{"id":"b0e53019-94a1-4ddb-9329-98d2c6b0d4eb","arxiv_id":"2603.13020","paper_version":2,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":5.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":3,"one_line_summary":"Inexact proximal ADMM optimizes quantum gates under amplitude, bandwidth, sparsity, and total-variation constraints, cutting total variation by ~10–13× versus L-BFGS-B while reporting mean fidelities of 0.6363 (qutrit) and 0.9541 (two-qubit).","lead":"This paper tests a constraint-native quantum pulse optimizer (inexact proximal ADMM) that builds amplitude, bandwidth, sparsity, and total-variation limits into the search rather than adding them after fidelity optimization. It reports large smoothness gains with only moderate fidelity on several gates, framing a reproducible fidelity–complexity trade-off for hardware-aware control design.","discovery_kind":"new_method","skeptic_critique":{"model":"grok-4.5","headline":"Wrong manuscript body supplied; abstract-only claims for PADMM-Warm trade-offs remain unverifiable.","rationale":"The reader correctly flagged that only the abstract of the quantum-control paper is available and that the full-text block is the unrelated fluid-dynamics manuscript. That is the dominant obstacle: every quantitative pillar of the strongest claim (fidelities, TV ratios, method ranking, “reproducible trade-off”) is body-dependent. No secondary technical soft spot in the abstract (e.g., constraint choice or gate suite) can be elevated above this, because those soft spots themselves cannot be inspected. Honest non-finding on internal math/implementation follows: there is nothing further to attack until the right manuscript is present. Verdict remains UNVERDICTED; no adjustment toward ACCEPT/CONDITIONAL/REJECT is justified from abstract text alone. Agreement with the reader is full on both the weakest-assumption framing and the low-confidence stance.","tokens_in":7783,"tokens_out":524,"duration_ms":10346,"concrete_test":"Retrieve the correct PDF/source for arXiv:2603.13020 and verify that the qutrit and two-qubit result tables (PADMM-Warm vs L-BFGS-B, paired seeds) reproduce the abstract’s fidelity and total-variation factors; if the body is missing, mismatched, or omits paired-seed statistics/leakage, keep the claim unverified.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The strongest claim (PADMM-Warm mean fidelities 0.6363/0.9541 with TV reductions 13.2×/10.7× vs L-BFGS-B, defining a reproducible fidelity–complexity trade-off) depends entirely on numerical comparisons, constraint projections, and paired-seed protocol that live in the methods, algorithms, tables, and figures of arXiv:2603.13020. The CACHEABLE full-text block is a different paper (dipolar-solvent nanofluidic electroosmosis, arXiv:2603.13021). Without the actual proximal-ADMM updates, bandwidth/sparsity/TV operators, gate Hamiltonians, leakage metrics, or baseline settings, one cannot check whether the reported numbers are correctly computed, fairly compared, or robust. That absence is load-bearing: the abstract’s self-limited framing is coherent, but the claim cannot be confirmed or refuted from the supplied body.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"The abstract claims that optimising quantum-control pulses for fidelity first and imposing waveform constraints afterwards conceals the fidelity cost of smooth, band-limited, amplitude-admissible controls. As an alternative, the authors propose a constraint-native formulation based on inexact proximal alternating-direction method of multipliers (PADMM), jointly minimising gate infidelity subject to amplitude bounds, Fourier-domain bandwidth projection, amplitude sparsity, and total-variation regularisation. On a single-qubit gate, a leakage-prone qutrit gate, and a two-qubit entangler without a directly controlled target generator, and with random seeds paired across methods, PADMM-Warm is reported to reach mean qutrit and two-qubit fidelities of 0.6363 and 0.9541 while reducing total variation by factors of 13.2 and 10.7 relative to L-BFGS-B (with GRAPE and Krotov as additional baselines). The abstract self-limits the contribution: the results define a reproducible fidelity–complexity trade-off, not a universal fidelity advantage, and position the method as a numerical tool for low-complexity control frontiers rather than a replacement for unconstrained high-fidelity solvers.","tokens_in":8033,"tokens_out":1087,"duration_ms":14413,"significance":"If the numerical comparisons hold under fair baselines and well-specified constraint operators, the work would be a useful methodological contribution to quantum optimal control: it makes the fidelity cost of hardware-relevant waveform constraints explicit rather than post-hoc, and it reports a multi-gate, paired-seed protocol with leakage alongside computational-subspace fidelity. The self-limited framing (trade-off tool, not universal solver) is scientifically appropriate. However, significance cannot be assessed from the abstract alone; load-bearing content (proximal operators, inexactness tolerances, Hamiltonians, tables, and figures) is required to credit reproducibility or algorithmic novelty.","major_comments":[{"comment":"The supplied full-text body does not match the title, abstract, or arXiv id under review. The body is a continuum electrohydrodynamics manuscript on dipolar-solvent corrections to transient nanoscale electroosmotic