{"id":"305d45cd-9c4c-4aa4-a70f-e158d4b65bbb","arxiv_id":"2608.06119","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Hadamard sensing channel deterministically suppresses finite-pulse artifacts in dynamical-decoupling quantum sensors, matching phase randomization's suppression factor with zero statistical variance.","lead":"A new protocol, called the Hadamard sensing channel, uses patterns of discrete pulse phases to cancel spurious signals in quantum sensors without the statistical averaging required by phase randomization. It could make nanoscale nuclear magnetic resonance measurements more reliable with simpler hardware.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Proof drops the run-dependent cos^2 phase factor from Eq. (10); exact M-fold suppression of the measured spurious intensity is not established, though a bound at 1/M survives.","rationale":"The reader's weakest assumption identifies the same gap: the proof of Eq. (12) averages only |H^{(n)}_M|^2, while Eq. (10) explicitly includes a cos^2 phase factor whose phase depends on n through H^{(n)}_M. This is load-bearing because the paper's headline claims are 'exactly and deterministically canceling spurious signals' and a suppression factor of M that is 'independent of Fourier index k.' Those claims are established only for the squared Fourier amplitude, not for the measured single-quadrature intensity that the simulations report. The concern is concrete: for M=4, the HSC-averaged intensity including cos^2 can be 1/8 or 1/4 of |\\tilde f^⊥|^2 depending on k and signal phase, so the exact factor M and k-independence fail at the level of the observable. This does not undermine the core proposal: HSC remains deterministic, has zero phase-randomization variance, and its residual spurious intensity is bounded by (1/M)|\\tilde f^⊥|^2 because cos^2 ≤ 1. The weak-coupling restriction is acknowledged in the paper, and the Hadamard orthogonality algebra in Eq. (12) is correct. The simulations are suggestive but do not settle the phase dependence because the signal phase is fixed in a given plot. A conditional verdict is therefore appropriate: the method is promising and likely useful, but the stated exactness needs a corrected derivation or a clear definition of which quantity is suppressed by exactly M.","tokens_in":9342,"tokens_out":21413,"duration_ms":180861,"concrete_test":"For a fixed signal phase θ and each relevant k, compute R_k(θ) = (1/M) Σ_n |H^{(n)}_M|^2 cos^2(arg H^{(n)}_M + θ) using the explicit Hadamard phase patterns of the paper (e.g., M=64), and compare with 1/M. Independently, re-run the Fig. 2 simulation with the initial phase of the AC field varied over [0, 2π) and measure the residual spurious peak height under HSC. If R_k(θ) and the simulated peak heights are constant and equal to the |f^⊥|^2 prediction, the concern is resolved; if they vary with k or θ, the exact suppression claim must be revised to a bound.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Eq. (10) states the spurious intensity is proportional to sin^2((1/2) A_perp |f^⊥_{n,k}| M T) cos^2(φ^⊥_k), with φ^⊥_k described as the phase of f^⊥_k. The proof then drops this factor and averages only |f^⊥_{n,k}|^2, concluding Eq. (12). Since f^⊥_{n,k}=H^{(n)}_M \\tilde f^⊥_{k/M}, the phase is φ^⊥_k = arg H^{(n)}_M + arg \\tilde f^⊥_{k/M}, which generally depends on the run index n. The actually averaged intensity is |\\tilde f^⊥|^2 (1/M) Σ_n |H^{(n)}_M|^2 cos^2(arg H^{(n)}_M + θ), which is not equal to (1/M)|\\tilde f^⊥|^2 for all k and signal phases. For Sylvester H4, this gives 1/8 for k=1 and up to 1/4 for k=2 when θ=0. Thus the exact, k-independent 1/M suppression is proven only for the squared Fourier amplitude, not for the measured single-quadrature response. The deterministic/variance claim survives, but the 'exact' factor-M claim needs qualification or an additional argument that cos^2 is run-independent.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript introduces the Hadamard sensing channel (HSC), a deterministic phase-cycling protocol for dynamical-decoupling quantum sensors. HSC assigns the rows of a Hadamard matrix as global phase patterns (0 or π) to M pulse units across M experimental runs, then averages the resulting signals. The paper derives that the average squared spurious Fourier amplitude is suppressed by a factor M relative to an unmodulated sequence, claims that this matches ideal phase randomization without statistical variance, and supports the claim with simulations of single-spin and dual-nuclear-spin NV sensing scenarios under increasing control errors.","tokens_in":9626,"tokens_out":14867,"duration_ms":110162,"significance":"If the suppression claim is established for the measured intensity, HSC is a useful and hardware-friendly alternative to phase randomization: it requires only 0/π phases, eliminates the statistical variance of random phase averaging, and the central Fourier-orthogonality argument is explicit and parameter-free. The simulations cover realistic