{"id":"e3c7deec-1ad6-4591-bb30-d979253a9272","arxiv_id":"2508.20116","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":3,"one_line_summary":"A 21-mode superconducting Bose-Hubbard chain shows a first-order dissipative phase transition in which all modes jump and switch together, with dwell times up to 143 s captured by a single-mode mean-field model.","lead":"A chain of 21 superconducting microwave resonators, driven at only one site, switches all at once between a dim and a bright state, with dwell times from a few milliseconds to 143 seconds. The experiment maps a driven-dissipative, non-equilibrium phase transition in a Bose-Hubbard metamaterial and matches a minimal single-mode theory at moderate pump power.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Single-mode reduction is empirically violated at high pump power; its validity range for the phase-diagram fit is unstated.","rationale":"The reader's weakest_assumption identifies the same load-bearing point: the single-mode reduction between Eq. (2) and Eq. (3), with alpha=0 for all modes except the quasi-resonant one and photon-conversion terms neglected. I agree this is the most load-bearing assumption because all quantitative theory lines derive from it. The paper's own admission of high-power multimode emission not captured by the model is direct in-scope evidence that the reduction has a finite validity range, and the text does not specify whether that range covers the powers used in the phase-diagram comparison. This is a genuine concern, but it is not fatal: the authors explicitly scope the model to 'moderate pump powers,' and the central observation of a first-order dissipative transition with simultaneous multimode frequency jumps, hysteresis, and long switching times is supported by complementary measurements. The appropriate response is to keep the CONDITIONAL verdict, with the added condition that the authors state the fitted power range and quantify non-resonant mode populations or exclude the high-power region from the quantitative claim. No basis for rejection or for full acceptance exists without these checks.","tokens_in":13066,"tokens_out":10601,"duration_ms":123373,"concrete_test":"Re-analyze the stored PSD data used for Fig. 3(e): for each point on the fitted transition line, integrate the emission of each non-pumped mode and form R = sum_{k != pumped} n_k / n_pumped. If R exceeds a small pre-registered threshold (e.g., 0.05) for any eps/gamma >= 9.5 on the compared boundary, the alpha=0 assumption is violated there and the theory line must not be compared in that region. Additionally, repeat the global epsilon-scale fit using only data with eps/gamma < 9.5; if the best-fit epsilon scale changes by more than the experimental power uncertainty, the single-parameter quantitative claim fails. This directly tests whether neglected multimode population invalidates the model in the regime where agreement is claimed.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section III reduces the 21-mode Bose-Hubbard model to an effective single driven mode by (i) setting alpha=0 for all non-resonant modes (footnote 1) and (ii) dropping photon-conversion terms as 'not energy conserving.' Every quantitative prediction—the frequency-shift curves of Fig. 2(b), the universal phase boundary of Fig. 3(e), and the transition line in Fig. 4(f)—flows from this reduction. The reduction is not validated at the powers used for the phase-diagram comparison. The paper itself reports that at eps/gamma=9.5 'additional emission peaks emerge at other mode frequencies' and concedes in Section VI that at high pump powers there is 'multimode emission not captured by our current model.' If non-resonant modes acquire population or conversion terms contribute at any point on the compared phase boundary, the 'excellent agreement' with a single fitted pump scale is a numerical coincidence rather than evidence for the single-mode mechanism. The text does not state the maximum eps/gamma included in the Fig. 3(e)/4(f) fits, nor report the measured population of non-resonant modes along the boundary. This is the load-bearing gap: the central quantitative claim is only as strong as the unstated domain of validity of the single-mode reduction.