{"id":"8a07d886-cf98-4fe8-a3ba-a0c0b2fba442","arxiv_id":"2412.13849","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"A Josephson-junction longitudinal readout achieves 99.8% measured fidelity in 202 ns and an estimated pure fidelity above 99.9% without first-stage amplification.","lead":"A superconducting qubit was read out through a resonator coupled by a Josephson junction, creating a longitudinal interaction that avoids the usual transverse-coupling errors. The team reports 99.8% measured readout fidelity in 202 ns without a first-stage amplifier, and estimates the pure measurement fidelity at above 99.9%.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The headline 99.9% pure measurement fidelity is an inferred number after subtracting preparation errors whose sizes are not independently quantified; the unquantified X-gate error alone could shift it below 99.9%.","rationale":"The Reader's conditional verdict is appropriate. The measured 99.8% readout fidelity at 202 ns is a direct experimental result, averaged over 10 rounds of 30,000 trials, and is credible even without a first-stage amplifier. However, the paper's most prominent claim—\"pure measurement fidelity above 99.9%\"—is not a directly measured quantity; it is obtained by subtracting estimated preparation errors from the raw 0.2% infidelity. The main text quantifies only the thermal-excitation contribution (about 0.1%) and leaves the X-gate error unquantified. Because the raw infidelity is only 0.2%, any unaccounted preparation error of order 0.05–0.1% directly changes the headline number. This is exactly the soft spot identified by the Reader, and the conditionality is warranted: the authors should supply the supplemental material, the uncertainties on the thermal estimate, a quantitative bound on the X-gate error, or an independent calibration of preparation errors. No evidence of fabrication or misconduct was found; the concern is about the completeness of the error budget, not the integrity of the measurement.","tokens_in":6380,"tokens_out":4489,"duration_ms":43044,"concrete_test":"Independently calibrate the preparation errors: measure the qubit population immediately after the X gate (and after any X12 pre-excitation sequence) and after the 1 μs vacancy after heralding, using either a separate high-fidelity reference readout or a randomized benchmarking protocol. Then recompute the pure measurement fidelity as raw infidelity minus the measured preparation-error contribution, with propagated uncertainties. If the resulting pure fidelity falls below 99.9% at the 202 ns operating point, the abstract's pure-fidelity claim should be downgraded to the directly measured 99.8%.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's headline claim of \"pure measurement fidelity above 99.9%\" rests on the final paragraph of the benchmarking section. The raw infidelity at long measurement time is 0.2% (Figure 3). To infer a pure measurement infidelity below 0.1%, the total preparation error must exceed 0.1%. The text assigns \"about 0.1%\" to thermal excitation accumulated in the 1 μs vacancy after heralding, and then says only that \"the non-ideal X gate from |0⟩ to |1⟩ is also a source of preparation error,\" without giving its magnitude. Since preparation errors enter almost linearly into the assignment probabilities P(0|1) and P(1|0), an unquantified X-gate error of even 0.05% would reduce the inferred pure fidelity below 99.9%. No uncertainty is reported on the 0.1% thermal estimate, and no independent calibration of the preparation errors is provided in the main text. The directly measured 99.8% fidelity is credible, but the more impressive 99.9% number in the title and abstract is not established by the evidence presented.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents a superconducting qubit readout architecture based on a Josephson-junction longitudinal coupler between a transmon qubit and a quarter-wavelength transmission-line resonator. The authors report that the transversal (XX) coupling can be largely canceled by destructive interference with capacitive contributions at zero external flux, and that the nonlinear resonator's steady states suppress decay and measurement-induced transitions. They benchmark the readout by measuring assignment probabilities and claim a 99.8% fidelity in 202 ns without a first-stage amplifier, and a 'pure measurement fidelity' above 99.9% after subtracting estimated preparation errors. The data are based on 10 rounds of 30,000 measurements.","tokens_in":6594,"tokens_out":10111,"duration_ms":77682,"significance":"If the directly measured 99.8% fidelity is confirmed, this is a significant advance: it demonstrates that a simple architecture without a cryogenic first-stage amplifier can reach state-of-the-art readout speed and fidelity, which would simplify large-scale quantum processors. The longitudinal-coupling design and the use of the resonator's nonlinear steady states are conceptually interesting. The paper also includes reasonable statistics (10 rounds of 30,000 measurements). However, the headline >99.9% pure measurement fidelity is not directly measured; it