{"id":"ee10b9ab-0bf6-4661-92fd-7e89aa0ce01c","arxiv_id":"2411.14065","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"In a two-giant-atom waveguide QED system, the number of bound states in the continuum determines whether the atoms exhibit Rabi oscillations (two BICs) or fractional population dynamics (one BIC).","lead":"Two giant atoms in a coupled resonator waveguide show two distinct behaviors depending on how many bound states in the continuum (BICs) exist: persistent Rabi oscillations for two BICs, and incomplete, fractional decay for one BIC. Because BICs prevent excitations from fully leaking away, the result offers a practical way to suppress decoherence and engineer interactions in superconducting quantum circuits.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The time-convolutionless reduction from exact memory equations (B7)-(B8) to Eq. (15) is asserted, not derived; the two-BIC/one-BIC dichotomy rests on this uncontrolled approximation and is tested only at g/ξ=0.1.","rationale":"The reader's weakest_assumption identifies the same load-bearing point: the reduction from the exact convolution equations to the time-local Eq. (15) is not shown, and its validity is only numerically checked at one coupling strength. I agree that this is the most vulnerable link in the argument. The paper does provide independent support: the exact BIC spectrum from Eq. (13) is derived from the Schrödinger equation, and the full-Schrödinger numerics at g/ξ=0.1 agree with Eq. (15). That agreement is real evidence, and it justifies keeping the verdict conditional rather than rejecting. However, because the claim is stated generally ('when the system supports two BICs, Rabi oscillations dominate'), the absence of a derived reduction or of coupling-strength scans leaves open the possibility that the dichotomy is a weak-coupling artifact. The proposed test directly probes the uncontrolled step. No change to the reader's CONDITIONAL verdict is needed.","tokens_in":13265,"tokens_out":13903,"duration_ms":137842,"concrete_test":"Numerically solve the exact integro-differential equations (B7)-(B8) with a Volterra or delay-differential solver for the N=6 (two-BIC) and N=8, Δ=2 (one-BIC) configurations, and compare |α1(t)|^2 and |α2(t)|^2 with the time-convolutionless solutions of Eq. (15) at g/ξ = 0.1, 0.2, and 0.5. If the deviations exceed a few percent, or if the one-BIC case shows additional damping not captured by Eq. (15), then the reduction is not controlled and the central claim needs a quantitative parameter-range qualification.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central dynamical predictions rest on Eq. (15), a time-local equation for α(t). The exact calculation in Appendix B yields integro-differential equations (B7)-(B8), in which the right-hand sides are convolutions of α1(τ) and α2(τ) with Bessel-function kernels. The text states that Eq. (15) follows 'under the Weisskopf-Wigner approximation but beyond the Markovian approximation,' yet Appendix B never performs this reduction; it stops at (B7)-(B8). The only evident route to the time-local form is to replace α(τ) by α(t) inside the memory kernels, an approximation that is uncontrolled and not error-bounded. Its validity is checked only at g/ξ = 0.1 by agreement with full Schrödinger evolution. If the replacement fails at stronger couplings, the quantitative Rabi period, the fractional population values, and possibly the dichotomy itself would be artifacts of the approximation rather than robust consequences of the BIC count. In addition, Sec. II.D extends Ref. [41]'s criterion for a time-independent generator (number of BICs equals number of zero eigenvalues of M) to the time-dependent M(t) of Eq. (15) without proof; instantaneous eigenvalues of a time-dependent generator do not in general determine the asymptotic decay of the propagator.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript studies two giant atoms in a braided configuration coupled to a coupled-resonator waveguide, working in the single-excitation subspace. The authors identify bound states in the continuum (BICs) from the eigenenergy equation (13), and they derive, in Appendix B, integro-differential equations (B7)-(B8) for the atomic amplitudes. The main text then uses a time-local equation (15) for the two atomic amplitudes and connects the number of BICs to the eigenvalues of the time-dependent matrix M(t). The reported dynamical regimes are: two BICs give persistent Rabi oscillations between the two giant atoms with period fixed by the BIC energy splitting, and one BIC gives fractional stationary populations. The analytical predictions are compared with direct Schr\\\"odinger evolution at g/xi = 0.1, and the photonic distributions are analysed to support the association of the long-time state with the BIC.","tokens_in":1653,"tokens_out":1769,"duration_ms":95112,"significance":"If the central claim is correct, the paper provides a useful example in which a structured environment suppresses its own dissipative action through BICs, leading to protected Rabi oscillations or fractional population trapping. The work has clear strengths: the eigenenergy equation (13) is derived in detail, the exact memory equations (B7)-(B8) are