{"id":"5824c7c0-b82b-4480-9dfc-0c06652df4c2","arxiv_id":"1908.04844","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Photon-pair scattering in qubit arrays is resonantly enhanced at double-excited subradiant states, and newly identified twilight states generate long-lived photon-photon correlations.","lead":"This paper calculates how pairs of photons scatter inelastically from a small row of qubits coupled to a one-dimensional waveguide. It finds that the scattering is strongly enhanced when the pair's total energy matches a long-lived double-excited subradiant state, and that this can create long-lived photon-photon correlations.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central subradiant resonances inherit the Markovian replacement of H(1)(ω) by H(1)(ω0); the paper states only subwavelength spacing, not the weak-coupling condition needed to protect the O(ϕ²Γ0) linewidths.","rationale":"The reader's weakest assumption is exactly the Markovian replacement of H(1)(ω) by H(1)(ω0), and my independent reading identifies the same step as the most load-bearing for the central claim. The reason it is load-bearing rather than cosmetic is that the claimed effect is resonant: the subradiant peaks are very narrow, with width ∼ϕ²Γ0, so small frequency-dependent phase errors are amplified in peak heights and positions. The paper does disclose the approximation and restricts to subwavelength spacing, which deserves credit; however, the subwavelength condition alone does not guarantee the approximation's accuracy unless Γ0/ω0 is also small, and that additional condition is not stated. I do not find an internal inconsistency in the main scattering derivation, and the analytic residue form of Eq. (10) plus the optical-theorem check give independent support for the two-photon resonance mechanism within the stated approximation. The unsupported general N=2M threshold and the speculative quantum-information applications are secondary concerns that also support a conditional verdict, but they do not threaten the two-photon resonance claim as directly as the Markovian scope condition. Therefore I recommend keeping the reader's conditional verdict unchanged.","tokens_in":20431,"tokens_out":37083,"duration_ms":386069,"concrete_test":"Recompute the N=4, ϕ=0.1, χ=10^4Γ0 two-photon scattering spectrum keeping the full frequency-dependent phases (ω/c)|z_i−z_j| in H(1)(ω), i.e., evaluate G(ω)=[ω−H(1)(ω)]^{-1} and Σ_ij(ε)=∫ G_ij(ω)G_ij(2ε−ω)dω/(2π) numerically without the Markovian replacement. Compare the positions and peak intensities of the ε1 and ε2 subradiant resonances with the Markovian Fig. 3 for Γ0/ω0=0.01 and 0.1. If the shifts are below the O(ϕ²Γ0) linewidth and the peak heights change by less than 50%, the Markovian condition is adequate; otherwise the central enhancement is not robust.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The strongest claimed prediction—sharp two-photon resonances at Re(2ε)=2Re(εν) with linewidths ∼ϕ²Γ0 that survive destructive interference and can exceed single-qubit scattering—is computed after replacing the frequency-dependent H(1)_ij(ω) in Eq. (3) by H(1)_ij(ω0). This replacement is applied both in the single-particle Green functions s±(ω) and in Σ(ε), fixing the inter-qubit phases at ω0 everywhere in the kernel Q of Eq. (8). For the N=4 subradiant eigenvalues of Eq. (7), the relevant detunings from ω0 are of order ϕΓ0, so the neglected phase variation is δ(ω|z_i−z_j|/c)∼(ϕΓ0)(ϕ/ω0)=ϕ²Γ0/ω0. The induced correction to the subradiant linewidth, which is itself ϕ²Γ0, is a relative fraction Γ0/ω0. The stated justification, subwavelength spacing ϕ≪1, does not control Γ0/ω0; in strongly coupled waveguide-QED implementations Γ0/ω0 can approach 0.1, at which the linewidths, resonance positions, and the order-of-magnitude enhancements in Fig. 3 are not protected. This is a scope condition rather than an internal inconsistency, but it is exactly the condition on which the headline enhancement rests.