{"id":"d32afade-94e1-4d10-9ce3-4b4a4d3f33a6","arxiv_id":"2411.18450","paper_version":2,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"An analytical decoupling-efficiency formula selects time-optimized AXY two-qubit gates, which are then used in a (2+1)-spin repetition code that corrects phase errors with simulated fidelity above 98.5%.","lead":"The paper derives a way to pick the best duration for two-qubit gates in diamond-based quantum computers, balancing speed against fidelity. It then shows that these gates can run a small error-correction code that protects a qubit against phase errors.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Equation (13) prints the wrong sign: substituting the paper's own r_z = -Δ(...) into Eq. (11) gives a minus sign, so the analytical gate-time selector is not correct as written and the optimal times used in the QEC simulation are not supported by the printed derivation.","rationale":"The central claim has two parts: analytical gate-time optimization via the decoupling-efficiency functions and the QEC demonstration. The analytical part is load-bearing, because the QEC simulation inherits whichever gate times the optimization selects. The weakest point is not the high-field approximation or the discarded internuclear coupling—those are stated approximations with supporting numerical evidence that they are mild for the time scales used. The weakest point is an internal mathematical inconsistency in the printed central formula. The paper's own definitions force a minus sign in Eq. (13), yet a plus sign is printed. This changes the predicted positions of the D_j extrema, which are exactly what Section III uses to choose gate times. The reader's verdict already flags the sign error in its rationale but lists a different weakest assumption; I am elevating the sign error because it is the cleanest decisive check and because the analytic selector is the paper's main methodological contribution. A single numerical trace calculation can settle it. If the corrected formula reproduces the same optimal N and the same QEC fidelity, then Eq. (13) is a typo and the conditional verdict can stand; if not, the central claim requires revision. I therefore keep the existing CONDITIONAL verdict rather than changing it.","tokens_in":12634,"tokens_out":9866,"duration_ms":92868,"concrete_test":"Recompute the single-spin decoupling efficiency directly from D_j(t) = (1/2) |tr(e^{iΔ t I_z} e^{-it(ĝ I_x - Δ I_z)})| for the two off-resonant spins used in Fig. 1D, fixing the target rotation φ = π/2. Scan N (or T) and compare the corrected Eq. (13) (minus sign) and the printed plus sign against the full-Hamiltonian simulation fidelity. Then regenerate the QEC curve in Fig. 3 using gate times selected from the corrected formula; if the optimal N values or the >98.5% fidelity point change, the central claim must be revised.","verdict_should_be":"UNCHANGED","load_bearing_attack":"With the stated definitions in Section III, r_z/|r| = -Δ_j / sqrt(ĝ_j^2 + Δ_j^2), so substituting into Eq. (11) with ϑ = Δ_j t/2 and μ = sqrt(ϑ^2 + (g_j φ / 2g_n)^2) gives D_j(ϑ) = |cos ϑ cos μ - (ϑ/μ) sin ϑ sin μ|. The paper prints a plus sign in Eq. (13). This is not a cosmetic typo: the extrema of |A+B| and |A-B| occur at different ϑ, and since ϑ = Δ_j T/2 maps directly to the gate time T, the predicted locally optimal sequence lengths N (and hence the fidelities in Fig. 3) shift when the sign is corrected. Eq. (12) makes the overall fidelity a product of these D_j, so the central claim that the decoupling-efficiency functions of Eqs. (12) and (13) identify locally optimal gate times is not supported by the printed formula. The internal inconsistency is between the displayed r_z = -Δ(...) and the displayed plus sign in Eq. (13). If the simulations used the correct sign, the fix is typographical, but that still needs verification.