obtained from the Schr\\\"odinger equation, and the approximate dynamics are checked against full numerical evolution at g/xi = 0.1, including real-space photonic distributions. However, the load-bearing step from the exact convolution equations to the time-local equation (15) is not shown, and the extension of a time-independent BIC-eigenvalue criterion to a time-dependent generator is asserted rather than proved. These gaps currently limit the claim to the weakly coupled, numerically tested regime.","major_comments":[{"comment":"The central dynamical equation is not derived. Equations (B7)-(B8) are exact integro-differential equations whose right-hand sides contain convolutions of alpha1(tau) and alpha2(tau) with Bessel-function kernels. The main text states that these reduce to the time-local equation (15) 'under the Weisskopf-Wigner approximation but beyond the Markovian approximation', but Appendix B ends at (B7)-(B8) without performing that reduction. Since the eigenvalues in Fig. 2 and the populations in Figs. 3-4 all rely on Eq. (15), the authors need to specify the reduction explicitly, for example as a second-order time-convolutionless approximation, and justify it with an error estimate or a consistency condition. The agreement with the full Schr\\\"odinger evolution at g/xi = 0.1 is encouraging, but it tests only one parameter point.","section":"Appendix B and Eq. (15)"},{"comment":"The criterion of Ref. [41] applies to a time-independent generator M whose eigenvalues have non-positive imaginary parts. Here M(t) is explicitly time-dependent, and the text asserts without proof that the number of BICs still equals the number of zero eigenvalues of M(t). For a time-dependent generator, instantaneous eigenvalues do not in general determine the asymptotic decay or the existence of invariant subspaces of the propagator; additional conditions such as adiabaticity, a commuting-family structure, or an explicit solution of the time-ordered exponential are needed. Please provide such a justification, or verify directly from Eq. (15) that the long-time propagator has the claimed protected subspace in the one- and two-BIC cases.","section":"Sec. II.D"},{"comment":"All numerical checks are performed at g/xi = 0.1. The experimental parameters quoted in Sec. IV are xi/(2pi) = 50 MHz and g/(2pi) = 124.6 MHz from Refs. [58] and [60], which give g/xi approximately 2.5 if these numbers are combined; the validity of the uncontrolled reduction behind Eq. (15) in that regime is therefore not established. The authors should either test stronger couplings, for example g/xi = 0.5 and 1, and show that the Rabi-oscillation and fractional-population behavior persists, or explicitly restrict the conclusions to the weak-coupling regime.","section":"Secs. III-IV and parameter regime"},{"comment":"The fractional-population claim is anchored by the numerical equality |alpha1(infinity)|^2 = |alpha2(infinity)|^2 = |<psi(0)|E_BIC><E_BIC|psi(0)>|^2, which is stated without derivation or error estimate. This equality is a consistency check between the approximate dynamics and the eigenstate calculation, not a proof. Given that the central one-BIC/two-BIC dichotomy rests on this relation, the authors should provide a formal argument, such as a decomposition into bound and scattering components with a demonstration that the scattering part vanishes at long times, or should show convergence of the equality with increasing system size and evolution time.","section":"Eq. (23) and Sec. III.B"}],"minor_comments":[{"comment":"The title contains a typo: 'bound st ates' should read 'bound states'.","section":"Title"},{"comment":"The caption of Fig. 3 states that solid lines are analytical and dashed lines are numerical, whereas the text for Fig. 4 states the opposite convention; please make the convention consistent and correct both captions.","section":"Figs. 3 and 4"},{"comment":"For N1 = N2 = 6 and Delta = 2 or 4, Table I gives E1 = E2 = 0, so the two BICs are degenerate; the formula T = 2pi/(E1 - E2) used for the Rabi period in Sec. III.A does not apply to the degenerate case and this should be discussed.","section":"Table I"},{"comment":"The horizontal axis in Fig. 2 is evolution time but the units are not stated; please specify that time is in units of 1/xi and comment on whether t = 100/xi is sufficient for the eigenvalues to reach their asymptotic values.","section":"Fig. 2"},{"comment":"The phrase 'the Weisskopf-Wigner approximation also allows us to derive the photonic dynamics' is misleading: Eqs. (B9)-(B11) follow exactly from the Schr\\\"odinger equation once the vacuum initial condition is imposed, with no approximation needed; please rephrase.","section":"Appendix B, Eq. (B9)"}],"recommendation":"major_revision","confidential_remarks":"The manuscript has an interesting and timely message, and the exact equations in Appendix B are useful. The main obstacle is the unshown reduction from (B7)-(B8) to (15) and the unproved time-dependent eigenvalue criterion in Sec. II.D; both are load-bearing for the claimed dichotomy. I believe these can be fixed with a proper derivation and additional numerical tests, so I recommend major revision rather than rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear [Colleague],\n\nThe punchline: the paper's central dichotomy is correct, and it is more robust than the derivation suggests. Two BICs give persistent Rabi oscillations; one BIC gives fractional population. I expected to find the TCL gap fatal, but on reading, the asymptotic behavior follows from exact BIC eigenstates, so the qualitative claim stands.