{"id":"6a8defc8-5332-49de-bbb7-04db73a82a51","arxiv_id":"2411.15039","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"An extended energy participation ratio method that uses the exact Josephson cosine reproduces the measured fluxonium qubit and resonator frequencies and dispersive shifts.","lead":"This paper extends the energy participation ratio (EPR) simulation method to highly anharmonic superconducting circuits such as fluxonium qubits, using the exact Josephson cosine instead of a truncated Taylor expansion. The authors fabricate and measure a fluxonium qubit and show that the extended EPR method agrees with experiment better than a lumped-element model.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No convergence check for the 30-Fock-state truncation underpins the EPR validation; the dispersive-shift agreement could be basis-size dependent.","rationale":"The reader's weakest assumption centers on Eq. (12) and the possibility that the linearized zero-point fluctuations are invalid for strong anharmonicity, making the linear-mode basis incomplete. My concern is the same root issue viewed through its operational consequence: the basis is truncated at 30 Fock states per mode, and no convergence evidence is provided. This is the most load-bearing concern because the method's novelty is the exact-cosine treatment, and the only approximations left are the linear-mode ZPFs and the finite Hilbert space. The experimental validation fits EJ, EL, and the resonator offset, so the dispersive shift is the key predictive test; it is also the observable most sensitive to the truncation, since high-lying Fock states contribute to the higher fluxonium levels that determine χ. The absence of a convergence study means the paper does not yet rule out the possibility that the agreement is an artifact of an under-converged basis. This is a concrete, fixable issue rather than a fundamental flaw; a basis-size check would settle it. The paper itself notes a discrepancy near 0.3 Φ0, which may signal such an artifact. Therefore the reader's CONDITIONAL verdict remains appropriate, and I do not recommend changing it.","tokens_in":17208,"tokens_out":10399,"duration_ms":105350,"concrete_test":"Recompute the EPR simulations at several flux points (notably Φ_ext/Φ0 = 0.3 and 0.5) using 40, 60, 80, and 100 Fock states per mode, and also test a three-mode truncation by including the next-highest-participation mode. If the predicted qubit frequency or dispersive shift changes by more than the experimental linewidth (≳1 MHz) between 30 and 100 states, the 30-state results are not converged and the agreement in Figs. 3–4 cannot be taken as validation of the method.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative claim—that the extended EPR analysis 'can fully describe the nonlinear coupling of a highly anharmonic circuit'—rests on diagonalizing the exact-cosine Hamiltonian of Eq. (17) in a basis of tensor-product Fock states of the two linear modes, truncated to 30 Fock states per mode (Sec. IV). The phase operators entering the cosine are built from the linear-mode zero-point fluctuations of Eq. (12). For the fluxonium at Φ_ext/Φ0 = 0.5, the qubit frequency is ~0.3 GHz while the linear mode frequency is ~6 GHz, so the potential minimum is displaced by many zero-point widths from the basis center; an accurate description of the low-lying eigenstates then requires a large number of Fock states. The paper does not report a convergence study with respect to basis size or the number of retained modes. The dispersive shift (Fig. 4) is the one observable that is not directly fitted, yet it is precisely the quantity most sensitive to the high-lying states that the truncation may miss. The unexplained deviation near 0.3 Φ0 is consistent with a basis-size artifact, so the claimed validation is not yet quantitatively secure.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper extends the energy participation ratio (EPR) method to highly anharmonic superconducting circuits by replacing the Taylor expansion of the Josephson potential with the exact cosine operator. The cosine is expressed in a Fock basis of the linear modes obtained from classical finite-element simulations, using the zero-point fluctuations extracted from the energy participation ratios. As a proof of concept, the authors design, fabricate, and measure a fluxonium qubit coupled to a readout resonator, and compare the measured qubit and resonator frequencies and the dispersive shift as functions of external flux with the extended EPR analysis and with a lumped-element model. They report good agreement for the frequencies and a better match for the dispersive shift than the lumped model, concluding that the extended EPR analysis can fully describe nonlinear coupling in highly anharmonic circuits.","tokens_in":17430,"tokens_out":5506,"duration_ms":54607,"significance":"If the numerical and conceptual approximations are adequately justified, this is a useful extension of black-box quantization to fluxonium-class circuits, which are of growing importance for quantum information processing. The manuscript is transparent about the three fitted parameters (EJ, EL, and resonator offset), and the dispersive shift is a genuinely higher-order prediction that is not directly fitted. The paper also ships open-source code and measurement data, which supports reproducibility. The main weakness is the absence of a convergence study for the truncated