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript proposes an analytic method for selecting the execution time of two-qubit gates generated by adaptive XY (AXY) dynamical decoupling sequences in NV-center registers. The method is based on a decoupling-efficiency function D_j(t) derived from the high-field effective Hamiltonian; the gate fidelity factorizes into a product of these functions, and local minima of the resulting curve are identified with optimal gate times, including outside the usual rotating-wave-approximation validity. The optimized gates are then used in a simulated (2+1)-spin repetition code that corrects phase errors, with average fidelities above 98.5% for an error probability p=5% in the absence of electron T1 relaxation.","tokens_in":12947,"tokens_out":20136,"duration_ms":158600,"significance":"If correct, the paper offers a practical analytical shortcut for finding high-fidelity AXY gate times without extensive parameter scans, and it demonstrates a small-scale QEC protocol built from these gates. The authors validate the analytic decoupling-efficiency model against direct numerical simulation (Fig. 1D), include realistic microwave control errors (detuning and Rabi error) in the gate simulations, and test the QEC protocol under finite electron T1 relaxation. These are genuine strengths. However, the central analytic formula contains a sign error that needs to be corrected and verified.","major_comments":[{"comment":"Substituting the authors' own definition r_z = -Δ_j (g̃_j^2 + Δ_j^2)^{-1/2} into Eq. (11) gives D_j(ϑ) = |cosϑ cosμ − (ϑ/μ) sinϑ sinμ|, but Eq. (13) prints a plus sign between the two terms. Because ϑ = Δ_j T/2 is directly proportional to the gate time T, the extrema of the printed function occur at different values of T than those of the correct expression. Since Eq. (12) expresses the gate fidelity as a product of these D_j, the printed equation does not support the paper's central claim that Eqs. (12)-(13) identify locally optimal gate times. The sign must be corrected and the analytical curve in Fig. 1D and all optimal repetition numbers used in Fig. 3 re-verified with the corrected formula.","section":"Section III, Eq. (13)"},{"comment":"The QEC simulation uses \"speed-optimized gates presented in Section III\", so the fidelity numbers in Fig. 3 depend on the gate-time selection from Eq. (13). If the sign in Eq. (13) is corrected, the locally optimal sequence lengths N may shift. The authors should state explicitly whether the simulations used the printed (plus) formula or the algebraically consistent (minus) formula, and if the latter, the printed text should be amended accordingly. Without this verification, the connection between the analytic method and the reported QEC performance is not established as written.","section":"Section IV, Fig. 3"}],"minor_comments":[{"comment":"The quantity φ appearing in μ(ϑ) is not defined; it should be identified with the reference-gate rotation angle θ.","section":"Section III, Eq. (13)"},{"comment":"The rotating-frame operator is written as e^{−Σ Δ_j I_z^j}, which is missing the imaginary unit and the time variable; the text should read e^{−i Σ Δ_j t I_z^j}.","section":"Section III, definition of U_Δ"},{"comment":"The sentence \"and are we can treat each nuclear spin evolution individually\" contains a grammatical error; it should read \"and we can treat each nuclear spin evolution individually\".","section":"Section III, after Eq. (8)"},{"comment":"The abstract's claim of \"average fidelity above 98.5% for p=5%\" should specify that this holds in the decoherence-free limit (T1 = ∞), as stated in Section IV.","section":"Abstract"},{"comment":"There is a typo: \"woe became aware\" should be \"we became aware\".","section":"Acknowledgments"}],"recommendation":"major_revision","confidential_remarks":"The sign error is the central technical issue and must be resolved before publication. If the simulations used the correct sign, the fix is typographical, but the authors need to state this explicitly and update the text and figure captions. The paper is otherwise a solid contribution that fits the journal's scope."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First thing to know: this paper gives NV groups a fast analytical way to pick two-qubit AXY gate times by maximizing decoupling-efficiency functions, and it shows a small (2+1)-spin repetition code with those gates. The idea is sound and useful. But the printed centerpiece formula, Eq. (13), has a sign error: substituting their own rz = -Δ_j/|r| into Eq. (11) gives a minus sign in front of the (ϑ/μ) sinϑ sinμ term, not a plus. That is not cosmetic; the extrema of the sum and difference occur at different ϑ, which maps directly to gate duration, so the optimal sequence lengths cannot be reproduced from the paper as written. The numerical simulations in Fig. 1D likely used the correct physics, so this is probably a typo, but it needs to be fixed and verified before anyone trusts the analytical selector.