{"id":"1dbcc27e-7877-4692-b5ad-7ba82ec6e9b9","arxiv_id":"2512.08393","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"A single phase-engineered reset segment appended to a square readout pulse empties the circuit-QED cavity ~6× faster than free decay with no extra hardware and no extra backaction.","lead":"Researchers appended a short reset pulse with a specially chosen amplitude and phase to the readout pulse in a superconducting quantum processor, emptying the measurement cavity about six times faster than natural decay. The method adds no hardware and may shorten the idle time between measurements in quantum error correction.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Unsupported 'arbitrarily fast' reset claim: only Δτ=50 ns is tested, and the linear-model reset amplitude diverges as 1/Δτ, so nonlinearity/backaction may invalidate the extrapolation.","rationale":"The reader's weakest assumption already flags the linear input-output model and the untested 'arbitrarily fast' extrapolation. My stress-test sharpens this into a concrete mechanism: the reset amplitude required by Eq. (5) diverges at short Δτ, which is exactly the regime where the weakly nonlinear Kerr term and drive-induced qubit transitions become relevant. This does not overturn the measured factor-six reset at Δτ=50 ns, but it means the headline 'arbitrarily fast' claim should carry a clear qualifier or be backed by multi-Δτ data. Since the reader's verdict is already CONDITIONAL, my read does not change the verdict. I agree partially because the reader identified the general area (model validity and parameter range) but not the specific 1/Δτ divergence of the reset drive and its implication for backaction.","tokens_in":13460,"tokens_out":23901,"duration_ms":227612,"concrete_test":"Keep the readout amplitude fixed at βn=1 and measure reset performance at Δτ = 100, 50, 25, and 12.5 ns. For each Δτ, numerically optimize (εr, ϕr) using Eq. (B.1) with the independently measured Kc, then on the same device measure residual photon number with the Ramsey method and qubit excitation/relaxation rates with the sequence of Fig. 4. If residual photons remain below 0.1 and the excitation/relaxation rates are flat as Δτ decreases, the 'arbitrarily fast' claim is supported; if either degrades, the conclusion should be restricted to the demonstrated duration.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The Abstract and Conclusion claim 'arbitrarily fast, overshoot-free depletion' of the cavity, but the experiment fixes the reset duration at Δτ = 50 ns and only demonstrates that one duration. In the linear model, Eq. (5) implies the reset amplitude needed to satisfy α_j(Δτ)=0 grows as εr ∼ εn/Δτ as Δτ→0, because the denominator 1−exp(−C_j Δτ/2) is proportional to Δτ for small Δτ. For example, reducing Δτ from 50 ns to 10 ns would require roughly a fivefold increase in reset amplitude. At such amplitudes, the Kerr nonlinearity calibrated in Appendix C (Kc/2π = −11 kHz) and the drive-induced qubit transitions discussed in Section III are no longer guaranteed to be negligible; the paper supplies no data or bound showing residual photon number remains below the 0.1-photon contour of Fig. 1 or that excitation/relaxation rates do not rise when Δτ is shortened. The experimental evidence supports a factor-six speed-up at the single demonstrated reset duration, but not the unrestricted 'arbitrarily fast' claim.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports an experimental implementation of a single-step phase-engineered pulse (SSPE) for resetting a superconducting readout cavity. A reset segment with optimized amplitude and phase is appended to a square readout pulse, and the coherent field amplitude is returned to zero at the end of the segment according to the linear input-output model. The authors claim the reset amplitude scales linearly with readout amplitude while the phase is invariant, simplifying calibration. They demonstrate residual photon numbers near zero at a reset duration of 50 ns, with effective decay rates approximately six times faster than passive free decay on qubit Q1, and they characterize measurement-induced backaction, reporting lower excitation and relaxation rates than Square or CLEAR pulses. The method is also tested on a second qubit in an appendix. The central claims are that SSPE enables arbitrarily fast, overshoot-free depletion and that it introduces no additional non-QND errors beyond those of free decay.","tokens_in":13820,"tokens_out":2171,"duration_ms":22808,"significance":"If the claims hold, the SSPE scheme is a simple, hardware-efficient pulse-level approach to cavity reset that