multi-nuclear scenarios and demonstrate qualitative robustness. However, the exactness of the suppression factor for the measured intensity, and the quantitative comparison with PR, need qualification before the claims as stated are justified.","major_comments":[{"comment":"The derivation of the suppression factor drops the cos^2(φ^⊥_k) factor in the spurious intensity. In Eq. (10), the measured intensity is proportional to (1/2 A⊥ |f^⊥_{n,k}| M T)^2 cos^2(φ^⊥_k), but the proof immediately replaces this by |f^⊥_{n,k}|^2 and averages only |H_M^{(n)}|^2 via Eq. (12). Since φ^⊥_k = arg H_M^{(n)} + arg \\tilde f_{k/M} is generally run-dependent, the averaged intensity is not (1/M)|\\tilde f_{k/M}|^2 for all k. For example, with Sylvester H_4 and k=2, taking the signal phase θ = arg \\tilde f = 0 gives an averaged intensity (1/4)|\\tilde f|^2, while k=1 gives (1/8)|\\tilde f|^2; Eq. (12) alone predicts a uniform 1/M. The exact factor-M claim should therefore be restricted to the squared Fourier amplitude, or an additional argument must show why the cos^2 factor is either run-independent or harmless for every measured quadrature.","section":"III.B, Eqs. (10)-(12)"},{"comment":"The statement that HSC 'matches the ideal suppression of PR' is not established for the measured intensity defined by Eq. (10). For PR, the random phases make φ^⊥_k random, so the cos^2 factor averages to 1/2; for HSC, for the standard spurious peaks at k multiples of M, H_M^{(n)} is real, so cos^2 = 1 in every contributing run. Thus the HSC intensity suppression factor is 1/M for those peaks, a factor of two weaker than PR's 1/(2M), while the manuscript reports only the 1/M factor for the squared amplitude. The comparison with PR needs to be restated in terms of the same quantity, and the factor of two should be acknowledged or explained.","section":"Abstract, Section V, and context around Eq. (12)"}],"minor_comments":[{"comment":"The text twice states that HSC requires only four distinct control phases (0, π/2, π, 3π/2), but the Hadamard construction in Eq. (2) and Section III.A uses only 0 and π; please reconcile this inconsistency.","section":"I and V"},{"comment":"The phase φ^⊥_k should carry a run index n, because it is the phase of f^⊥_{n,k}; the current notation obscures the run dependence that is central to the proof.","section":"III.B, Eq. (10)"},{"comment":"The word 'expolit' should be 'exploit'.","section":"III.A"},{"comment":"The product notation '1Y_{m=M}' is nonstandard and should be written as \\prod_{m=1}^M (or with the intended ordering) to be unambiguous.","section":"III, Eq. (3)"},{"comment":"The phrase 'high-resolution A WGs' contains a spacing typo; it should be 'AWGs'.","section":"I"}],"recommendation":"major_revision","confidential_remarks":"The cos^2 issue and the factor-of-two comparison with PR are substantive but fixable; they should be resolved before publication. I would not reject on these grounds alone. The manuscript should also clarify whether the claimed suppression factor applies to the measured intensity or to the squared Fourier amplitude, since the abstract and conclusions currently use the stronger phrasing."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know about arXiv:2608.06119. First, the Hadamard sensing channel is a real, useful idea: the authors take the orthogonal phase-cycling trick from their earlier HPC work and apply it to the spurious-response problem in DD-based NV sensing, proving it works for any Hadamard matrix rather than just the Sylvester construction. Second, the central suppression claim is sound, but it is over-stated as \"exact\" in one place that a referee will want tightened.\n\nThe core calculation, Eq. (12), is correct. Averaging |H^{(n)}_M|^2 over the M rows of a Hadamard matrix gives exactly 1/M by row orthogonality, so the squared spurious Fourier amplitude is deterministically suppressed by a factor M, matching the mean suppression of PR with zero statistical variance. The simulations are believable and support the qualitative claims: HSC cleans up the 4x13C artifact that sits on top of the 1H resonance and degrades more gracefully than PR under injected pulse errors. The generalization to any Hadamard matrix is a genuine extension over the Sylvester-only HPC framework, and the derivation is self-contained with no fitted parameters.\n\nThe soft spot is the step from Eq. (10) to the next line. The spurious intensity for a single run carries a cos^2(phi) phase factor, and phi = arg(f_perp_{n,k}) = arg(H^{(n)}_M) + arg(tilde f) depends on the run index n. The proof drops this factor without comment, so \"the measured intensity is exactly 1/M of the unmodulated value\" is not what is proven. What is proven is the averaged squared amplitude, which bounds the averaged intensity from above: since cos^2 <= 1, the measured intensity is suppressed by at least a factor M, and for Sylvester matrices and generic k it actually comes out closer to 1/(2M). The deterministic hardware-friendly pitch survives; only the word \"exactly\" needs qualifying. The authors should compute the cos^2-weighted average or state the bound explicitly. Minor: H^{(n)}_M in Eq. (8) silently drops the k index.