\n\nWhat is actually new: the decoupling-efficiency derivation in Section III goes beyond the usual RWA bound and uses the periodic non-RWA oscillations to find locally optimal times. That is a real contribution, supported by simulation in Fig. 1D. The paper is honest about discarding H_nn analytically and including it in simulation, and it acknowledges Ref. [22] as a related idea. The repetition-code construction itself is a routine phase-flip code, so novelty is concentrated in the gate-time method.\n\nSoft spots: the sign error is the main one. The QEC part lacks a direct baseline comparison – there is no curve for what the achieved fidelity would be without the QEC protocol, so the 'above 98.5% for p=5%' claim is not contextualized. And calling this a 'demonstration' is generous; it is a numerical simulation. Minor: the notation switches from θ to φ for the rotation angle in Eq. (13), and the high-field approximation of Eq. (6) is assumed throughout. The internuclear coupling is argued away but not quantified in the analytical model.\n\nIf the sign fix checks out, this is a solid methods paper for experimental groups working on NV registers and AXY sequences. It deserves a serious referee – I would send it to review, not desk reject it – but the authors should be asked to correct Eq. (13), add a no-QEC baseline, and soften 'demonstrated'.","headline":"Useful analytical gate-time selection for AXY two-qubit gates, but a sign error in Eq. (13) makes the central formula wrong as printed; fixable, and worth refereeing.","tokens_in":13449,"tokens_out":3806,"would_cite":false,"duration_ms":31463,"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":"The paper claims that gate fidelity in NV-diamond registers factorises into oscillatory decoupling-efficiency functions, and that their local maxima select gate times enabling a compact phase-error code with 98.5% average fidelity.","keywords":["nitrogen-vacancy centers","adaptive XY sequences","dynamical decoupling","two-qubit gates","quantum error correction","repetition code","gate time optimization","decoherence efficiency functions"],"falsifier":"Measure the fidelity of an AXY-8 $A^x_1(\\pi/2)$ gate on a single NV center with two resolved $^{13}$C spins at $B=600$ G while sweeping the repetition number $N$ across the values where the decoupling-efficiency formula predicts local infidelity minima, including values below the RWA bound; if the measured minima do not align with the predicted times within the combined control-error budget, the selection rule is falsified.","tokens_in":12448,"feed_emoji":"💎","tokens_out":12736,"duration_ms":107173,"temperature":0.7,"pith_summary":"The paper tries to show that the speed–fidelity trade-off for selective two-qubit gates between an NV-center electron spin and a nearby 13C nuclear spin can be resolved by a formula instead of a parameter scan. It derives a decoupling-efficiency function for each off-resonant nuclear spin whose oscillations mark gate times at which the perturbing terms discarded by the rotating-wave approximation (RWA) cancel, and it verifies that these times give high-fidelity gates even outside the parameter range the RWA would normally allow. Using these gates, the paper builds a (2+1)-spin repetition code in which the two carbon nuclei form the code space, and its simulations show that the protocol recovers a stored qubit with average fidelity above 98.5% when each physical qubit undergoes a phase error with probability $p=5\\%$. The practical point is that an analytical curve, not gate tomography, can pick the best working point for each gate.","feed_headline":"Diamond spin register corrects phase errors at 98.5%","feed_subtitle":"Adaptive XY pulses pick optimal gate times by formula, then run a compact error-correction code.","key_machinery":"The load-bearing object is the decoupling-efficiency function $D_j(t)$, a closed-form expression for how much an AXY sequence disturbs an off-resonant nuclear spin while a gate is performed on the target spin. It emerges from writing the evolution in the electron-spin $\\sigma_z$ eigenbasis, where each nuclear spin sees an effective constant Hamiltonian $\\tilde g_j I^x_j - \\Delta_j