is an inference from subtracting preparation errors that are only loosely quantified in the main text.","major_comments":[{"comment":"The conclusion that 'the pure measurement fidelity is larger than 99.9%' is not supported by the numbers reported. The measured infidelity converges to 0.2%. To infer a pure infidelity below 0.1%, the total preparation error must exceed 0.1%. The text only states that 'about 0.1% population of excited state can be accumulated in the 1 µs vacancy after the heralding measurement' and that 'the non-ideal X gate from |0⟩ to |1⟩ is also a source of preparation error,' without giving the magnitude of the X-gate error or an uncertainty on the thermal estimate. For example, if the thermal estimate is 0.08% and the X-gate error is 0.01%, the inferred pure infidelity would be 0.11%, below 99.9%. Please provide numerical values and uncertainties for both contributions and a propagation-of-error analysis, or revise the headline claim to match the directly measured 99.8%.","section":"Finally, we benchmarked... (last paragraph of benchmarking section)"},{"comment":"The abstract states 'a measurement fidelity of 99.8% in 202 ns', but the benchmarking section states 'When the measurement time reaches 202 ns, the readout fidelity is 99.5%.' These two numbers are inconsistent as written. The authors should clarify which curve (with or without X12 pre-excitation) is referred to in each case, and reconcile the numbers.","section":"Abstract vs. benchmarking section"}],"minor_comments":[{"comment":"The phrase 'the Transmon qubit' should be 'a transmon qubit' for grammatical correctness.","section":"Introduction"},{"comment":"The sentence 'The non-ideal X gate from |0⟩ to |1⟩' is ambiguous: it could refer to the state-preparation pulse or to the X12 gate used for pre-excitation from |1⟩ to |2⟩. Please disambiguate these two gates and their respective errors.","section":"Benchmarking section"},{"comment":"The caption states 'Presentation of the multiple steady states' but Figure 2b shows the readout signal distributions; please rephrase for clarity.","section":"Figure 2 caption"},{"comment":"The statement 'fine-tuning the device parameters, such as the Josephson energy of CJJ and the coupling quality factor of the resonator' is vague; consider providing a sentence on how these parameters affect the readout fidelity.","section":"Conclusion"}],"recommendation":"major_revision","confidential_remarks":"The paper contains a strong, credible direct measurement (99.8% fidelity) and a conceptually interesting architecture. However, the title and abstract emphasize the inferred >99.9% pure measurement fidelity, which is not sufficiently supported by the main-text evidence. The authors should either provide a full preparation-error budget with uncertainties, or change the headline claim to the directly measured value. Also, the internal inconsistency between 99.5% and 99.8% at 202 ns must be resolved before publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe headline 99.9% number is the one thing to check before trusting this paper, but the core result — 99.8% readout fidelity in 202 ns with no cryogenic amplifier — is directly measured and credible. The architecture is genuinely new: a transmon coupled to a quarter-wave resonator through a small Josephson junction gives longitudinal (ZZ) coupling, and the residual transverse (XX) coupling is cancelled by destructive interference with parasitic capacitance. That is a real advance over earlier longitudinal proposals that stayed theoretical, and it is supported by the measured flux-dependent dispersive shift showing 90% ZZ contribution at zero flux.\n\nThe readout scheme itself also makes sense. The junction adds nonlinearity to the resonator, and driving into the bifurcated bright/dark steady states suppresses decay error and measurement-induced excitation. The multilevel pre-excitation improves the error by roughly 50%. The 99.8% figure comes from 10 rounds of 30,000 measurements with standard deviation error bars; that part is solid.\n\nThe soft spot is exactly where the stress-test lands. The abstract's \"pure measurement fidelity above 99.9%\" is not a direct measurement. The raw infidelity at long time is 0.2%. To get below 0.1% pure infidelity they subtract \"about 0.1%\" thermal excitation accumulated in the 1 µs vacancy after heralding, plus an unquantified X12-gate error. The text gives no magnitude or uncertainty for the X-gate error, and no independent calibration of the preparation errors is shown in the main text. Since the correction is roughly comparable to the gap between 0.2% and 0.1%, an X-gate error of even 0.05% would pull the inferred pure fidelity below 99.9%. So the title and abstract overreach relative to the presented evidence. The supplementary material may fix this, but it was not available for review and the main text alone does not close the gap.\n\nThere is also a smaller point: the complete cancellation of transverse coupling rests on simulated parasitic capacitance values (4 fF and 2.9 fF). That is plausible but not independently verified; the measured flux dependence is consistent, so I would not call it a flaw, just a caveat.