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports the first experimental observation of a multimode dissipative first-order phase transition in a one-dimensional Bose-Hubbard chain of 21 superconducting nonlinear resonators. When a single mode is pumped, the measured resonance frequencies of all modes shift abruptly, with hysteresis, as the pump frequency or power is varied. The authors characterize the dim-to-bright transition using transmission spectroscopy, power spectral density measurements, and time-resolved switching statistics, obtaining dwell times from milliseconds to 143 s. They propose a single-mode mean-field model with self- and cross-Kerr nonlinearities and show that it reproduces the observed frequency shifts and the phase boundary, with the global pump amplitude scale as the only fitted parameter. A parameter-free prediction for the ratio of non-resonant mode shifts (4/3) is also confirmed. The paper concludes that this is the first observation of a multimode dissipative first-order phase transition in a Bose-Hubbard metamaterial and discusses the platform's potential for collective switching and sensing.","tokens_in":13332,"tokens_out":4021,"duration_ms":45583,"significance":"If the central claim holds, this is a significant advance in driven-dissipative many-body physics: a 21-site Bose-Hubbard array is one of the largest circuit-QED lattices to show a first-order dissipative transition, and the multimode character with cross-Kerr-mediated coupling is novel. The paper's strengths include three complementary experimental probes (frequency-jump spectroscopy with hysteresis, PSD phase diagrams, and switching-time statistics), an exceedingly long observed metastable lifetime, and at least one parameter-free prediction (the 4/3 cross-Kerr shift ratio) that matches the data. The analytical response-function framework and the careful statistical treatment of dwell times are also positive features. However, the central quantitative claim hinges on the single-mode reduction of the 21-mode model, and the manuscript does not yet establish the range of validity of that reduction over the parameter region used for the phase-diagram comparison.","major_comments":[{"comment":"The model reduction sets the mean-field amplitude of every non-resonant mode to zero and neglects photon-conversion terms as 'not energy conserving.' Every quantitative prediction shown in Figs. 2(b), 3(e), and 4(f) follows from this reduction. Yet the paper itself reports that at epsilon/gamma = 9.5 additional emission peaks appear at other mode frequencies (Fig. 3(b)) and concedes in Section VI that high pump powers give 'additional multimode emission not captured by our current model.' The text never states the maximum epsilon/gamma included in the fits of Figs. 3(e) and 4(f), nor does it report the measured population of non-resonant modes along the phase boundary. Please specify the exact parameter range used in the quantitative comparison, quantify the non-resonant mode occupations within that range, and estimate the magnitude of the neglected conversion terms for the fitted parame","section":"Section III, between Eq. (2) and Eq. (3); footnote 1"},{"comment":"The claim that the theory 'quantitatively reproduces the transition line' with 'pump amplitude as the only free parameter' is weakened by the calibration procedure. The global epsilon scale is fitted to the frequency-shift measurements in Section III, and an additional 4 dBm offset between PSD and spectroscopy/jump-rate data is introduced in Appendix G and 'accounted for' in the fitting of Fig. 4(d). Thus the absolute position of the transition line in the (Delta/gamma, epsilon/gamma) plane is not a parameter-free prediction. Please state explicitly every parameter that is adjusted (global pump scale, 4 dBm offset, any additional offsets) and show the raw, unshifted PSD transition points alongside the theory line. The shape of the line can still be a meaningful test, but the present text overstates the predictive content.","section":"Sections III, IV, and Appendix G"},{"comment":"The phase diagram in Fig. 3(e) is compared with a 'numerically computed transition line' obtained from exact diagonalization of the Liouvillian in the U/(gamma N) -> 0 limit, while the analytical boundary used in Fig. 4(f) is attributed to Ref. [53], which is cited as 'In preparation.' The central quantitative comparison therefore relies partly on an unpublished derivation. Please include the derivation of the analytical phase boundary in an appendix or cite a published version, so that the agreement is independently verifiable.","section":"Section IV and Fig. 3(e)"}],"minor_comments":[{"comment":"There is a notation inconsistency: the Hamiltonian in Eq. (1) uses omega_r for the bare resonator frequency, but the text says 'omega_0 is the resonant frequency of each cavity,' and Eq. (3) defines Delta = omega_p - omega_0. Please unify the symbols.","section":"Eq. (1) and surrounding text"},{"comment":"The sentence 'we conducted an Anderson-Darling test and rejected time traces failing at 15%' is ambiguous. Is 15% a significance level or a