flow (Poisson–Nernst–Planck–Stokes with Langevin–Bikerman permittivity and viscoelectric viscosity; arXiv-style content consistent with 2603.13021), not a quantum-control paper. Consequently there are no PADMM updates, constraint projections, gate Hamiltonians, leakage metrics, baseline settings, tables, or figures with which to verify the reported fidelities (0.6363 / 0.9541) or TV reduction factors (13.2 / 10.7). This mismatch is load-bearing: the central empirical claim cannot be checked, confirmed, or refuted from the provided manuscript.","section":null},{"comment":"Even taking the abstract at face value, the strongest claim depends on numerical protocol details that are not present in the supplied body: (i) definitions and implementation of the Fourier bandwidth projection, amplitude sparsity, and total-variation proximal maps; (ii) inexactness tolerances and warm-start policy for proximal ADMM; (iii) fairness of GRAPE / Krotov / L-BFGS-B baselines under the same amplitude and bandwidth limits (or a clear statement that baselines are unconstrained); and (iv) the paired-seed design and leakage definition for the qutrit. Without those sections, one cannot assess whether the trade-off is reproducible or an artifact of unequal constraint handling.","section":null},{"comment":"The abstract’s weakest load-bearing premise—that the chosen constraint set and the three selected gates suffice to map a general low-complexity control frontier—cannot be stress-tested without methods and results. Robustness to other device Hamiltonians, noise models, or alternative constraint encodings is not available for review in the supplied text.","section":null}],"minor_comments":[{"comment":"Abstract only: the acronym PADMM-Warm is used without expansion of the warm-start policy; a one-clause definition would help readers who stop at the abstract.","section":null},{"comment":"Abstract only: reporting mean fidelities without uncertainty (e.g., standard error over paired seeds) makes the 0.6363 / 0.9541 figures hard to interpret even at the abstract level.","section":null},{"comment":"Editorial: the CACHEABLE source appears to have swapped arXiv 2603.13020 (quant-ph control) with 2603.13021 (flu-dyn electroosmosis). The correct quantum-control PDF/source must be provided before a technical review can proceed.","section":null}],"recommendation":"uncertain","confidential_remarks":"I cannot issue accept / minor_revision / major_revision / reject on scientific grounds because the full text provided is a different paper from the abstract and arXiv id under review. Please re-supply the actual manuscript for 2603.13020 (methods, algorithms, tables, figures). Until then the only honest recommendation is uncertain. Possible cause: adjacent arXiv ids (13020 vs 13021) were swapped in the review packet."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"The first thing to know: the cacheable full text is not this paper. It is Srinivasula’s dipolar-solvent electroosmosis work (2603.13021). For Song’s quant-ph piece we effectively have the abstract only, so confidence has to stay low.\n\nFrom that abstract, the contribution is clear and modestly useful. Instead of optimising fidelity first and clipping or filtering the pulse later, the author folds amplitude bounds, Fourier bandwidth projection, sparsity, and total-variation into an inexact proximal ADMM loop and compares it head-to-head with GRAPE, Krotov, and L-BFGS-B on three gates (single qubit, leaky qutrit, two-qubit entangler without a direct target generator). Paired seeds and explicit leakage reporting are good practice. The self-limitation is also honest: PADMM-Warm is sold as a tool for mapping fidelity–complexity frontiers (mean fidelities 0.6363 / 0.9541 with TV cuts of 13.2× and 10.7× vs L-BFGS-B), not as a universal high-fidelity solver. That framing is rarer than it should be in this literature.\n\nWhat we cannot check is everything that would make the claim solid: the actual proximal updates, inexactness tolerances, regulariser weights, bandwidth cutoffs, Hamiltonians, baseline settings, and tables. Those free parameters are load-bearing. The three-gate suite is also thin for a “reproducible trade-off” claim. None of that means the abstract is wrong; it means the evidence is not in front of us.\n\nWho it is for: people who design experimental pulses under hardware filters and care about TV/sparsity as much as raw fidelity. A serious editor should still send it to referees if the real manuscript matches the abstract’s scope—engineering comparison with open code would make it stronger. I would not cite or put it in reading group until the correct PDF is available. Engage only after the body is in hand.","headline":"We only have the abstract for the quantum-control PADMM paper; the supplied body is a different nanofluidics manuscript, so the reported trade-off numbers stay unverifiable.","tokens_in":8637,"tokens_out":520,"would_cite":false,"duration_ms":10993,"reading_group":"no","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.5","headline":"Building hardware constraints into the optimizer itself yields a clear fidelity–complexity trade-off for quantum control pulses, not a free fidelity win.","keywords":["quantum control","proximal ADMM","pulse shaping","fidelity-complexity trade-off","total variation","band-limited controls","GRAPE","Krotov"],"falsifier":"On the same three gates (or a new device Hamiltonian), re-run the