\n\nBottom line: this is a serious experimental paper with a solid direct result and an interesting new coupling architecture. The 99.9% claim needs either the SM numbers or a more careful statement before it should stand as written. I would send it to peer review — the referee should have access to the supplement and should ask the authors to report the preparation-error budget with uncertainties.\n\nRecommendation: engage with it, cite the 99.8% direct result and the longitudinal-coupling architecture, but treat the 99.9% as provisional.","headline":"Direct 99.8% readout is real and impressive; the 99.9% pure-fidelity claim is an unquantified correction and should be treated as provisional.","tokens_in":7157,"tokens_out":1899,"would_cite":true,"duration_ms":15420,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["85.25.Cp","03.67.Lx","42.50.Pq"],"model":"deepseek-v4-flash","headline":"A superconducting qubit readout hits 99.8% fidelity in 202 ns without a first-stage amplifier, with pure measurement fidelity above 99.9%.","keywords":["superconducting qubit","quantum measurement","longitudinal coupling","readout fidelity","Josephson junction","nonlinear resonator","quantum error correction","multiplexed readout"],"falsifier":"A direct experiment that measures the pure measurement fidelity without relying on preparation-error subtraction — for example, by varying the post-heralding vacancy time from 1 µs to 10 µs and checking whether the extracted 0.1% thermal population stays consistent, or by performing a full state-tomography-based readout calibration that independently quantifies the X12 gate error — would settle whether the >99.9% figure is real.","tokens_in":6137,"feed_emoji":"⚙️","tokens_out":1498,"duration_ms":15706,"temperature":0.7,"pith_summary":"The paper claims a new readout architecture for superconducting qubits that reaches a measurement fidelity of 99.8% in 202 ns — and an estimated pure measurement fidelity above 99.9% — without using any cryogenic first-stage amplifier. It argues that a Josephson junction coupling the qubit to a transmission-line resonator realizes a genuine longitudinal (ZZ) interaction, eliminating residual transversal (XX) coupling by destructive interference. The nonlinearity of the junction also creates bistable resonator states that suppress qubit decay error and measurement-induced excitation. A sympathetic reader would care because readout has been the slow, lossy bottleneck of superconducting quantum computing.","feed_headline":"Qubit readout hits 99.9% fidelity without an amplifier","feed_subtitle":"A Josephson-junction longitudinal coupler reads a superconducting qubit in 202 ns, removing the need for cryogenic first-stage…","key_machinery":"The key object is the Josephson-junction-coupled quarter-wavelength resonator, whose interaction Hamiltonian contains a longitudinal ZZ term, gZZ σz a†a, alongside a transversal XX term that can be cancelled by capacitive-coupling interference at ϕext = 0. The junction also gives the resonator nonlinearity, producing bistable 'bright' and 'dark' photon-number branches; operating on the correct branch provides a large frequency shift for |1> while keeping |0> nearly empty, and the bistability holds the resonator state steady even if the qubit decays during the measurement.","core_discovery":"The central claim is that a longitudinal-interaction readout scheme, implemented with a Josephson junction as the coupler, achieves both high speed and ultrahigh fidelity: a measured average readout fidelity of 99.5% at 202 ns without pre-excitation, improving to 99.8% with an X12 gate that pre-excites |1> to |2>, and a pure measurement fidelity estimated above 99.9% after subtracting preparation errors. The architecture provides genuine longitudinal coupling by cancelling the unwanted transversal interaction at zero external flux, and the junction's nonlinearity creates a bright/dark bistable resonator response that holds the |0> state even under strong drive, decoupling the resonator from later qubit decay.","pith_inferences":["The inference from 99.8% raw to >99.9% 'pure' fidelity rests on two preparation-error estimates — a ~0.1% thermal population accumulating in the 1 µs post-heralding vacancy and an unquantified non-ideal X12 gate error — so the true pure-measurement number is only as good as those estimates.","If the same architecture scales to a multi-qubit chip with multiplexed readout, it could simultaneously solve the two classic readout bottlenecks — amplifier noise and decay-induced misclassification — which is the main reason the result matters for error correction.","A natural testable extension is to vary the post-heralding vacancy time and the X12 gate calibration, independently check the preparation-error budget, and verify whether the inferred pure fidelity stays above 99.9% under those variations.","The paper does not report the resonator's measured photon occupancy or the exact discrimination threshold used; an independent reproduction that measures the IQ-separation SNR directly would clarify how much of the 99.8% comes from the longitudinal shift versus