could be valuable for mid-circuit measurement and fast feedback in circuit QED. The linear-scaling calibration rule (amplitude proportional, phase invariant) is a useful practical simplification, and the comparison with CLEAR addresses a known transient overshoot issue. The work combines standard input-output theory, Ramsey/ac-Stark photon-number measurement, and backaction characterization. The manuscript includes a two-qubit test and a nonlinear (Kerr) model fit, which are positive features. However, the strongest claim—'arbitrarily fast' reset—is an extrapolation from a single reset duration, and the quantitative comparisons lack error bars, so the significance of the reported speedups and backaction advantages is not yet fully established.","major_comments":[{"comment":"The abstract and conclusion claim 'arbitrarily fast, overshoot-free depletion,' but the experiment only tests a single reset duration Δτ = 50 ns. In Eq. (5), the reset amplitude required to satisfy |α_j(Δτ)|^2 = 0 grows as ε_r ∝ 1/Δτ for Δτ → 0, since the denominator 1 − exp(−C_j Δτ/2) is linear in Δτ. At shorter Δτ, the larger drive amplitude will enter the nonlinear regime where Eq. (B.1) with Kc/2π = −11 kHz applies, and drive-induced qubit transitions (discussed in Section III) may become non-negligible. No data or bound is provided showing that the residual photon number remains below the 0.1-photon contour or that backaction rates stay low when Δτ is reduced. Please either soften the 'arbitrarily fast' claim to what is demonstrated, or provide measurements at additional Δτ values (e.g., 25 ns and 10 ns) with the nonlinear model and backaction characterization.","section":"Section II, Eq. (5); Section III, Fig. 3"},{"comment":"The central quantitative comparisons—residual photon population, decay rates (κ_SSPE vs κ_free), and backaction probabilities—are presented without error bars or statistical uncertainty. For example, Fig. 3(d,e) reports κ_SSPE/2π = 10.968 and 11.618 MHz with no confidence intervals, and Fig. 4(b,c) reports excitation and relaxation rates (0.05%, 7.22%, 9.54%) without uncertainty. The claim that SSPE yields the lowest backaction among the three protocols depends on the significance of these differences. Please add error bars, repeat counts, or confidence intervals for all quantitative claims, especially the factor-of-six speedup and the ordering of backaction rates.","section":"Section III, Figs. 2–4"},{"comment":"The decay rates for SSPE are obtained by exponential fits, but the SSPE reset process is a driven coherent trajectory, not an exponential decay. It is not clear over which time window the fit is performed and whether the fitted rate is a well-defined physical quantity or merely an empirical descriptor. Please specify the fitting range, the functional form used for the driven segment, and the goodness-of-fit. Without this, the comparison 'six times faster' is ambiguous.","section":"Section III, Fig. 3(d,e)"}],"minor_comments":[{"comment":"Typo: 'Thses' should be 'These'.","section":"Abstract"},{"comment":"The text says 'residual photon population ... measured either at a delay of Δτ = 50 ns following the end of a Square pulse or immediately after the SSPE pulse.' It would be clearer to specify whether the SSPE residual is measured exactly at t = Δτ or at a later fixed delay; the caption should also state the number of experimental repetitions used for each point.","section":"Section III, Fig. 2"},{"comment":"The term 'π-pusle' appears twice; should be 'π-pulse'.","section":"Appendix B, Fig. 5"},{"comment":"The Kerr coefficient Kc/2π = −11 kHz is reported without uncertainty. A confidence interval and the fit range would help assess the validity of the nonlinear model at the drive amplitudes used in the reset segment.","section":"Appendix C"},{"comment":"The manuscript would benefit from a table summarizing the measured residual photon numbers, decay rates, and backaction rates for Q1 and Q2, with uncertainties, to make the cross-qubit comparison easier.","section":"General"}],"recommendation":"major_revision","confidential_remarks":"The core idea is sensible and the experimental demonstration at Δτ = 50 ns appears internally consistent. The main concern is over-claiming in the abstract/conclusion: 'arbitrarily fast' is a theoretical extrapolation that the current data do not support, and the missing error bars make it hard to judge the significance of the speedup and backaction comparisons. The paper is likely publishable after these points are addressed, but I would not accept it in present form."