{"id":"f7eea720-2306-433f-b146-29dae24f095a","arxiv_id":"2512.08328","paper_version":2,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Demonstrates deterministic quantum state transfer and remote entanglement between fixed-frequency superconducting qubits on separate chips using broadband resonators and frequency-tunable photon generation.","lead":"Two separate superconducting chips, each with a fixed-frequency qubit, exchange a single microwave photon to transfer a quantum state and create remote entanglement, without using any frequency-tunable circuit elements. The trick is a broadband two-resonator design plus a Raman-based photon-generation method that lets the photon frequency be tuned electronically.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Receiver leakage beyond |f> is acknowledged but never bounded; because fidelities are computed after projecting qutrit data onto the qubit subspace, the reported 78%/73% may be optimistic.","rationale":"The reader's verdict is well-supported: the experiment is direct, loss and absorption are characterized independently, and the fidelities are well above classical thresholds. However, the paper itself flags an unresolved receiver leakage (Appendix C2) and then computes all headline fidelities after projecting qutrit data onto the qubit subspace (Appendix C3). This is the least secure link between the data and the quantitative central claim. The Pg-based loss correction is a related but secondary issue; even if perfect, it does not address leakage that is discarded by the projection. The proposed check is decisive: if the full-state fidelity including leakage is within error of the reported qubit-subspace fidelity, the concern is dismissed; if not, the correct statement is that qubit-subspace fidelities are 78%/73% with leakage contributing additional infidelity. Because this is a missing analysis rather than a demonstrated error, conditional acceptance is appropriate.","tokens_in":19693,"tokens_out":14157,"duration_ms":158927,"concrete_test":"Reanalyze the raw single-shot qutrit tomography data from Fig. 4 to compute the fidelity of the full unnormalized (two-)qutrit output—including all population in |f> and higher states as error—against the target state embedded in the qubit subspace, without renormalizing the qubit block. If the full-state fidelities remain within ~1–2% of the reported 78%/73%, the concern is resolved. If they drop by the 3–10% scale of the observed leakage, the headline should be revised to state clearly that the reported fidelities are qubit-subspace fidelities and that leakage contributes an additional uncharacterized infidelity.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The paper's own Appendix C2 reports that the receiver exhibits unexpectedly large decay from |f>–|e> and leakage to states beyond |f> during the photon-emission drive, with leakage population fluctuating between 0.03 and 0.1. In Appendix C3, state-transfer process tomography and Bell-state tomography are performed on (two-)qutrit density matrices and then projected onto the qubit subspace. Any population in |f> or higher is therefore discarded and the surviving qubit block is renormalized. The headline numbers Fp≈0.78 and FBell≈0.73 are qubit-subspace fidelities, not full-state fidelities. The transfer/Bell protocols drive the same |f0>–|g1> transition in reverse at the receiver, so the leakage observed in emission should also have been quantified for absorption. The numerical model in Appendix C4 is a single-pole cascaded model with only T1, Tφ, photon loss, and absorption inefficiency; it contains no leakage channel. The error budget sums to ~21% infidelity with no explicit leakage term, so the agreement between simulation and the projected fidelities does not test the leakage hypothesis. The central quantitative claim is therefore not yet bounded against a known, unresolved device defect.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports deterministic quantum state transfer and remote Bell-state generation between fixed-frequency transmon qubits on separate chips, using itinerant microwave photons. A frequency-tunable photon-generation method from the authors' earlier work is combined with two-resonator broadband transfer resonators to compensate a ~50 MHz fabrication offset, yielding a 30 MHz operating range. Quantum process tomography yields state-transfer process fidelities around 78% and Bell-state fidelities around 73%. Photon propagation loss and receiver absorption efficiency are characterized separately, and a numerical cascaded-system simulation using measured parameters reproduces the reported fidelities.","tokens_in":20048,"tokens_out":8783,"duration_ms":86169,"significance":"If the reported fidelities can be taken at face value, this is an important experimental step: it removes flux-tunable elements from remote microwave quantum communication, with potential scalability benefits for modular superconducting processors. The two-resonator broadband design, the separate loss/absorption determination, and the Monte Carlo study of fabrication tolerance are concrete strengths. However, the unresolved receiver leakage documented in Appendix C2 bears directly on the headline quantitative claims, and the paper does not currently bound its effect. The central idea is sound and likely correct, but the fidelity numbers need to be placed on a firm footing before publication.","major_comments":[{"comment":"The reported fidelities F_p≈0.78 and F_Bell≈0.73 are qubit-subspace fidelities: Appendix C3 states that qutrit tomograms are projected onto the qubit subspace. Appendix C2 reports an unresolved leakage of the receiver qubit to states above |f> during the |f0>–|g1> drive, with populations fluctuating between 0.03 and 0.1. The same drive is used in the communication protocols, but the simulation in Appendix C4 (Eqs. C1–C2) contains no higher-level leakage channel, and the error budget in Sec. III B lists only photon loss, absorption inefficiency, T1, and Tφ. The simulation/error-budget agreement therefore does not constrain the fidelity reduction from leakage. Because projecting and