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper develops a Green's-function formalism for inelastic two-photon scattering in a 1D waveguide with an array of two-level (or anharmonic) qubits, and uses it to identify double-excited subradiant states as the origin of sharp resonances. The authors classify double-excited eigenstates into superradiant, twilight, and subradiant states using the single-photon decay amplitudes d_j defined in Eq. (4), and they derive a closed-form expression for the scattering kernel Q in the Markovian approximation, Eqs. (8) and (9). For N=4 qubits they give explicit eigenvalues for the two subradiant states, Eq. (7), and show numerically that the incoherent scattering can exceed the single-qubit value by orders of magnitude (Fig. 3). They also argue, on the basis of a numerical scan in Fig. S10, that M-excitation subradiant states require N ≥ 2M qubits, and they illustrate long-lived photon-photon correlations from twilight and subradiant states in the supplemental figures.","tokens_in":20706,"tokens_out":1683,"duration_ms":18688,"significance":"If the central claims hold, the paper provides a useful conceptual addition to waveguide-QED few-photon scattering: it isolates a mechanism by which collective subradiant two-qubit states produce narrow resonances in the inelastic channel, and it introduces the 'twilight state' classification that connects one-photon and two-photon decay properties. The explicit N=4 analytic eigenvalues, Eqs. (6)–(7), the closed-form scattering kernel, Eq. (9), and the optical-theorem-like identity, Eq. (10), are concrete and checkable results. The paper also ships reproducible numerical spectra and contrasts N=2,3 behavior, which strengthens the presentation. However, the general N ≥ 2M threshold is not proven analytically, and the Markovian approximation—while stated—is not accompanied by a quantitative validity condition, which matters because the predicted linewidths are of order ϕ²Γ0.","major_comments":[{"comment":"The Markovian replacement of H^(1)(ω) by H^(1)(ω0) is the foundation for the resonance positions and linewidths in Eqs. (7), (8), and (9), yet the paper justifies it only by subwavelength spacing, ϕ << 1. For the N=4 subradiant eigenvalues the relevant detunings from ω0 are of order ϕΓ0, so the neglected frequency dependence of the phase (ω/c)|z_i−z_j| is of order (ϕΓ0)(ϕ/ω0) = ϕ²Γ0/ω0. This correction is comparable to the claimed linewidth itself when Γ0/ω0 is not much smaller than unity. In strongly coupled waveguide-QED implementations Γ0/ω0 can approach 0.1, at which point the predicted linewidths, resonance positions, and the order-of-magnitude enhancements in Fig. 3 are not protected. The authors should state the required weak-coupling condition Γ0/ω0 ≪ 1 (or an equivalent bound) alongside the subwavelength condition, and should specify whether the figures and the N ≥ 2M scan are within such a regime.","section":"Main text, after Eq. (3); Supplemental S4.B"},{"comment":"The general claim that M-excitation subradiant states exist only for N ≥ 2M is inferred from a numerical scan of minimal decay rates (Fig. S10) and stated as a general fact without an analytic argument. This is load-bearing for the paper's advertised message. The N=4 double-excitation case is supported by the explicit counting argument in the main text (conditions (i)–(iii) with z_j ≡ 0), but that argument does not obviously extend to M>2. The authors should either provide a proof of the N=2M counting result for general M or clearly label it as a conjecture supported by numerics for the studied parameters.","section":"Main text, 'Multi-excited states' and Fig. S10"},{"comment":"The central quantitative claim—that the subradiant resonances survive destructive interference and exceed single-qubit scattering by orders of magnitude—is shown in Fig. 3 for a specific set of parameters (χ=10^4Γ0, ϕ=0.1, N=4). Since Eq. (10) is derived in the χ→∞ two-level limit, the authors should state how sensitive the enhancement is to finite χ and to the choice of ϕ. In particular, the figure appears to be taken at χ=10^4Γ0; the text says the two-level limit is χ→∞, but does not quantify how large χ must be for Eq. (10) to be quantitatively reliable. A brief statement of the finite-χ correction would make the comparison to experiment more meaningful.","section":"Main text, Fig. 3 and Eq. (10)"},{"comment":"The