Fock basis, which is load-bearing because the dispersive shift is the key unfitted observable and could be sensitive to the truncation.","major_comments":[{"comment":"The manuscript truncates the Hilbert space of each eigenmode to 30 Fock states without a convergence study. The dispersive shift defined in Eq. (18) is the central unfitted observable and is known to receive contributions from high-lying qubit levels; therefore it is precisely the quantity most likely to be affected by the truncation. The unexplained deviation near Φ_ext/Φ0 = 0.3 in Fig. 4 could be a truncation artifact. Please provide a convergence check as a function of Fock truncation (e.g., 20, 30, 40, and 60 states per mode) at several flux points, including 0.3, and report the resulting sensitivity of χ and of the qubit and resonator frequencies. If the results are stable, state this explicitly; if not, the validation is not yet quantitatively secure.","section":"Sec. IV (paragraph beginning 'During EPR simulations')"},{"comment":"The zero-point fluctuations φ_mj are obtained from the linearized energy participation via φ²_mj = p_mj ℏω_m/(2E_j). The nonlinear Hamiltonian in Eq. (17) is treated exactly, but the basis itself is built from these linear zero-point fluctuations. For a fluxonium at Φ_ext/Φ0 = 0.5, the qubit frequency is ~0.3 GHz while the linear mode frequency is ~6 GHz, so the cosine term strongly mixes many Fock states and may also couple to modes beyond the two retained modes. The completeness of the two-mode Fock basis constructed from the linear modes is therefore not self-evident. Please justify this approximation, for example by comparing with exact circuit quantization for the same circuit parameters (e.g., using scqubits), or by studying convergence as additional modes are included in the EPR Hamiltonian.","section":"Sec. II, Eq. (12)"},{"comment":"The statement that the extended EPR analysis 'can fully describe the nonlinear coupling of a highly anharmonic circuit' is stronger than the evidence presented. The qubit and resonator frequency agreements in Fig. 3 rest partly on the fitted parameters EJ, EL, and the resonator offset, as the paper itself acknowledges. The genuinely predictive test is the dispersive shift in Fig. 4, but that display shows an unexplained discrepancy near 0.3 Φ0. The conclusion should be tempered accordingly, or additional parameter-free predictions should be provided (for example, the qubit frequency at flux points outside the two used for fitting, or an uncertainty estimate propagated from the junction capacitance CJ).","section":"Sec. IV (last paragraph) and Sec. V"}],"minor_comments":[{"comment":"The sentence 'see red points in Fig. 3(b)' following the discussion of the resonator avoided crossings should refer to Fig. 3(a), since the resonator frequency is shown in panel (a) and the qubit frequency in panel (b).","section":"Sec. IV (paragraph after Fig. 3)"},{"comment":"The expression 2χ = E_j φ²_qj φ²_rj /12 is given without specifying the sign convention; please state whether the usual dispersive shift is negative and whether the Taylor-expansion result reproduces the sign expected for the fluxonium-resonator system.","section":"Sec. II, Eq. (15)"},{"comment":"The junction capacitance is given as CJ = 50 ± 12 fF/μm², but the simulation results in Figs. 3 and 4 are presented without uncertainty bars. Please propagate the uncertainty in CJ (and, if possible, in the fitted parameters) into the predicted frequencies and dispersive shift, or state why these uncertainties are negligible.","section":"Sec. IV, Eq. (19)"},{"comment":"The definitions of C⋆ and Ccoup in Eqs. (A16) and (A17) are hard to parse because of the nested subscripts; a short verbal description of the effective capacitance network would improve readability.","section":"Appendix A, Eqs. (A16)-(A17)"}],"recommendation":"major_revision","confidential_remarks":"The paper fits within the scope of a quantum engineering or applied physics journal. The central methodological claim is promising, but the missing convergence study is a load-bearing gap that should be addressed before publication. The authors' transparency about the fitted parameters is a positive feature, and the open-source release of code and data makes the work easy to check."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Straight talk: this paper is a genuine extension of EPR to strongly anharmonic circuits, and the experimental validation on a real fluxonium is the right test. The exact-cosine implementation via matrix exponential, the inclusion of junction capacitance, and the comparison with a lumped model make it a useful contribution to the design-toolbox literature. The dispersive shift is the most convincing part—it is not a fitted quantity, and the EPR result tracks the data better than the lumped model, although there is a real bump near 0.3 Phi0 that the authors leave unexplained.\n\nThe weak spots are the ones the reader identified, and I agree with the conditionality. First, the frequency agreement is partially a fitting exercise: EJ, EL, and the resonator offset are adjusted to the same measured data. The authors acknowledge this, but the abstract and conclusion still say 'excellent agreement' and 'fully describe', which overstates what was shown. Second, the stress-test note is on the money: no convergence study is presented for the 30-Fock-state truncation. At half flux, the qubit well is displaced by many zero-point widths from the harmonic basis center, and the dispersive shift is exactly the observable most sensitive to high-lying states. The missing scan is a one-hour numerical check, not a research project, but it matters. Third, the linearized zero-point fluctuation in Eq. (12) is an assumption; the measured agreement suggests it works in practice, but a statement about the expected error would help.