\n\nWhat's new: a concrete two-giant-atom braided setup in a CRW where the number of BICs—two vs one—changes the reduced dynamics. The paper shows via spectral analysis and real-space photonic distributions that the long-time Rabi oscillation is exactly oscillation between the two BIC components, and the single-BIC steady state is just the BIC projection. That's a clean result and a useful design rule. The derivations in the appendices follow standard methods, and the numerical check at g/ξ=0.1 matches the approximate equations.\n\nThe soft spots, in order:\n\n1. The reduction from the exact convolution equations (B7)-(B8) to the time-local Eq. (15) is asserted, not shown. 'Weisskopf-Wigner beyond Markovian' is doing a lot of work. For the long-time behavior this doesn't matter, because the BIC components are exact. But the transient shape, the Rabi period, and the fractional values do depend on the approximation. At stronger coupling the transient could be wrong. The authors should either supply the TCL derivation or state that the asymptotic claims are exact and the approximation is only for the transient.\n\n2. Sec. II.D extends Longhi's criterion to a time-dependent M(t) without proof. Instantaneous eigenvalues of a time-dependent generator don't generally control asymptotic decay. The agreement with the exact BIC count here is a consistency check, not a derivation. Should be flagged and proved or reframed.\n\n3. Numerical details are missing: no chain length, no convergence data, no code/data. For a paper making quantitative predictions, that's a reproducibility gap. Minor, but easy to fix.\n\nI don't think the paper overclaims. The phrase 'challenge the conventional notion' is a bit grand, but the underlying phenomenon is real. The citation pattern is fine; self-citations to [56,57] are relevant.\n\nWho it's for: anyone in waveguide QED interested in BIC-protected dynamics or non-Markovian effects. It's a moderate step forward, not a revolution, but it deserves a serious referee.\n\nMy recommendation: send to peer review, conditional on the authors clarifying the approximation, proving or reframing the eigenvalue criterion, and providing numerical details. I'd accept it as a solid but incremental contribution.\n\nBest.","headline":"The BIC-count dichotomy is real and robust, but the time-local derivation is asserted, not derived, and the numerics lack details.","tokens_in":14094,"tokens_out":4593,"would_cite":true,"duration_ms":46023,"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":"In a braided two-giant-atom setup coupled to a resonator waveguide, the number of bound states in the continuum dictates the dynamics: two BICs give persistent Rabi oscillations, one BIC gives fractional population decay.","keywords":["giant atoms","coupled resonator waveguide","bound states in the continuum","Rabi oscillations","fractional population","non-Markovian dynamics","waveguide quantum electrodynamics","open quantum systems"],"falsifier":"Solve the exact integro-differential equations (B7)–(B8) directly, without the reduction to the time-local equation, at couplings $g/\\xi \\ge 0.3$; if the two-BIC case shows decaying oscillation amplitude or the one-BIC plateau drifts from $|\\langle \\psi(0)|E_{\\mathrm{BIC}}\\rangle\\langle E_{\\mathrm{BIC}}|\\psi(0)\\rangle|^2$, the BIC classification is an artifact of the weak-coupling reduction. A complementary experiment: in a superconducting coupled-resonator array with inter-site hopping $\\xi/(2\\pi) = 50$ MHz, prepare one pair of giant atoms in the $N=6$ braided geometry and one in the $N=8$, $\\Delta=2$ geometry, and measure the long-time resonator populations; the prediction is persistent shuttling of the photon in the first case and a frozen two-site pattern in the second.","tokens_in":13024,"feed_emoji":"⚛️","tokens_out":22899,"duration_ms":176229,"temperature":0.7,"pith_summary":"The paper claims that the fate of excitation in two 'giant' atoms — emitters each coupled to a resonator waveguide at two separated points, braided so their coupling regions overlap — is decided entirely by the number of bound states in the continuum (BICs) the joint system supports. When the configuration supports two BICs, the atomic excitation oscillates back and forth between the two atoms without decaying into the waveguide, with a period set by the energy splitting of the BICs. When only one BIC exists, the atoms decay only partially and converge to equal nonzero steady-state populations. The authors tie both behaviours to a single criterion, whether the imaginary parts of the eigenvalues of their reduced dynamical matrix vanish, and confirm it by watching where the photons go in the waveguide. If the claim holds, it overturns the default expectation that a reservoir always drives a quantum system to complete decay.","feed_headline":"Two bound states make a lossy waveguide stop leaking atomic