the bistability."],"forward_implications":["Readout no longer requires a Josephson parametric amplifier or other first-stage cryogenic amplification, simplifying the millikelvin wiring and reducing device cost for large-scale processors.","The suppression of Purcell decay and measurement-induced excitation means readout time can be pushed well below 200 ns without sacrificing fidelity, approaching gate-operation speeds.","The genuine longitudinal coupling opens the possibility of frequency shifts beyond the dispersive limit, potentially enabling even faster or more selective readout in future designs.","The scheme is compatible with multiplexing and with reset protocols, so it can be integrated into quantum error correction cycles.","Fine-tuning junction parameters and resonator coupling quality factor is predicted to further raise fidelity beyond the demonstrated 99.9%.","The demonstrated fidelity surpasses the state-of-the-art dispersive readout result without needing the amplifier that prior work used, setting a new benchmark for amplifier-free measurement."],"supporting_citations":[{"why":"Defines the dispersive readout protocol that this work contrasts with and extends by replacing transversal coupling with longitudinal coupling.","marker":"[1]"},{"why":"Identifies measurement-induced excitation as a key error channel that the nonlinear-resonator scheme is designed to suppress.","marker":"[8]"},{"why":"Introduces the nonlinear-resonator (Josephson bifurcation) mechanism whose bistable steady states this paper exploits to hold the resonator state.","marker":"[15]"},{"why":"Provides the prior use of first-stage parametric amplifiers and multilevel readout that this work eliminates or improves upon; also the basis for the X12 pre-excitation protocol.","marker":"[9]"},{"why":"The state-of-the-art dispersive readout result (99%+ with amplifier) that this paper's 99.8% is claimed to surpass.","marker":"[27]"},{"why":"Shows the dressed-dephasing effect from strong transversal readout that the genuine longitudinal coupling avoids.","marker":"[23]"}],"fun_headline_variants":["Longitudinal readout hits 99.9% qubit fidelity in 202 ns","No amplifier needed: qubit readout reaches 99.9% fidelity","Josephson-junction coupler enables 99.9% qubit measurement","Superconducting qubit readout: 99.9% pure fidelity"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The claim that the pure measurement fidelity is above 99.9% depends on estimates that subtract about 0.1% thermal excitation accumulated in a 1 µs vacancy after heralding and an unquantified X12 gate error; if those preparation-error numbers are wrong or incomplete, the inferred pure measurement fidelity would change.","fun_headline_variants_meta":{"raw":{"variants":["Longitudinal readout hits 99.9% qubit fidelity in 202 ns","No amplifier needed: qubit readout reaches 99.9% fidelity","Josephson-junction coupler enables 99.9% qubit measurement","Superconducting qubit readout: 99.9% pure fidelity"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000422,"raw_usage":{"total_tokens":2121,"prompt_tokens":854,"completion_tokens":1267,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":470,"completion_tokens_details":{"reasoning_tokens":1182}},"tokens_in":470,"tokens_out":1267,"duration_ms":9223,"temperature":1.0,"reasoning_tokens":1182,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T12:43:32.203752+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A direct experiment that measures the pure measurement fidelity without relying on preparation-error subtraction — for example, by varying the post-heralding vacancy time from 1 µs to 10 µs and checking whether the extracted 0.1% thermal population stays consistent, or by performing a full state-tomography-based readout calibration that independently quantifies the X12 gate error — would settle whether the >99.9% figure is real.","supporting_citations":[{"cited_title":"Blais, R.-S","cited_arxiv_id":null,"evidence_quote":"Defines the dispersive readout protocol that this work contrasts with and extends by replacing transversal coupling with longitudinal coupling."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Identifies measurement-induced excitation as a key error channel that the nonlinear-resonator scheme is designed to suppress."},{"cited_title":"Mallet, F","cited_arxiv_id":null,"evidence_quote":"Introduces the nonlinear-resonator (Josephson bifurcation) mechanism whose bistable steady states this paper exploits to hold the resonator state."},{"cited_title":"Mutus, T","cited_arxiv_id":null,"evidence_quote":"Provides the prior use of first-stage parametric amplifiers and multilevel readout that this work eliminates or improves upon; also the basis for the X12 pre-excitation protocol."},{"cited_title":"Swiadek, R","cited_arxiv_id":null,"evidence_quote":"The state-of-the-art dispersive readout result (99%+ with amplifier) that this paper's 99.8% is claimed to surpass."},{"cited_title":"Boissonneault, J","cited_arxiv_id":null,"evidence_quote":"Shows the dressed-dephasing effect from strong transversal readout that the genuine longitudinal coupling avoids."}],"review_version":1}