\n\nNone of this is fatal. The code and data are available (including the measurement repository), which is a real plus. The paper deserves a serious referee. It should come back with a convergence study, error bars on fitted parameters, and a more measured abstract. I would bring it to the group and cite the method in my own design work.","headline":"A useful EPR extension for fluxonium with a genuinely predictive dispersive-shift test, but the validation is softened by fitted parameters and a missing Hilbert-space convergence check.","tokens_in":17973,"tokens_out":2931,"would_cite":true,"duration_ms":30112,"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"],"model":"deepseek-v4-flash","headline":"The paper claims that extending the energy participation ratio method to keep the exact Josephson cosine makes black-box quantization accurate for strongly anharmonic circuits such as fluxonium.","keywords":["energy participation ratio","fluxonium qubit","superconducting circuits","circuit quantization","dispersive shift","strong anharmonicity","Josephson junction","finite-element simulation"],"falsifier":"The claim would be settled by applying the same pipeline to a new fluxonium device with a different geometry and, using $E_J$ and $E_L$ determined only from room-temperature junction measurements rather than fit to the qubit spectrum, comparing the predicted qubit and resonator frequencies and dispersive shift across the flux range; systematic disagreement would show that the linearized zero-point fluctuations in Eq. (12) break down under strong anharmonicity.","tokens_in":17032,"feed_emoji":"⚛️","tokens_out":7574,"duration_ms":67631,"temperature":0.7,"pith_summary":"The paper extends the energy participation ratio (EPR) method—which builds quantum Hamiltonians from the energy distribution of classical electromagnetic simulations—to strongly anharmonic circuits such as the fluxonium qubit. The standard EPR approach expands the Josephson cosine to low order, which fails when anharmonicity is large; the authors instead keep the exact cosine and implement it numerically through a matrix exponential of the junction phase operator. As a proof of concept, they design, fabricate, and measure a fluxonium qubit coupled to a readout resonator, and compare the measured qubit and resonator frequencies and the dispersive shift as functions of external flux with the extended EPR predictions. They find close agreement, including avoided crossings with higher fluxonium states, and better accuracy than a lumped-element model that treats the circuit islands as equipotential. The claim is that the extended EPR analysis can fully describe the nonlinear coupling of a highly anharmonic circuit.","feed_headline":"Exact-cosine EPR reproduces fluxonium and resonator spectra","feed_subtitle":"Keeping the full Josephson cosine matches measured flux dependence and beats lumped-element models.","key_machinery":"The machinery is the energy participation ratio (EPR), the fraction of a mode's total inductive energy stored in a given Josephson junction, together with the zero-point-fluctuation formula $\\varphi_{mj}^2 = p_{mj}\\hbar\\omega_m/(2E_j)$, which converts classically computed participation ratios into the quantum phase operator. The extension is to keep the full Josephson cosine, $-E_J[\\cos(\\varphi_j - \\varphi_{\\text{ext}}) + \\varphi_j^2/2]$, with the phase operator expanded in the two linear modes (qubit and resonator), implemented numerically in a truncated Fock basis via the matrix exponential of the phase operators. This replaces the usual fourth-order expansion, which would give an analytic but inaccurate dispersive shift, with an exact diagonalization from which the dispersive shift is read off as $2\\chi = (\\omega_{|1,1\\rangle} - \\omega_{|1,0\\rangle}) - (\\omega_{|0,1\\rangle} - \\omega_{|0,0\\rangle})$.","core_discovery":"The central claim is that a black-box quantization method based on energy participation ratios can handle circuits whose nonlinearity is far beyond the weak-anharmonicity regime. The authors replace the truncated Taylor expansion of the Josephson energy with the exact cosine term, evaluated in the Fock basis of the linear modes as a matrix exponential, and include the external flux as a phase in that term. Applied to a measured fluxonium-resonator system, the method reproduces the flux dependence of the qubit frequency from about 5 GHz down to about 300 MHz, the resonator frequency including avoided crossings, and the dispersive shift $2\\chi$ across the flux range, whereas a lumped-element model using the same electrostatic parameters is less accurate. The paper takes this as evidence that the mode distribution captured by the finite-element simulation renormalizes effective parameters such as charging energy and coupling strength, and that the extended EPR analysis 'can fully describe the nonlinear coupling of a highly anharmonic circuit.'","pith_inferences":["A natural next step, not taken in the paper, is to test the method on a device where the junction's Josephson energy