long-lived photon-photon correlation claim relies on the kinetic cascade model in S2, which assumes that the single-photon decay channels of the double-excited state are Markovian and independent. The paper notes that the first photon is emitted quickly and the second slowly for twilight states; this is consistent with the kinetic equations, but the equations (S14)–(S20) appear to treat the occupation of the single-excited state as a classical probability without including the interference between multiple decay channels. The authors should state whether the cascade model is exact in the Markovian limit or is a simplifying approximation, since the claim of long-lived correlations is used to motivate quantum-information applications.","section":"Supplemental S4.C and S6"}],"minor_comments":[{"comment":"The abstract states that 'the N-excitation subradiant states can be engineered only if the number of qubits exceeds 2N,' but the paper demonstrates this only for M=2 analytically and for M>2 numerically. Suggest rewording to 'the numerics indicate' or adding a proof.","section":"Abstract and Introduction"},{"comment":"The symbol Γ0 is used for the single-qubit decay rate, but in Eq. (5) the decay rate is written as Γ1; the notation Γν21, Γµ10 in the supplement is clearer. Consider defining these once in the main text to avoid confusion.","section":"Main text, Eq. (3) and after"},{"comment":"The caption uses 'Scat.(4)/Scat.(1)' in panel (f) but does not define the normalization of the single-qubit scattering; the text refers to 'Scat.(1)' only after Eq. (11). Please define all abbreviations in the caption.","section":"Fig. 3 caption"},{"comment":"The plot shows decay rates as a function of M and N, but the color scale and the exact definition of the 'most subradiant' state (e.g., which eigenvalue is chosen when several have similar rates) are not specified. This is important for reproducing the N=2M threshold line.","section":"Supplementary Fig. S10"},{"comment":"The identity −2Γ0 Re TrQ = |TrQ|² is stated without derivation in the main text and derived in the supplement; please include the derivation in the main text or point explicitly to Eq. (S47), since this identity is used to argue that the scattering is not suppressed by destructive interference.","section":"Eq. (10) and optical-theorem identity"},{"comment":"Minor grammatical issues: 'amplidute' in the text near Eq. (4), 'the photons can be traced out' is vague, and 'the other parameters in the second and fourth panels' in Fig. S9 is unclear. These do not affect the science but should be corrected.","section":"Language and typos"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"You should know this one is a genuine advance in waveguide QED, not a repackaging. The paper shows that double-excited subradiant states of a qubit array act as sharp resonances in two-photon inelastic scattering, and that these resonances survive the destructive interference that usually suppresses the process. That's a new mechanism, and the analytic treatment is convincing.\n\nThe matrix Green's function formulation is compact and genuinely useful. The explicit subradiant double-excited states for N=4 (Eq. 6) and their eigenvalues (Eq. 7) are derived cleanly, and the twilight-state classification (bright times subradiant) organizes the spectrum well. The numerical comparison with N=2,3 in the supplement shows that the enhancement appears exactly when the subradiant states appear. They also properly distinguish their small-N subradiant states from the fermionic ansatz of Zhang and Mølmer, which only works at larger N.\n\nSoft spots are real but not fatal. The Markovian replacement H^(1)(ω)→H^(1)(ω0) is justified by subwavelength spacing, but the linewidths you care about are order ϕ²Γ0. The phase error from detuning near resonance is order ϕ²Γ0/ω0, i.e. a relative correction Γ0/ω0 to the linewidth. Subwavelength spacing does not control Γ0/ω0. For strongly coupled implementations where Γ0/ω0 approaches 0.1, the specific linewidths and the enhancement factors in Fig. 3 are only good to ~10%, and the justification should be stated as requiring both ϕ≪1 and Γ0/ω0≪1. This is a scope condition, not an inconsistency.