\n\nReproducibility is the other gap: the simulations carry the robustness claims, but no code or data are released. For a methods paper that should be fixable in revision.\n\nThis paper is for experimentalists running multi-pulse NV sensing who currently rely on PR and want deterministic suppression without AWG-grade hardware. It is an incremental but practical contribution and deserves a serious referee. My recommendation: send it out, with the referee told to push on the exactness claim and to ask for the simulation code.","headline":"A sound, incremental protocol that deterministically matches PR's suppression of spurious responses in DD sensing, but the 'exact' factor-M claim outruns the proof, which drops a run-dependent phase factor; the 1/M bound survives and the paper deserves review.","tokens_in":10117,"tokens_out":13554,"would_cite":false,"duration_ms":95853,"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":"Hadamard phase patterns exactly cancel every spurious response in dynamical-decoupling quantum sensors, matching phase randomization's suppression without its statistical variance.","keywords":["Hadamard sensing channel","quantum sensing","dynamical decoupling","phase randomization","spurious responses","nanoscale NMR","control errors","pulse-sequence phase cycling"],"falsifier":"Compute the full run-averaged spurious intensity, keeping the $\\cos^2(\\varphi_k)$ factor, for a finite-duration pulse sequence under 10–20% pulse amplitude and detuning errors, and check whether the residual scales as $1/M$ with zero run-to-run variance; equivalently, an experiment that measures the shot-to-shot variance of the sensor spectrum across $M$ HSC runs would settle whether the deterministic cancellation holds beyond the ideal weak-coupling model.","tokens_in":9133,"feed_emoji":"🧲","tokens_out":9209,"duration_ms":82679,"temperature":0.7,"pith_summary":"The paper introduces the Hadamard sensing channel (HSC), a deterministic replacement for phase randomization in dynamical-decoupling quantum sensors. It claims that by assigning global phases to $M$ repeated pulse units according to the rows of a Hadamard matrix and averaging the outcomes of $M$ experimental runs, every spurious response caused by finite-duration pulses is suppressed by the factor $M$, with no statistical variance left behind. In the weak-coupling regime this exactly matches the suppression that phase randomization achieves only on average, while removing the need for continuous random phases and the hardware that generates them. The practical payoff is a hardware-friendly route to reliable nanoscale nuclear magnetic resonance, where the spurious response of one nuclear species can otherwise be mistaken for the genuine signal of another.","feed_headline":"Hadamard phases wipe out spurious quantum sensing signals","feed_subtitle":"A finite set of orthogonal phases suppresses artifacts exactly, matching random averaging without the variance or hardware cost.","key_machinery":"The load-bearing object is the Hadamard matrix $H_M$ with entries $s_{n,m}\\in\\{+1,-1\\}$ and the orthogonality identity $\\sum_n s_{n,m}s_{n,l}=M\\,\\delta_{m,l}$. Each row $n$ of the matrix specifies the phases for the $M$ pulse units in run $n$, with $+1$ mapped to phase $0$ and $-1$ to phase $\\pi$. These phase shifts commute with the sensor's $\\sigma_z$-type signal coupling, so the genuine target signal is untouched, while the spurious transverse modulation $F_\\perp(t)$ is multiplied by $s_{n,m}$; this changes only the spurious Fourier amplitude, making it $H_M^{(n)}$ times a common base amplitude. The orthogonality identity then turns the average of $|H_M^{(n)}|^2$ into exactly $1/M$, which is the whole suppression mechanism. The effective operation is a convex combination of unitary channels, which ensures HSC is a legitimate quantum channel.","core_discovery":"The central claim is captured by Eq. (12): the run-averaged squared Fourier weight of the spurious transverse modulation is exactly $1/M$ for every Fourier index $k$. Since the spurious signal intensity is quadratic in this weight in the weak-coupling regime, HSC suppresses all spurious responses by the factor $M$, matching ideal phase randomization but with zero statistical variance because the cancellation is deterministic. The derivation uses Hadamard orthogonality: summing the phase-entry products $s_{n,m}s_{n,l}$ over runs $n$ gives $M$ when $m=l$ and $0$ otherwise, which collapses the double sum to a single diagonal term. The authors support the claim with simulations of single- and dual-nuclear-spin sensing, showing that HSC keeps a genuine 1H signal distinguishable from a 13C artifact under pulse-angle and detuning errors where the standard sequence fails and phase-randomization contrast degrades.","pith_inferences":["The same sign-flip trick may