Goldblatt, N

    U. Goldblatt, N. Kahn, S. Hazanov,et al., Phys. Rev. X 14, 041056 (2024)

    2024

  29. [37]

    A. Z. Ding, B. L. Brock, A. Eickbusch,et al., Nat. Com- mun.16, 5279 (2025)

    2025

  30. [38]

    D. C. McKay, S. Filipp, A. Mezzacapo,et al., Phys. Rev. Appl.6, 064007 (2016)

    2016

  31. [39]

    Maiti, J

    A. Maiti, J. W. O. Garmon, Y. Lu,et al., PRX Quantum 6, 040326 (2025)

    2025

  32. [40]

    D. G. Atanasova, I. Yang, T. H¨ onigl-Decrinis,et al., PRX Quantum6, 020318 (2025)

    2025

  33. [41]

    Z. Li, E. Gupta, F. Zhao,et al., arXiv:2503.13953 (2025)

    2025

  34. [42]

    Valadares, A

    F. Valadares, A. Dorogov, T. Krisnanda,et al., arXiv:2602.18122 (2026)

    2026

  35. [43]

    Copetudo, A

    A. Copetudo, A. M. Kasper, T. Krisnanda,et al., arXiv:2603.15587 (2026)

    2026 arXiv

  36. [44]

    Neill, P

    C. Neill, P. Roushan, K. Kechedzhi,et al., Science360, 195 (2018)

    2018

  37. [45]

    Barends, J

    R. Barends, J. Kelly, A. Megrant,et al., Nature508, 500 (2014)

    2014

  38. [46]

    Gargiulo, S

    O. Gargiulo, S. Oleschko, J. Prat-Camps,et al., Appl. Phys. Lett.118, 012601 (2021)

    2021

  39. [47]

    Y. Sung, L. Ding, J. Braumueller,et al., Phys. Rev. X 11, 021058 (2021)

    2021

  40. [48]

    F. Yan, P. Krantz, Y. Sung,et al., Phys. Rev. Appl.10, 054062 (2018)

    2018

  41. [49]

    Stehlik, D

    J. Stehlik, D. M. Zajac, D. L. Underwood,et al., Phys. Rev. Lett.127, 080505 (2021). 24

    2021

  42. [50]

    Didier, E

    N. Didier, E. A. Sete, J. Combes,et al., Phys. Rev. Appl. 12, 054015 (2019)

    2019

  43. [51]

    S. S. Hong, A. T. Papageorge, P. Sivarajah,et al., Phys. Rev. A101, 012302 (2020)

    2020

  44. [52]

    P. S. Mundada, A. Gyenis, Z. Huang,et al., Phys. Rev. Appl.14, 054033 (2020)

    2020

  45. [53]

    Huang, P

    Z. Huang, P. S. Mundada, A. Gyenis,et al., Phys. Rev. Appl.15, 034065 (2021)

    2021

  46. [54]

    J. A. Valery, S. Chowdhury, G. Jones, and N. Didier, PRX Quantum3, 020337 (2022)

    2022

  47. [55]

    Blais, A

    A. Blais, A. L. Grimsmo, S. M. Girvin,et al., Rev. Mod. Phys.93, 025005 (2021)

    2021

  48. [56]

    Lambert, E

    N. Lambert, E. Gigu‘ere, P. Menczel,et al., Phys. Rep. 1153, 1 (2026)

    2026

  49. [57]

    M. D. Reed, L. DiCarlo, B. R. Johnson,et al., Phys. Rev. Lett.105, 173601 (2010)

    2010

  50. [58]

    S. P. Premaratne, F. C. Wellstood, B. S. Palmer,et al., Nat. Commun.8, 14148 (2017)

    2017

  51. [59]

    R. W. Heeres, P. Reinhold, N. Ofek,et al., Nat. Commun. 8, 94 (2017)

    2017

  52. [60]

    Z. Wang, S. Shankar, Z. K. Minev,et al., Phys. Rev. Appl.11, 014031 (2019)

    2019

  53. [61]

    A. A. Clerk and D. W. Utami, Phys. Rev. A75, 042302 (2007)

    2007

  54. [62]

    A. Kou, W. C. Smith, U. Vool,et al., Phys. Rev. X7, 031037 (2017)

    2017

  55. [63]

    M. H. Michael, M. Silveri, R. T. Brierley,et al., Phys. Rev. X6, 031006 (2016)

    2016

  56. [64]

    K. S. Chou, T. Shemma, H. McCarrick,et al., Nat. Phys. 20, 1454 (2024)

    2024

  57. [65]

    G. Zhu, D. G. Ferguson, V. E. Manucharyan, and J. Koch, Phys. Rev. B87, 024510 (2013)

    2013

  58. [66]

    Zurich Instruments AG,HDAWG User Manual: Specifi- cations(2026),https://docs.zhinst.com/hdawg_user_ manual/specifications.html

    2026

  59. [67]

    Timmer and M

    J. Timmer and M. Koenig, Astron. Astrophys.300, 707 (1995)

    1995

  60. [68]

    C. W. Duncan, P. M. Poggi, M. Bukov,et al., PRX Quan- tum6, 040201 (2025)

    2025

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.