\n\nThe general claim that M-excitation subradiant states require N>2M rests on a numerical scan (Fig. S10) together with a clean counting argument for M=2. That's plausible but the wording 'can be engineered only if' is stronger than the evidence. Soften it.\n\nThe final quantum-information storage remark is a one-sentence overreach, but it's not load-bearing.\n\nOverall: central result holds, derivations are coherent, data support the claims. Send it to review. I'd cite it next time I work on few-photon scattering.","headline":"A genuine new mechanism in waveguide QED with a clean analytic treatment; the soft spots are scope conditions, not fatal flaws.","tokens_in":21267,"tokens_out":5340,"would_cite":true,"duration_ms":50448,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["42.50.Nn","42.50.Ct"],"model":"deepseek-v4-flash","headline":"Subradiant double-excited states create sharp resonances in inelastic photon-pair scattering and can beat single-qubit output by orders of magnitude.","keywords":["subradiant states","photon-pair scattering","inelastic scattering","waveguide quantum electrodynamics","qubit arrays","two-photon correlations","twilight states","superradiance"],"falsifier":"Measure the total forward inelastic scattering of a photon pair in a waveguide coupled to four qubits with spacing $\\phi \\approx 0.1$: the theory predicts sharp peaks at mean pair energies $2(\\omega_0-\\phi\\Gamma_0)$ and $2(\\omega_0-7\\phi\\Gamma_0/3)$ with widths $\\sim\\phi^2\\Gamma_0$ that exceed the single-qubit scattering maximum; if no such peaks appear, or if the scattering stays at or below the two-qubit value in the suppression dip, the central claim fails.","tokens_in":20238,"feed_emoji":"⚛️","tokens_out":10975,"duration_ms":93985,"temperature":0.7,"pith_summary":"This paper seeks to establish that the inelastic scattering of photon pairs in a waveguide-coupled array of two-level qubits is sharply enhanced when the total energy of the incoming pair matches a double-excited subradiant state of the array. It shows that these subradiant resonances survive the destructive interference that normally suppresses inelastic two-photon scattering, so an array of four qubits can scatter photon pairs more strongly than a single qubit by orders of magnitude. It also classifies the double-excited states into superradiant, subradiant, and twilight states, and shows that $M$-excitation subradiant states exist only when the array contains more than $2M$ qubits. The paper argues this opens a route to long-lived photon-photon correlations useful for quantum information storage and processing.","feed_headline":"Subradiant resonances amplify photon-pair scattering in qubit arrays","feed_subtitle":"Matching a photon pair's total energy to a double-excited subradiant state beats the suppression that kills normal two-photon scattering.","key_machinery":"The load-bearing object is the non-Hermitian effective qubit Hamiltonian obtained after tracing out the photons, $H^{(1)}_{ij}(\\omega_0)=\\hbar\\omega_0\\delta_{ij}-i\\hbar\\Gamma_0 e^{i\\omega_0|z_i-z_j|/c}$, evaluated in the Markovian approximation at the qubit frequency. From it the paper builds the two-particle Hamiltonian $H^{(2)}+U$ and the matrix scattering kernel $Q=-i\\chi(1-i\\chi\\Sigma)^{-1}$, which in the Markovian limit simplifies to $Q=i\\chi[(2\\varepsilon-H)/(H+U-2\\varepsilon)]$ and, near a resonance, to $Q_{ij}\\approx 2i\\Gamma_0^2 d_i d_j^*/(\\varepsilon_\\nu-\\varepsilon)$. The radiative-transition amplitudes $d_i$ do double duty: they determine the decay rate $\\Gamma_1=\\Gamma_0\\sum_j|d_j|^2$ in the denominator and the oscillator strength in the numerator, which is why states with small but nonvanishing $d_i$ (subradiant) or out-of-phase $d_i$ (twilight) can still dominate the scattering. The classification of double-excited states as superradiant, twilight, or subradiant by the magnitudes of $\\sum_j|d_j|^2$ and $|\\sum_j d_j|^2$ is the organizational device that makes the resonance structure legible.","core_discovery":"The central discovery is that the two-photon scattering kernel $Q$ is controlled by the eigenstates of the effective two-particle