renormalizing the qubit block discards leakage events, the reported fidelities could overstate the actual process by up to ~0.1. The authors should extract the leakage population from the same qutrit tomograms","section":"Appendix C2/C3 and Sec. III B"},{"comment":"The photon-loss estimate L is obtained by comparing sender and receiver emission amplitudes after normalizing each waveform by sqrt(P_g), under the assumption that P_g equals the average emitted photon number. The receiver's P_g is degraded by the same unresolved leakage/decay mechanism described in Appendix C2. If the leakage channel contributes incoherently or modifies the waveform shape beyond a global efficiency factor, the extracted L≈29% could be biased. Since photon loss dominates the infidelity budget (≈15%), a sensitivity analysis of the sqrt(P_g) normalization would be needed to firmly support the quantitative fidelity claims.","section":"Sec. III A, Eq. (2), and Appendix C2"}],"minor_comments":[{"comment":"The text says 'Figure 7(a) shows the qubit population dynamics during the |f0>–|g1> photon-emission pulse', but the correct panel appears to be Fig. 7(c). Please fix the cross-reference.","section":"Appendix C2"},{"comment":"The sentence 'One of the design strategies described in Sec. B 2 involves maximizing...' appears to be a self-reference; it should refer to the relevant section of the main text or Appendix.","section":"Appendix B 2"},{"comment":"The text states that the measured absorption inefficiency is incorporated directly into η while also listing absorption inefficiency as a separate error contribution. Please clarify how the 1.5–3.0% absorption contribution is isolated from the combined η parameter in the simulations.","section":"Appendix C4"},{"comment":"The error-budget percentages are given without explicit uncertainty ranges. Given the observed coherence-time fluctuations (Appendix A) and the leakage fluctuations (Appendix C2), providing uncertainties on each contribution would make the budget more informative.","section":"Sec. III B"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"You should know two things about arXiv:2512.08328. First, it delivers the first deterministic state transfer and remote Bell-state generation between separate chips using only fixed-frequency transmons, no flux-tunable elements anywhere. That is a genuine enabler for modular superconducting networks. Second, the headline fidelities (process ~78%, Bell ~73%) are qubit-subspace projections: the paper explicitly projects qutrit tomographies onto the qubit block, discarding population in |f> and above. The receiver is known to leak to those higher levels during emission (0.03–0.1 population, Appendix C2), and the numerical model has no leakage channel at all. The error budget sums to ~21% infidelity with no leakage term, so the simulation–data agreement does not test the leakage hypothesis. The reported numbers are likely optimistic by a few percent, maybe more if the leakage also affects absorption.\n\nWhat is genuinely new: the broadband two-resonator transfer resonator (a flat Butterworth-type receiver, an emission-rate-optimized sender) and the extension of the authors' own frequency-tunable shaped-photon technique to inter-chip communication. Previous photon links used flux-tunable qubits or couplers; this work shows a path around that scaling bottleneck. The experiments are careful: three-state readout with assignment-matrix correction, independent loss and absorption measurements exploiting the circulator, tomography on qutrits, and a cascaded-system simulation that reproduces the projected fidelities using independently measured parameters. The Monte Carlo design study with 90% matching probability is a useful engineering contribution.\n\nWhere I push back on the reader's enthusiasm: the leakage problem is acknowledged but not bounded in a way that protects the quantitative claims. The photon-loss estimate is corrected by the Pg-based normalization, which itself rests on an assumption the paper states but does not justify. That said, this is not a fatal flaw. The qualitative result—fixed-frequency, broadband, frequency-tunable inter-chip photon link—stands on the tomography and the absorption-efficiency measurements. The leakage is a device-level defect that a referee could reasonably ask the authors to quantify in revision.\n\nThis paper deserves serious peer review: it is novel, reproducible in the experimental sense, and the right kind of engineering advance for the modular-superconducting community. Send it to referees, but insist that the leakage be either bounded, included in the error model, or reported through full-state fidelities before publication. I would bring it to a reading group focused on interconnects, and I would cite it in work on scalable networks—though with a note about the subspace projection.","headline":"A real first: deterministic inter-chip microwave quantum communication with only fixed-frequency qubits; the reported fidelities are qubit-subspace numbers that could be a few percent optimistic given unquantified receiver leakage.","tokens_in":20475,"tokens_out":4663,"would_cite":true,"duration_ms":47263,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Two fixed-frequency superconducting qubits on separate chips exchanged quantum states through a microwave photon, at around 78% process fidelity.","keywords":["superconducting qubits","quantum state transfer","remote entanglement","itinerant microwave photons","fixed-frequency transmons","broadband resonators","frequency-tunable photon generation","modular quantum networks"],"falsifier":"Perform qudit-resolved single-shot readout on the receiver immediately after the emission pulse, resolving populations in |g>, |e>, |f>, and higher levels, and compare the |g> population with an independently calibrated average photon number; if they disagree beyond uncertainty, recompute the propagation loss from the waveform ratio and rerun the simulated fidelity budget.","tokens_in":1227,"feed_emoji":"📡","tokens_out":2097,"duration_ms":69628,"temperature":0.7,"pith_summary":"This paper tries to establish that