p-value threshold? Please state the exact rejection criterion and how many traces were rejected.","section":"Section V, Anderson-Darling test"},{"comment":"The color bar for pump amplitudes would be easier to read if the numerical values of epsilon/gamma were given in the caption. Also, adding representative error bars would help the reader judge the agreement with the theory.","section":"Fig. 2(b)"},{"comment":"The hysteresis loops are convincing, but adding vertical dashed lines marking the predicted bistability boundaries would make the comparison with the model quantitative.","section":"Appendix F, Fig. 8"}],"recommendation":"major_revision","confidential_remarks":"The paper is technically impressive and likely publishable after revision. The main risk is not the honesty of the experimental data but the explanatory claim: the single-mode reduction's domain of validity must be pinned down. The reliance on an 'in preparation' reference for the analytical phase boundary is also something the editor should watch, since it is load-bearing for the central quantitative statement. I would not reject, but the revision should include a clear statement of the fitted parameters, the range of validity, and a quantitative check of the neglected multimode terms."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a good experimental paper. The new thing is the simultaneous frequency jump of all 21 modes when only one mode is pumped, with a phase boundary cross-checked by two independent methods and switching times up to 143 s. That is a genuine advance over earlier single-resonator or two-photon bistability work, and the 4/3 cross-Kerr ratio is a parameter-free prediction that matches.\n\nWhat's done well: the experiment is careful. Three complementary measurements agree: spectroscopy frequency shifts, integrated PSD phase diagram, and dwell-time switching. The hysteresis in App. F supports first-order character. The paper honestly states the model is for moderate pump powers and that high-power multimode emission is not captured.\n\nSoft spots: the stress-test note has a point but it's not as damaging as it sounds. The single-mode reduction is explicitly scoped, and the phase-diagram comparisons appear to use powers where only the quasi-resonant mode emits. Still, the paper never states the maximum eps/gamma included in the fits, nor does it report non-resonant mode populations along the compared boundary. That should be added. The probe-induced 4 dBm offset is investigated in App. G but absorbed into fits rather than independently calibrated; that's a minor weakness. The switching-rate theory is in an unpublished companion paper [53], so the exponential fit is somewhat assumption-driven. And the relation to the thermodynamic limit (Vicentini et al. saying 1D loses the transition) is acknowledged but not deeply addressed; the N=21 result is what it is, but a short discussion would help.\n\nVerdict: this deserves peer review. The central observation is strong and the limitations are mostly stated. I'd want data/code release and the validity-range statement before signing off, but not a desk reject.","headline":"Solid, well-scoped experiment on a multimode dissipative phase transition in a 21-site circuit QED chain; the main caveat is the single-mode model's unstated validity range.","tokens_in":13883,"tokens_out":1561,"would_cite":true,"duration_ms":17760,"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":"Pumping one site of a 21-resonator Bose-Hubbard chain makes all modes jump at once—a multimode dissipative first-order phase transition.","keywords":["multimode dissipative phase transition","superconducting Bose-Hubbard chain","driven-dissipative systems","Kerr nonlinearity","first-order phase transition","metastable switching","quantum metamaterial","microwave resonators"],"falsifier":"Measure the photon population of an off-resonant mode directly at a pump power just above the onset of the extra multimode emission (around ε/γ ≈ 9.5). If that mode shows a clear emission peak—a nonzero α—while the model predicts α=0, the single-mode reduction is violated and the reported agreement on the transition line would need a collective description. A less direct check is to measure the frequency-shift ratio between two non-resonant modes in a chain with engineered disorder; the closed-boundary prediction 4/3 should break in a calculable way.","tokens_in":12928,"feed_emoji":"⚛️","tokens_out":7544,"duration_ms":75875,"temperature":0.7,"pith_summary":"The paper reports the first experimental observation of a multimode dissipative first-order phase transition in a one-dimensional Bose-Hubbard chain built from 21 superconducting resonators. When