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a legitimate, workmanlike demonstration of a one-segment active cavity reset pulse. The measured ~6x speedup over free decay at 50 ns is supported by a standard input-output model and consistent Ramsey and ac-Stark data. The scaling law—reset amplitude proportional to readout amplitude, phase invariant—is clearly derived and confirmed in Fig. 2(b). That alone makes the paper worth having in the literature.\n\nWhat's genuinely new: SSPE is a single-segment variant of CLEAR, with a simpler calibration rule. The comparison against CLEAR and Square in Figs. 3–4 is fair and shows SSPE avoids the overshoot that CLEAR produces and has the lowest excitation/relaxation rates. The reproducibility on a second qubit (Q2 in Appendix B) is a nice touch. The modeling is honest: they include a Kerr term in Appendix B to fit the nonlinear regime, and they explicitly note that the reset pulse is limited by available drive amplitude.\n\nSoft spots, in proportion. The biggest is the 'arbitrarily fast, overshoot-free depletion' claim in the abstract and conclusion. The experiment only tests Δτ = 50 ns. The linear solution (Eq. 5) shows the required reset amplitude grows roughly as 1/Δτ, so pushing to 10 ns would need ~5x more power—where the Kerr nonlinearity and drive-induced qubit transitions they themselves calibrate are no longer negligible. There is no data showing the residual photon number stays below the 0.1 contour or that backaction stays flat at shorter times. The stress-test note is correct. The authors do hedge in Section II ('in practice limited by maximum drive amplitude'), but the abstract does not carry that caveat. That needs fixing.\n\nSecond, error bars are absent from Figs. 2–4, and the decay rate fits (κ_SSPE ~ 11 MHz) are quoted without uncertainties. Given the Ramsey method, the zero-residual claim at βn up to 1.5 is believable, but I can't tell if the floor is 0.01 or 0.05 photons. The data are not public, only available on request.\n\nThis is not a fatal flaw. The central 50 ns result, the scaling law, and the backaction comparison are all reproducible in principle and mutually consistent. The paper deserves a serious referee. I would send it out with a request to temper the abstract, add error bars, and include a discussion—ideally one measurement—of what happens at shorter Δτ.","headline":"Solid experimental reset result with a real scaling law, but the 'arbitrarily fast' claim outruns the data.","tokens_in":14252,"tokens_out":2489,"would_cite":true,"duration_ms":22763,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Adding one phase-tailored microwave segment to a standard readout pulse empties the superconducting readout cavity in about 50 ns, roughly six times faster than passive decay, without extra qubit backaction.","keywords":["cavity reset","circuit quantum electrodynamics","dispersive readout","input-output theory","measurement-induced backaction","superconducting qubit","coherent state cancellation","pulse shaping"],"falsifier":"Measure the residual photon number immediately after the SSPE pulse for a reset duration of, say, 20 ns (shorter than the demonstrated 50 ns) at the same readout amplitude; the model predicts the optimized (εr, φr) still cancels the field exactly. If any nonzero residual appears, the 'arbitrarily fast' assertion and the linear-cancellation picture are wrong.","tokens_in":13421,"feed_emoji":"⚛️","tokens_out":6705,"duration_ms":56914,"temperature":0.7,"pith_summary":"This paper reports an experimental method for actively resetting the readout cavity of a superconducting qubit after a measurement. Rather than waiting for the cavity to decay passively, the authors append a single microwave segment with a carefully chosen amplitude and phase to the end of the standard square readout pulse. This segment cancels the coherent field left in the cavity, driving the photon number to zero within 50 ns for both qubit states |0> and |1>. The reset is smooth (no photon overshoot) and introduces no additional measurement-induced backaction relative to free decay, which makes fast, low-backaction cavity reset possible without any hardware overhead.","feed_headline":"One reset pulse empties qubit readout cavities 6x faster","feed_subtitle":"Phase-tailored microwave segment returns the readout cavity to vacuum in 50 ns, without extra qubit backaction.","key_machinery":"The mechanism is a two-segment pulse: a standard square readout segment of duration τ, followed by a reset segment of duration Δτ with complex amplitude εr e^{iφr}. The cavity field is governed by the input–output equation dα_j/dt = -i ε_d(t) - i(Δ_r + χ_j)α_j - (κ/2)α_j, whose piecewise-constant solution yields an analytic expression for