Hamiltonian $H^{(2)}+U$: near a double-excited eigenenergy $\\varepsilon_\\nu$, $Q_{ij}\\approx 2i\\Gamma_0^2 d_i d_j^*/(\\mathrm{Re}\\,\\varepsilon_\\nu - i\\Gamma_0\\sum_j |d_j|^2 - \\varepsilon)$, where $d_i=\\sum_j e^{i\\omega_0 z_j/c}\\Psi_{ij}$ are the radiative-transition amplitudes. The same amplitudes set both the linewidth and the oscillator strength of the two-photon resonance, and different patterns of $d_i$ separate superradiant ($\\sum_j |d_j|^2\\sim N$), twilight (finite $\\sum_j|d_j|^2$ but $|\\sum_j d_j|^2\\ll 1$), and subradiant ($\\sum_j|d_j|^2\\ll 1$) states. For $N=4$ qubits with subwavelength spacing $\\phi$, the two subradiant double-excited states have energies $\\varepsilon_1=\\omega_0-\\phi\\Gamma_0-i\\phi^2\\Gamma_0/2$ and $\\varepsilon_2=\\omega_0-7\\phi\\Gamma_0/3-157i\\phi^2\\Gamma_0/54$, and these resonances appear as sharp peaks in the inelastic forward-scattering spectrum, exceeding the single-qubit result by orders of magnitude. The paper further finds that $M$-excitation subradiant states require more than $2M$ qubits, and that twilight states, formed as products of bright and subradiant single-excitation states, emit one photon quickly and the second after $\\sim 1/(\\phi^2\\Gamma_0)$, giving photon-photon correlations far longer-lived than a single qubit's decay time.","pith_inferences":["A natural extension the authors do not pursue is applying the same compact matrix method to non-periodic or disordered arrays; if subradiant resonances survive disorder, the enhancement would not require precise periodicity.","The triple-resonance condition suggests a practical way to create entangled photon pairs with tunable correlation lifetimes by tuning the array period $\\phi$; this application is implied but not demonstrated in the paper.","Since twilight states have linewidth $\\sim\\Gamma_0$ but lifetime $\\sim 1/(\\phi^2\\Gamma_0)$, they could be easier to address with broad-band pulses than fully subradiant states while still storing correlations; this trade-off is an inference from the paper's spectra.","For $M>2$ photons, the threshold $N>2M$ implies that a six-qubit array should support three-photon subradiant states; observing them would extend the same resonance mechanism to multiphoton scattering, which the paper leaves beyond its scope."],"forward_implications":["For an array of four qubits, the double-excited subradiant resonances appear exactly in the suppression dip where two- and three-qubit arrays show destructive interference, so the enhancement is not washed out by that interference.","At the triple-resonance point $\\omega_1=\\omega_2=\\omega_0-\\phi\\Gamma_0$, the single- and double-excited subradiant resonances coincide and further amplify the inelastic forward scattering.","$M$-excitation subradiant states require more than $2M$ qubits; with $N\\ge 4$, the lowest double-excited subradiant decay rate scales as $\\Gamma_1\\sim \\Gamma_0 \\phi^2/N^3$, making the states optically accessible while long-lived.","Twilight states, being products of bright and subradiant single-excitation states, emit one photon quickly and the second only after a time $\\sim 1/(\\phi^2\\Gamma_0)$, producing photon-photon correlations that outlive the single-qubit decay time.","The subradiant and twilight resonances both appear in inelastic scattering of photon pairs, giving a route to long-lived photon-photon correlations for quantum information storage."],"supporting_citations":[{"why":"Supplies the exact Bethe-wavefunction treatment of two-photon scattering in a waveguide, including the destructive interference that suppresses ordinary inelastic scattering.","marker":"[28]"},{"why":"Provides the concept and prior treatment of subradiant states of qubits coupled to a one-dimensional waveguide that the paper builds upon for double excitations.","marker":"[32]"},{"why":"Gives the theory of subradiant states of a one-dimensional two-level atom chain and the fermionic ansatz that the paper compares with its exact double-excited states.","marker":"[34]"},{"why":"Offers the subradiant dimer excited states and fermionic ansatz used in the supplemental comparison for small and large qubit numbers.","marker":"[36]"},{"why":"Provides