I^z_j$ in a rotating frame; the trace of that local evolution yields $D_j$. The product formula $F=\\prod_{j\\neq n}D_j(T)$ converts the full gate fidelity into a product of single-spin factors, and the adaptive character of the sequence enters because keeping the rotation angle fixed forces the Fourier coefficient to scale as $t^{-1}$, leaving each factor dependent on time only through $\\vartheta=\\Delta_j t/2$. The oscillations of $D_j$ are the actual mechanism: periodic non-RWA perturbations cancel at selected times, which locates infidelity minima without any experimental parameter scan.","core_discovery":"On the paper's own terms, the central claim is that the fidelity of an AXY-generated two-qubit gate acting on a target nuclear spin $n$ equals the product, over every other nuclear spin $j$, of the decoupling-efficiency function $D_j(T)$, defined as the fidelity of the local evolution on spin $j$ with respect to the identity. In the high-field regime, and with the rotation angle kept fixed as the adaptive Fourier coefficient is re-scaled with time, this function becomes $$D_j(\\vartheta)=\\left|\\cos\\vartheta\\,\\cos\\mu(\\vartheta)+\\frac{\\vartheta}{\\mu(\\vartheta)}\\sin\\vartheta\\,\\sin\\mu(\\vartheta)\\right|,\\qquad \\mu(\\vartheta)=\\left(\\$vartheta^{2}$+\\left(\\frac{g_j\\phi}{2g_n}\\right)^2\\right)^{1/2},$$ with $\\vartheta=\\Delta_j t/2$. Because $D_j$ oscillates, its local maxima are gate times of locally minimal infidelity; these occur where the periodic non-RWA perturbation cancels, which is why the best times can lie outside the usual RWA bound (5). The paper further claims that the same gates, encoded with $U_{\\rm enc}=A^x_2(\\pi/2)A^x_1(\\pi/2)$ and read out through iSWAP operations, form a phase-error repetition code whose simulated average correction fidelity stays above 98.5% at $p=5\\%$, and that electron-spin longitudinal relaxation with $T_1\\approx1$ s at temperatures up to 77 K does not significantly degrade that performance.","pith_inferences":["Going beyond the paper, the factorization $F=\\prod_j D_j$ implies that searching for a good gate time in a register of $d$ nuclear spins remains a one-dimensional problem in the sequence length: one only needs a time where the product of $d-1$ oscillatory functions is maximal, so the method may scale to larger carbon registers without expensive master-equation simulations.","The same periodic-perturbation-cancels-at-special-times mechanism is not specific to AXY sequences; any periodic pulse sequence with a tunable Fourier coefficient should exhibit analogous local minima, so the analytical time-selection recipe could be exported to other pulse families or other spin platforms.","Because the simulation assumes ideal noise-free syndrome measurements, a natural next test is to add readout noise; I would expect the 98.5% recovery fidelity to decrease roughly by the readout error rate plus any extra idling time introduced by the measurement step.","The paper's rough counting estimate puts the chance of finding two carbon spins with the required coupling strengths at around 39% (neglecting the geometric factor), which I read as a warning that a practical implementation would need to characterize several candidate NV centers before finding a suitable one."],"forward_implications":["Gate times for a desired fidelity can be selected by evaluating a closed-form product of oscillatory functions instead of performing gate tomography, which the paper argues becomes indispensable as the register grows beyond a few nuclear spins.","Locally optimal times exist outside the RWA coupling-limit condition (5), so the usable parameter region for high-fidelity gates is broader than the approximation condition alone would suggest.","The repetition-code protocol requires only $A^x_j$ gates for encoding and uses iSWAP operations for readout, where each iSWAP is decomposed into local electron rotations plus $A^x_j$/$A^y_j$ gates, the same AXY-generated two-body gate family that is optimized.","With the optimized gates, the (2+1)-spin code corrects phase errors with average fidelity above 98.5% for $p=5\\%$, and adding electron-spin relaxation with $T_1=1$ s at 77 K changes the result negligibly.","Gates with infidelity below $10^{-2}$ run at about $2.2\\,T_{\\rm min}$, and gates with infidelity below $10^{-3}$ at about $4.4\\,T_{\\rm min}$, where $T_{\\rm min}$ is the shortest theoretically possible two-spin gate time without decoupling."],"supporting_citations":[{"why":"Supplies the adaptive XY sequence construction and the analytical pulse-position solution used to set Fourier coefficients.","marker":"[8]"},{"why":"Demonstrates that adaptive sequences can act as high-fidelity gates, the starting point for the gates optimized here.","marker":"[9]"},{"why":"Supplies the repetition-code structure from which the (2+1)-spin phase-error protocol is derived.","marker":"[10]"},{"why":"Provides the experimentally measured hyperfine couplings used as simulation parameters.","marker":"[14]"},{"why":"Gives the zero-field-splitting temperature dependence used to set the microwave detuning error in the simulations.","marker":"[15]"},{"why":"Defines the Haar-averaged state fidelity used to score the correction protocol.","marker":"[17]"},{"why":"Supplies the measured relaxation rates and temperatures used for the dissipative electron-spin simulations.","marker":"[18–20]"},{"why":"Provides the soft-control alternative against which the fixed-coefficient AXY approach is compared in the outlook.","marker":"[21]"}],"fun_headline_variants":["Adaptive gates push diamond spin code past 98.5%","NV register uses adaptive XY pulses to cut gate errors","Optimal gate timing lifts diamond error correction to 98.5%","Two-body gates in diamond hit 98.5% correction fidelity","Pulse-timing formula sharpens NV error correction"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The selection rule assumes the gate's error is dominated by the off-resonant nuclear spins described by the decoupling-efficiency functions, with the target spin exactly on resonance, the magnetic field high enough to suppress higher-harmonic terms, and internuclear couplings too weak to matter; if control errors, higher harmonics, or nuclear–nuclear couplings shift the infidelity minima, the predicted optimal times will be off.","fun_headline_variants_meta":{"raw":{"variants":["Adaptive gates push diamond spin code past 98.5%","NV register uses adaptive XY pulses to cut gate errors","Optimal gate timing lifts diamond error correction to 98.5%","Two-body gates in diamond hit 98.5% correction fidelity","Pulse-timing formula sharpens NV error correction"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000231,"raw_usage":{"total_tokens":1509,"prompt_tokens":990,"completion_tokens":519,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":606,"completion_tokens_details":{"reasoning_tokens":433}},"tokens_in":606,"tokens_out":519,"duration_ms":4571,"temperature":1.0,"reasoning_tokens":433,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T11:12:24.131552+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the fidelity of an AXY-8 $A^x_1(\\pi/2)$ gate on a single NV center with two resolved $^{13}$C spins at $B=600$ G while sweeping the repetition number $N$ across the values where the decoupling-efficiency formula predicts local infidelity minima, including values below the RWA bound; if the measured minima do not align with the predicted times within the combined control-error budget, the selection rule is falsified.","supporting_citations":[{"cited_title":"Casanova, Z.-Y","cited_arxiv_id":null,"evidence_quote":"Supplies the adaptive XY sequence construction and the analytical pulse-position solution used to set Fourier coefficients."},{"cited_title":"Casanova, Z.-Y","cited_arxiv_id":null,"evidence_quote":"Demonstrates that adaptive sequences can act as high-fidelity gates, the starting point for the gates optimized here."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the experimentally measured hyperfine couplings used as simulation parameters."},{"cited_title":"Assuming a constant detun- 4 ing represents a worst case scenario, since we approxi- mate the fluctuating process by its maximum|∆(t)| ≤∆","cited_arxiv_id":null,"evidence_quote":"Gives the zero-field-splitting temperature dependence used to set the microwave detuning error in the simulations."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the Haar-averaged state fidelity used to score the correction protocol."},{"cited_title":"Bar-Gill, L","cited_arxiv_id":null,"evidence_quote":"Provides the soft-control alternative against which the fixed-coefficient AXY approach is compared in the outlook."}],"review_version":1}