a single mode is driven, the entire chain jumps between a dim and a bright phase: the resonance frequencies of all 21 modes shift abruptly and simultaneously as the pump frequency or power is swept. The authors map the phase boundary in two independent ways—from the emitted power spectrum and from the statistics of switching times between the two metastable states—and find both agree with a minimal theory that reduces the chain to one effectively driven Kerr oscillator, with other modes only frequency-shifted by cross-Kerr interactions. The single free parameter, the pump-amplitude scale, fixes the whole transition line. The longest measured switching time, 143 seconds, shows how metastable the bright state can be.","feed_headline":"All 21 modes of a superconducting chain switch at once","feed_subtitle":"One pump flips a 21-resonator Bose-Hubbard array from dim to bright; theory predicts the whole jump.","key_machinery":"The argument is carried by a single-mode reduction of the Bose-Hubbard Hamiltonian in the Fourier-mode basis. The reduction postulates that only the quasi-resonant mode acquires a photon population (all other modes keep α=0 and are shifted only through cross-Kerr interactions), and that photon-conversion terms, which would transfer population between modes, can be neglected because they are not energy conserving. The resulting effective driven-dissipative Kerr oscillator has its mean-field photon amplitude fixed by the self-consistency equation |α|² [1 + (Δ/γ + |α|²)²] = ε²/γ², and the renormalized mode frequencies are read off from the pole of the response function, ω* = sqrt((Δ + 2γ|α|²)²","core_discovery":"On the paper's own terms, the discovery is that a driven 1D Bose-Hubbard chain with 21 sites exhibits a genuine first-order dissipative phase transition with multimode character: driving a single site renormalizes the frequencies of all modes through cross-Kerr coupling, and at the transition the system switches collectively between a dim state with near-zero photon population and a bright state with a macroscopic population. The transition line obtained from the emission spectrum and from the crossing of dim and bright dwell-time distributions is consistent with a single-mode mean-field model, and the frequency shifts of non-resonant modes follow the predicted 4/3 cross-Kerr ratio. The auth","pith_inferences":["If the single-mode reduction is the right effective description, then increasing the chain length N should make the transition sharper because the effective nonlinearity U/(γN) shrinks; the same device with more sites should display longer dwell times and a sharper jump, approaching the thermodynamic limit.","The additional multimode emission seen at high pump powers (ε/γ ≳ 9.5) may be the signature of the neglected photon-conversion terms becoming energy-conserving once the modes shift with power; if so, that regime is a natural place to look for true many-body corrections to mean-field theory.","The measured probe effect—where adding a few probe photons shifts the apparent transition—implies that any readout scheme perturbs the phase boundary; a probe-free or single-shot measurement protocol would be needed to confirm the intrinsic transition line.","The 4/3 cross-Kerr ratio was derived for closed boundary conditions; repeating the frequency-shift measurement on a chain with engineered disorder or open boundaries would provide a quantitative test of the mode-structure assumption."],"forward_implications":["A single pump tone can act as a collective switch: all 21 modes jump coherently between dim and bright states, so the chain may function as a multimode bistable element with one drive.","Two independent measurements—integrated emission spectrum and switching-rate statistics—locate the same transition line, meaning the phase boundary is a reproducible feature of the steady state, not an artifact of a particular readout.","Dwell times grow exponentially with pump strength up to 143 s, far beyond the intrinsic decay timescale, demonstrating metastability that could be exploited as a classical or quantum memory.","The single-mode mean-field theory, calibrated with one free parameter (the pump amplitude scale), reproduces the observed transition line across detunings and powers at moderate drive; this makes the model predictive within that regime.","The critical endpoint of the first-order line, where bistability disappears, is predicted to show critical behavior; the authors place it out of reach of time-resolved measurements but accessible to PSD, guiding future lower-nonlinearity experiments."],"supporting_citations":[{"why":"Supplies the single-mode optical bistability steady state