the multiphoton scattering formalism in a one-dimensional waveguide that underpins the two-photon scattering amplitude and its interference structure.","marker":"[37]"},{"why":"Establishes the destructive-interference effect that suppresses disallowed two-atom transitions, the background against which the subradiant resonances stand out.","marker":"[42]"},{"why":"Supplies the Green-function method for waveguide-mediated interactions that the paper's approach conceptually resembles and extends to a compact matrix formulation.","marker":"[43]"},{"why":"Provides the prior treatment of biexciton-mediated superradiant photon blockade, used for the large-anharmonicity two-level limit and the structure of the scattering kernel.","marker":"[44]"}],"fun_headline_variants":["Subradiant states boost photon-pair scattering in qubit arrays","Photon pairs scatter strongly via subradiant qubit states","Long-lived correlations from subradiant photon-pair scattering","Subradiant double-excited states enhance inelastic photon scattering"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"Everything hinges on replacing the frequency-dependent photon-mediated coupling by its value at the qubit resonance frequency $\\omega_0$, which is valid only for subwavelength spacing; the predicted resonance positions, lifetimes, and scaling are all computed inside that Markovian approximation.","fun_headline_variants_meta":{"raw":{"variants":["Subradiant states boost photon-pair scattering in qubit arrays","Photon pairs scatter strongly via subradiant qubit states","Long-lived correlations from subradiant photon-pair scattering","Subradiant double-excited states enhance inelastic photon scattering"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000811,"raw_usage":{"total_tokens":3622,"prompt_tokens":1077,"completion_tokens":2545,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":693,"completion_tokens_details":{"reasoning_tokens":2474}},"tokens_in":693,"tokens_out":2545,"duration_ms":19078,"temperature":1.0,"reasoning_tokens":2474,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:32:07.825942+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the total forward inelastic scattering of a photon pair in a waveguide coupled to four qubits with spacing $\\phi \\approx 0.1$: the theory predicts sharp peaks at mean pair energies $2(\\omega_0-\\phi\\Gamma_0)$ and $2(\\omega_0-7\\phi\\Gamma_0/3)$ with widths $\\sim\\phi^2\\Gamma_0$ that exceed the single-qubit scattering maximum; if no such peaks appear, or if the scattering stays at or below the two-qubit value in the suppression dip, the central claim fails.","supporting_citations":[{"cited_title":"Exact Dicke superradiance theory: Bethe wavefunctions in the discrete atom model,","cited_arxiv_id":null,"evidence_quote":"Supplies the exact Bethe-wavefunction treatment of two-photon scattering in a waveguide, including the destructive interference that suppresses ordinary inelastic scattering."},{"cited_title":"Subradiant states of quan- tum bits coupled to a one-dimensional waveguide,","cited_arxiv_id":null,"evidence_quote":"Provides the concept and prior treatment of subradiant states of qubits coupled to a one-dimensional waveguide that the paper builds upon for double excitations."},{"cited_title":"Multiphoton scattering in a one-dimensional waveguide with resonant atoms,","cited_arxiv_id":null,"evidence_quote":"Provides the multiphoton scattering formalism in a one-dimensional waveguide that underpins the two-photon scattering amplitude and its interference structure."},{"cited_title":"Persistent quantum beats and long-distance entanglement from waveguide-mediated interactions,","cited_arxiv_id":null,"evidence_quote":"Supplies the Green-function method for waveguide-mediated interactions that the paper's approach conceptually resembles and extends to a compact matrix formulation."},{"cited_title":"Biexciton- mediated superradiant photon blockade,","cited_arxiv_id":null,"evidence_quote":"Provides the prior treatment of biexciton-mediated superradiant photon blockade, used for the large-anharmonicity two-level limit and the structure of the scattering kernel."}],"review_version":1}