α_j(t) during both segments. The reset parameters are chosen so that |α_j(Δτ)|^2 = 0 simultaneously for j=0 and j=1. The practical key is the scaling law: optimal ε_r is proportional to the readout amplitude, while optimal φ_r is invariant, so calibration is a one-time measurement. For weakly nonlinear operation, the model is extended with a Kerr term K_c|α","core_discovery":"The central claim is that, in the linear-response regime, for any fixed readout amplitude and duration there exists a single pair of reset parameters—amplitude εr and phase φr—such that the cavity field returns exactly to the vacuum at the end of the reset interval for both qubit states. The experiment verifies this by measuring the residual photon number after the pulse and by time-resolved photon dynamics. When the readout amplitude is rescaled by a factor βn, the optimal reset amplitude rescales proportionally while the optimal reset phase remains fixed; this scaling law reduces calibration to one measurement per device. Fitted reset rates are about 10.97 MHz (|0>) and 11.62 MHz (|1>), ro","pith_inferences":["The same scaling law suggests that qubit-state-independent reset could be achieved by selecting a phase that cancels the field for both |0> and |1>; the paper leaves this as future work, but the fixed-phase property makes it a natural next test.","The 'arbitrarily fast' claim rests on the linear model; a short-window test (e.g., Δτ = 20–30 ns) would verify whether optimized (εr, φr) still cancels the field and would locate the real bandwidth limit.","The Kerr-corrected model should predict a maximum usable reset amplitude before nonlinearity spoils cancellation; sweeping εr at fixed phase and comparing residual photons would test this directly, and might even provide a way to measure K_c more precisely."],"forward_implications":["SSPE empties the readout cavity to near-zero photon population in 50 ns for both qubit states across a wide range of readout amplitudes, while a square pulse leaves a residual population that grows with drive amplitude.","Because the reset trajectory avoids photon overshoot, the protocol does not experience the transient backaction seen in CLEAR; measured excitation and relaxation rates are the lowest of the three protocols.","The amplitude-linear, phase-invariant scaling law means that after a one-time phase calibration, reset parameters for any readout amplitude are obtained by proportional scaling, greatly simplifying use in multiplexed systems.","The ultimate reset speed is limited by available drive amplitude and analog bandwidth, not by the model; the paper states that arbitrarily fast, overshoot-free depletion is achievable in principle."],"fun_headline_variants":["Single pulse resets readout cavity to vacuum in 50 ns","Phase-tailored pulse empties cavity 6x faster than CLEAR","One-shot reset: cavity vacuum in 50 ns, no extra backaction","Single-step pulse resets readout cavity with zero overshoot","Cavity reset via phase-engineered pulse: 6x faster, no QND penalty"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The protocol assumes that a single (εr, φr) pair optimized in the linear input–output model (with constant decay rate and dispersive shift) continues to cancel the cavity field at the experimentally used drive amplitudes and within the 50 ns reset window; if nonlinearities or state-dependent effects shift the field away from this trajectory, residual photons remain.","fun_headline_variants_meta":{"raw":{"variants":["Single pulse resets readout cavity to vacuum in 50 ns","Phase-tailored pulse empties cavity 6x faster than CLEAR","One-shot reset: cavity vacuum in 50 ns, no extra backaction","Single-step pulse resets readout cavity with zero overshoot","Cavity reset via phase-engineered pulse: 6x faster, no QND penalty"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00024,"raw_usage":{"total_tokens":1366,"prompt_tokens":766,"completion_tokens":600,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":510,"completion_tokens_details":{"reasoning_tokens":501}},"tokens_in":510,"tokens_out":600,"duration_ms":5284,"temperature":1.0,"reasoning_tokens":501,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-03T17:39:44.861425+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the residual photon number immediately after the SSPE pulse for a reset duration of, say, 20 ns (shorter than the demonstrated 50 ns) at the same readout amplitude; the model predicts the optimized (εr, φr) still cancels the field exactly. If any nonzero residual appears, the 'arbitrarily fast' assertion and the linear-cancellation picture are wrong.","supporting_citations":[],"review_version":1}