used as the effective model's core.","marker":"[51]"},{"why":"Provides the analytical phase boundary and exponential switching-rate form used to locate the transition.","marker":"[53]"},{"why":"The theoretical study that dismissed first-order transitions in 1D chains, which this work claims to overturn.","marker":"[45]"},{"why":"Previous observation of bistability in a Josephson-junction array resonator; this paper extends it to multimode and collective behavior.","marker":"[27]"},{"why":"Establishes the quantum-tunneling-activated exponential dwell times expected across a first-order dissipative transition.","marker":"[4]"},{"why":"The scaling argument U/(γN N_JJ^2) that sets the effective nonlinearity and hence the thermodynamic-limit parameter.","marker":"[54]"},{"why":"The Josephson-junction resonator array platform the experiment is built on and whose parameters are characterized.","marker":"[46]"},{"why":"Identifies the double-peak emission as the self-Kerr nonlinear analog of the Mollow triplet, used to interpret the bright-phase spectrum.","marker":"[56]"}],"fun_headline_variants":["21 resonators flip together in a dissipative quantum phase transition","One pump drives a collective jump in a 21-site Bose-Hubbard chain","Dissipative phase transition flips all 21 superconducting resonator modes","A single drive makes 21 superconducting modes switch as one","Collective dim-to-bright jump in a 21-site quantum simulator"],"cache_read_input_tokens":2688,"weakest_assumption_plain":"The theory's quantitative agreement rests on the assumption that only the mode closest to the pump holds photons, that all other modes only have their frequencies pushed by cross-Kerr effects, and that photon-conversion processes can be ignored; if those processes matter at the powers used to map the phase diagram, the agreement is not explained by the claimed mechanism.","fun_headline_variants_meta":{"raw":{"variants":["21 resonators flip together in a dissipative quantum phase transition","One pump drives a collective jump in a 21-site Bose-Hubbard chain","Dissipative phase transition flips all 21 superconducting resonator modes","A single drive makes 21 superconducting modes switch as one","Collective dim-to-bright jump in a 21-site quantum simulator"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001066,"raw_usage":{"total_tokens":4305,"prompt_tokens":748,"completion_tokens":3557,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":492,"completion_tokens_details":{"reasoning_tokens":3465}},"tokens_in":492,"tokens_out":3557,"duration_ms":29350,"temperature":1.0,"reasoning_tokens":3465,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T17:55:02.133442+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the photon population of an off-resonant mode directly at a pump power just above the onset of the extra multimode emission (around ε/γ ≈ 9.5). If that mode shows a clear emission peak—a nonzero α—while the model predicts α=0, the single-mode reduction is violated and the reported agreement on the transition line would need a collective description. A less direct check is to measure the frequency-shift ratio between two non-resonant modes in a chain with engineered disorder; the closed-boundary prediction 4/3 should break in a calculable way.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the single-mode optical bistability steady state used as the effective model's core."},{"cited_title":"S´ epulcre, Analytical phase boundary of a quantum driven-dissipative Kerr oscillator from classical stochastic instantons, In preparation (2025)","cited_arxiv_id":null,"evidence_quote":"Provides the analytical phase boundary and exponential switching-rate form used to locate the transition."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Previous observation of bistability in a Josephson-junction array resonator; this paper extends it to multimode and collective behavior."},{"cited_title":"Risken, C","cited_arxiv_id":null,"evidence_quote":"Establishes the quantum-tunneling-activated exponential dwell times expected across a first-order dissipative transition."},{"cited_title":"Pechal, J.-C","cited_arxiv_id":null,"evidence_quote":"The scaling argument U/(γN N_JJ^2) that sets the effective nonlinearity and hence the thermodynamic-limit parameter."},{"cited_title":"Scigliuzzo, G","cited_arxiv_id":null,"evidence_quote":"The Josephson-junction resonator array platform the experiment is built on and whose parameters are characterized."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Identifies the double-peak emission as the self-Kerr nonlinear analog of the Mollow triplet, used to interpret the bright-phase spectrum."}],"review_version":1}