is set independently by room-temperature resistance measurements, removing the two fitted energies and making the comparison fully parameter-free.","The linearized zero-point-fluctuation assumption could be probed by comparing the exact diagonalization of the full distributed Hamiltonian (where accessible) with the EPR prediction as $E_J/E_L$ is varied; deviations would reveal where the two-mode basis begins to fail.","Because the paper attributes the lumped model's error to the equipotential-island approximation, the EPR approach might be the practical route to computing mode-dependent 'renormalized' parameters for cross-chip fluxonium architectures, including tunable couplers."],"forward_implications":["If the extended EPR method is correct, designers can predict the full flux-dependent spectrum of a fluxonium-resonator system, including higher-level avoided crossings, from a classical eigenmode simulation plus two calibrated energy scales, without a lumped-element circuit model.","The predicted dispersive shift, which depends on many higher fluxonium levels, becomes a reliable design target, so qubit-resonator couplings and readout operating points can be chosen before fabrication.","The method's success implies that distributed mode structure renormalizes effective circuit parameters such as charging energy and coupling strength, so lumped-element fits should be checked against EPR simulations for small-capacitance devices.","Because the exact cosine is handled by a matrix exponential and only the Fock-basis truncation sets the accuracy, the same pipeline should extend to other strongly nonlinear elements and to multi-qubit circuits without changing the formalism."],"supporting_citations":[{"why":"Defines the energy participation ratio and the zero-point-fluctuation formula that the paper extends to strong anharmonicity.","marker":"[20]"},{"why":"Supplies the exact-cosine treatment of fluxonium in a lumped model, which the paper adapts to the EPR basis, and the statement that no analytic dispersive shift exists.","marker":"[40]"},{"why":"Introduces the fluxonium qubit, the device class whose strong anharmonicity motivates the extension.","marker":"[22]"},{"why":"Provides the open-source design-and-simulation tool whose eigenmode simulations yield the classical energy distribution.","marker":"[44]"},{"why":"Provides the numerical package used for the lumped-element comparison model.","marker":"[49]"},{"why":"Supplies the measured junction capacitance density used to include the small Josephson capacitance.","marker":"[45]"},{"why":"Supplies the flux-pulse-assisted reset protocol needed to measure the dispersive shift when the qubit frequency is below 1 GHz.","marker":"[47]"},{"why":"Used to correct the effective coupling for the resonator's modified mode distribution.","marker":"[12]"}],"fun_headline_variants":["EPR tames fluxonium with exact cosine","Exact cosine EPR captures fluxonium spectra","EPR with full Josephson cosine fits fluxonium data","Beyond weak anharmonicity: EPR with exact cosine","Exact-cosine EPR matches fluxonium and resonator"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole prediction rests on assuming that the size of the quantum jitter of the phase across the junction, computed from the linearized classical modes, is the same even when the nonlinearity is so large that the cosine cannot be truncated.","fun_headline_variants_meta":{"raw":{"variants":["EPR tames fluxonium with exact cosine","Exact cosine EPR captures fluxonium spectra","EPR with full Josephson cosine fits fluxonium data","Beyond weak anharmonicity: EPR with exact cosine","Exact-cosine EPR matches fluxonium and resonator"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001588,"raw_usage":{"total_tokens":6360,"prompt_tokens":1000,"completion_tokens":5360,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":616,"completion_tokens_details":{"reasoning_tokens":5283}},"tokens_in":616,"tokens_out":5360,"duration_ms":32584,"temperature":1.0,"reasoning_tokens":5283,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T14:33:32.612394+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"The claim would be settled by applying the same pipeline to a new fluxonium device with a different geometry and, using $E_J$ and $E_L$ determined only from room-temperature junction measurements rather than fit to the qubit spectrum, comparing the predicted qubit and resonator frequencies and dispersive shift across the flux range; systematic disagreement would show that the linearized zero-point fluctuations in Eq. (12) break down under strong anharmonicity.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the energy participation ratio and the zero-point-fluctuation formula that the paper extends to strong anharmonicity."},{"cited_title":"Dunsworth, A","cited_arxiv_id":null,"evidence_quote":"Provides the open-source design-and-simulation tool whose eigenmode simulations yield the classical energy distribution."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the measured junction capacitance density used to include the small Josephson capacitance."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the flux-pulse-assisted reset protocol needed to measure the dispersive shift when the qubit frequency is below 1 GHz."},{"cited_title":"Bourassa, F","cited_arxiv_id":null,"evidence_quote":"Used to correct the effective coupling for the resonator's modified mode distribution."}],"review_version":1}