extend to other control artifacts, such as crosstalk or leakage in multi-qubit gates, whenever the artifact can be switched by a sign and enters the signal linearly or quadratically.","If the $\\cos^2(\\varphi_k)$ phase factor is benign or bounded in the strong-coupling regime, the deterministic cancellation could extend beyond the paper's weak-coupling proof; a direct check would be to compute the full run-averaged intensity at larger coupling.","Instead of using all $M$ rows, one could select a subset of Hadamard rows to trade suppression depth against measurement time for a targeted artifact frequency.","The observed robustness to pulse-angle and detuning errors hints that HSC may also partially filter systematic control errors, a property stronger than the analytic proof alone establishes."],"forward_implications":["Every spurious spectral peak, at every Fourier index $k$, is uniformly suppressed by the same factor $M$ under HSC.","HSC removes the statistical noise floor of phase randomization, so the same suppression can be reached with fewer repeated measurements and shorter measurement campaigns.","HSC needs only the finite phase set $0,\\pi/2,\\pi,3\\pi/2$, enabling implementation with standard IQ mixers and avoiding high-resolution arbitrary waveform generators.","In the simulated 13C/1H scenario, HSC preserves the genuine proton peak under 10–20% control errors, a regime where the standard sequence fails and phase-randomization contrast drops.","Because the HSC operation is a convex combination of unitaries, it is certified as a CPTP channel and can be composed with other quantum-control primitives."],"supporting_citations":[{"why":"Defines the spurious harmonic response of multipulse sensing sequences that HSC targets.","marker":"[21]"},{"why":"Supplies the finite-pulse-duration spurious intensity formula used for the weak-coupling scaling.","marker":"[22]"},{"why":"The phase randomization protocol whose suppression factor and statistical variance HSC matches and eliminates.","marker":"[23]"},{"why":"The authors' previous Hadamard phase cycling work whose phase-stacking mechanism HSC adapts to sensing.","marker":"[20]"},{"why":"Provides the recursive Sylvester construction of Hadamard matrices used for the phase patterns in the examples.","marker":"[24]"}],"fun_headline_variants":["Hadamard phases cancel quantum sensing artifacts exactly","Deterministic Hadamard design wipes spurious signals from quantum sensors","Hadamard matrices make quantum sensing artifact-free without averaging","Hadamard design cancels quantum sensor artifacts with zero variance","No averaging, no variance: Hadamard phases suppress quantum sensing artifacts"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof assumes the weak-coupling regime, where the spurious intensity is quadratic in the Fourier amplitude, and it treats the $\\cos^2(\\varphi_k)$ phase factor as effectively constant across runs; if that phase factor varies from run to run, the exact factor-$M$ suppression may no longer be exact.","fun_headline_variants_meta":{"raw":{"variants":["Hadamard phases cancel quantum sensing artifacts exactly","Deterministic Hadamard design wipes spurious signals from quantum sensors","Hadamard matrices make quantum sensing artifact-free without averaging","Hadamard design cancels quantum sensor artifacts with zero variance","No averaging, no variance: Hadamard phases suppress quantum sensing artifacts"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000637,"raw_usage":{"total_tokens":2894,"prompt_tokens":860,"completion_tokens":2034,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":476,"completion_tokens_details":{"reasoning_tokens":1947}},"tokens_in":476,"tokens_out":2034,"duration_ms":12548,"temperature":1.0,"reasoning_tokens":1947,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T14:30:51.554617+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the full run-averaged spurious intensity, keeping the $\\cos^2(\\varphi_k)$ factor, for a finite-duration pulse sequence under 10–20% pulse amplitude and detuning errors, and check whether the residual scales as $1/M$ with zero run-to-run variance; equivalently, an experiment that measures the shot-to-shot variance of the sensor spectrum across $M$ HSC runs would settle whether the deterministic cancellation holds beyond the ideal weak-coupling model.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the spurious harmonic response of multipulse sensing sequences that HSC targets."},{"cited_title":"Loretz, J","cited_arxiv_id":null,"evidence_quote":"Supplies the finite-pulse-duration spurious intensity formula used for the weak-coupling scaling."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"The phase randomization protocol whose suppression factor and statistical variance HSC matches and eliminates."},{"cited_title":"Lombardi, W","cited_arxiv_id":null,"evidence_quote":"The authors' previous Hadamard phase cycling work whose phase-stacking mechanism HSC adapts to sensing."}],"review_version":1}