deterministic quantum state transfer and remote entanglement between fixed-frequency superconducting qubits on separate chips are possible without flux-tunable circuit elements, using an itinerant microwave photon as the carrier. The key move is to compensate for fabrication-related frequency mismatch from the emitter side: a resonator-assisted Raman transition lets the sender choose the photon frequency without modifying circuit parameters, and two coupled coplanar-waveguide resonators widen the transfer bandwidth past 100 MHz. The paper reports process fidelities around 78% for state transfer and Bell-state fidelities around 73% across a 30-MHz frequency range, with photon absorption around 95%. If correct, this removes a significant control complexity from modular superconducting processors and points toward fixed-frequency quantum networks.","feed_headline":"One microwave photon links fixed-frequency qubits at 78%","feed_subtitle":"Twin resonators broaden bandwidth, bridging a 50-MHz fabrication offset without tunable hardware.","key_machinery":"The carrying mechanism is a transfer resonator made of two coupled coplanar-waveguide resonators, giving two eigenmodes whose combined linewidth exceeds 100 MHz. On the sender side the two modes are placed to maximize the range where the photon-emission rate, computed from the two-mode response functions, stays above a design threshold; on the receiver side the modes are separated by roughly their linewidth to give a flat, wide absorption spectrum. Complementing this, a resonator-assisted Raman transition between the qubit's |f0> and |g1> states is driven off-resonantly, letting the photon frequency be chosen independently of the qubit's fixed transition frequency.","core_discovery":"The central claim is that frequency mismatch between sender and receiver—here about 50 MHz—does not have to be engineered away by tunable qubits. Instead, the sender's photon-emission process is driven off-resonantly so the emitted photon's frequency can be tuned to a frequency where both broadband transfer resonators respond; the receiver, designed for a flat broad response, absorbs it. In the overlapped 30-MHz band the paper demonstrates both quantum state transfer and remote Bell-state generation, and numerical simulations reproduce the measured fidelities with photon loss during propagation (about 15%) as the dominant error source.","pith_inferences":["The paper leaves implicit that the photon-emission-rate formula encodes a design continuum: adding more coupled resonators could further widen bandwidth or raise emission rate, though this is not demonstrated and would add engineering complexity.","The practical value of the result hinges on whether the photon-loss correction via the measured |g>-state population holds under the receiver's unexplained leakage; an independent qudit-resolved measurement after emission would confirm or correct the reported fidelities.","Combining frequency-tunable photon generation with non-adiabatic wave-packet shaping could remove the current 2-MHz photon-bandwidth limitation, making the protocol faster while keeping fixed-frequency hardware.","If the two-resonator approach extends to more coupled resonators, each node's communication spectrum could be shaped independently, potentially improving multi-node network integration; this is an extrapolation from the paper's analysis."],"forward_implications":["Modular superconducting processors can be built from fixed-frequency qubits, eliminating flux-bias lines and the associated control complexity and noise channels.","A fabrication offset of about 50 MHz between otherwise identical chips can be tolerated across a 30-MHz overlapping communication band; Monte Carlo modeling estimates a near-90% matching probability within a fabrication batch.","The broad operational bandwidth supports frequency-division multiplexing, so several quantum channels could share one transmission line at different carrier frequencies simultaneously.","Because the photon frequency is selectable in software, the protocol can avoid two-level-system defects by hopping to a cleaner frequency without hardware retuning.","The infidelity budget is dominated by propagation loss, so improvements in cable loss or receiver absorption efficiency would directly raise both state-transfer and Bell-state fidelities."],"fun_headline_variants":["Fixed-frequency qubits link via tuned photons, 79% fidelity","Photon tuning, not qubit tuning, bridges fixed-frequency chips","Remote quantum entanglement on fixed-frequency qubits","Broadband resonators let fixed qubits swap states remotely"],"cache_read_input_tokens":21888,"weakest_assumption_plain":"The reported fidelities assume that the receiver's measured |g>-state population after an attempted emission equals the number of photons actually emitted; if leakage to states above |f> biases that correction, the inferred photon loss and the reported fidelities could be inflated.","fun_headline_variants_meta":{"raw":{"variants":["Fixed-frequency qubits link via tuned photons, 79% fidelity","Photon tuning, not qubit tuning, bridges fixed-frequency chips","Remote quantum entanglement on fixed-frequency qubits","Broadband resonators let fixed qubits swap states remotely"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000714,"raw_usage":{"total_tokens":3034,"prompt_tokens":720,"completion_tokens":2314,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":464,"completion_tokens_details":{"reasoning_tokens":2246}},"tokens_in":464,"tokens_out":2314,"duration_ms":16279,"temperature":1.0,"reasoning_tokens":2246,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-03T17:46:13.040756+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Perform qudit-resolved single-shot readout on the receiver immediately after the emission pulse, resolving populations in |g>, |e>, |f>, and higher levels, and compare the |g> population with an independently calibrated average photon number; if they disagree beyond uncertainty, recompute the propagation loss from the waveform ratio and rerun the simulated fidelity budget.","supporting_citations":[],"review_version":1}