paired-seed comparison with a different constraint model or with realistic noise; if PADMM-Warm no longer simultaneously lowers total variation by an order of magnitude and keeps comparable subspace fidelity, the claimed trade-off does not hold.","tokens_in":8643,"feed_emoji":"⚛️","tokens_out":886,"duration_ms":16556,"temperature":0.7,"pith_summary":"Standard quantum-control design first maximises gate fidelity and only later enforces waveform limits such as amplitude caps, bandwidth, sparsity, and smoothness. That order hides how much fidelity is lost once the pulse must actually run on hardware. This paper instead optimises gate fidelity and those constraints together using inexact proximal alternating-direction updates. On a single-qubit gate, a leakage-prone qutrit gate, and a two-qubit entangler, the warm-started method reaches mean fidelities of about 0.64 and 0.95 for the harder cases while cutting total variation by factors of roughly 13 and 11 relative to an unconstrained quasi-Newton baseline. The result is framed as a reproducible map of low-complexity control frontiers, not a claim of universally higher fidelity. Readers who care about implementable pulses get a concrete numerical tool for trading smoothness and bandwidth against gate error under paired random seeds.","feed_headline":"Constraint-first pulses cut complexity 10× with usable fidelity","feed_subtitle":"Proximal ADMM trades gate error for smoothness and bandwidth on qubit, qutrit, and two-qubit targets.","key_machinery":"Inexact proximal ADMM (PADMM-Warm): alternating-direction updates that jointly minimise gate infidelity and apply proximal projections for amplitude bounds, band-limited Fourier content, sparsity, and total variation, so constraints are native to the search rather than post-processed.","core_discovery":"When amplitude bounds, Fourier bandwidth projection, amplitude sparsity, and total-variation regularisation are enforced inside the optimiser via inexact proximal ADMM rather than after unconstrained fidelity maximisation, the resulting pulses define a reproducible fidelity–complexity trade-off: PADMM-Warm attains mean qutrit and two-qubit fidelities of 0.6363 and 0.9541 while reducing total variation by factors of 13.2 and 10.7 versus L-BFGS-B, without claiming superiority over unconstrained high-fidelity solvers.","pith_inferences":["The same constraint-native template could be attached to other pulse optimisers (not only ADMM) to test whether the trade-off is method-specific or constraint-driven.","Hardware teams that already measure filter bandwidth and amplitude slew rates could use these regularisers as soft models of their electronics and re-rank candidate pulses before calibration.","If the two-qubit entangler without a directly controlled target generator is representative, similar gains may appear on any gate whose generator is only indirectly actuated."],"forward_implications":["Control designers can report fidelity together with total variation, bandwidth, and sparsity instead of fidelity alone.","Warm-started proximal ADMM becomes a practical tool for charting low-complexity pulse frontiers on leakage-prone and multi-qubit targets.","Unconstrained high-fidelity solvers remain the right choice when the only figure of merit is peak fidelity.","Paired random seeds and subspace-plus-leakage reporting become standard for fair method comparisons."],"fun_headline_variants":["Constraint-native ADMM maps fidelity-complexity trade-offs in pulse design","Proximal ADMM enforces bandwidth and TV limits inside quantum optimisers","PADMM-Warm cuts total variation 13× on qutrit gates at 0.64 fidelity","Inexact proximal updates yield smoother controls than post-hoc filtering","Constraint-first quantum control trades gate error for 10× lower complexity"],"cache_read_input_tokens":128,"weakest_assumption_plain":"The chosen constraint set and the three selected gates with paired seeds are enough to represent the dominant hardware limits and to establish a general, useful control frontier.","fun_headline_variants_meta":{"raw":{"variants":["Constraint-native ADMM maps fidelity-complexity trade-offs in pulse design","Proximal ADMM enforces bandwidth and TV limits inside quantum optimisers","PADMM-Warm cuts total variation 13× on qutrit gates at 0.64 fidelity","Inexact proximal updates yield smoother controls than post-hoc filtering","Constraint-first quantum control trades gate error for 10× lower complexity"]},"model":"grok-4.5","effort":"low","cost_usd":0.005646,"raw_usage":{"total_tokens":1525,"prompt_tokens":828,"num_sources_used":0,"completion_tokens":104,"cost_in_usd_ticks":56460000,"prompt_tokens_details":{"text_tokens":828,"audio_tokens":0,"image_tokens":0,"cached_tokens":128},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":593,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":828,"tokens_out":104,"duration_ms":5190,"temperature":1.0,"reasoning_tokens":593,"cache_read_input_tokens":128,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-14T21:55:27.619840+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"On the same three gates (or a new device Hamiltonian), re-run the paired-seed comparison with a different constraint model or with realistic noise; if PADMM-Warm no longer simultaneously lowers total variation by an order of magnitude and keeps comparable subspace fidelity, the claimed trade-off does not hold.","supporting_citations":[],"review_version":1}