energy","feed_subtitle":"Trapped photon modes decide whether atomic excitation oscillates forever or settles at a fraction","key_machinery":"The carrier of the argument is the reduced single-excitation dynamical equation $i\\,d\\vec{\\alpha}(t)/dt = M(t)\\vec{\\alpha}(t)$, whose entries are memory integrals over Bessel functions: the diagonal terms $A_i(t)$ describe each atom's free oscillation plus the backaction of the waveguide, and the off-diagonal term $B(t)$ is the waveguide-mediated coupling between the atoms. The number of BICs is read off from the long-time imaginary parts of the eigenvalues $\\lambda(t)$ of $M(t)$: an eigenvalue with vanishing imaginary part flags a channel that does not dissipate, hence one BIC. That eigenvalue criterion, adapted from the Markovian case, connects the transcendental eigenenergy equation of the full system to the reduced atomic dynamics, and it converts the geometric parameters — atom size $N = n_2 - n_1 = m_2 - m_1$ and relative distance $\\Delta = m_1 - n_1$ — into a prediction about oscillation versus fractional decay.","core_discovery":"A coupled resonator waveguide, despite hosting a continuous photonic band that normally acts as a dissipative reservoir, can harbour bound states inside that band — modes with real-space localization whose coupling to the emitters is cancelled by interference between the two contact points of each giant atom. The paper's central claim is that the number of such BICs fixes the reduced dynamics of the two braided atoms: with two BICs, an initial excitation of the first atom performs undamped Rabi oscillations between the atoms, and the waveguide photon periodically moves between the two waveguide regions the atoms touch; with one BIC, the excitation decays only partially, and both atoms settle at identical finite populations set by the projection of the initial state onto the BIC eigenstate. The same BIC count follows from two independent routes — the eigenenergy equation of the full atom–waveguide system and the long-time imaginary parts of the eigenvalues of the reduced $2\\times 2$ dynamical matrix $M(t)$ — so the paper establishes a consistent map from the global spectrum to the open-system dynamics.","pith_inferences":["Because the same waveguide can be switched between full decay, fractional survival, and undamped oscillation by choosing atom size and relative distance, the braided geometry is a tunable resource: a testable extension is to encode one qubit in the two-BIC protected subspace of a superconducting circuit array and compare its coherence time with that of an unprotected configuration in the same devi","The eigenvalue criterion — vanishing imaginary part of a memory-matrix eigenvalue equals one BIC — looks like a special case of a more general counting rule for structured reservoirs; if so, the same argument should predict oscillation versus decay for three or more giant atoms, or for emitters in other band-gap environments such as photonic crystals.","Because the effective equations are checked numerically only at the weak coupling $g = 0.1\\xi$, a natural next test is to solve the exact delay equations at stronger coupling (say $g/\\xi \\sim 0.5$) and see whether the Rabi period still tracks the BIC splitting and the fractional plateau still tracks the BIC projection of the initial state."],"forward_implications":["Two BICs (for example atom size $N=6$ at any relative distance $\\Delta$) turn the braided pair into a closed two-level system: the excitation oscillates indefinitely between the atoms with period $T = 2\\pi/(E_1 - E_2)$ set by the BIC energies, and nothing leaks into the waveguide.","One BIC (for example $N=8$ with even $\\Delta$) makes the atoms reach a shared fractional steady state: both atoms keep the same nonzero population $|\\langle \\psi(0)|E_{\\mathrm{BIC}}\\rangle\\langle E_{\\mathrm{BIC}}|\\psi(0)\\rangle|^2$ at infinite time, with the surviving photon frozen in a symmetric profile around the two atoms.","The environment's role is not fixed: the same waveguide that dissipates excitation when no BIC is present protects a decoherence-free subspace when BICs are present, so dissipation can be suppressed by choosing the atom–waveguide geometry.","The waveguide's photonic distribution tracks the atomic dynamics — oscillating between the two atom-covered regions in the two-BIC case and settling into the localized BIC pattern in the one-BIC case — giving an observable signature of the mechanism."],"supporting_citations":[{"why":"Supplies the braided two-coupling-point giant atom model and the decoherence-free interaction that the two-atom braided configuration builds on.","marker":"[8]"},{"why":"Supplies the non-Markovian time-delay treatment of giant atoms underlying the derivation of the dynamical equations.","marker":"[23]"},{"why":"Supplies the criterion that the number of BICs equals the number of zero eigenvalues of the effective matrix, extended here to the time-dependent case.","marker":"[41]"},{"why":"Supplies the notion of fractional population used to characterize the single-BIC steady state.","marker":"[47]"},{"why":"Provides the coupled-resonator-waveguide Hamiltonian with nearest-neighbour hopping that forms the structured environment.","marker":"[52]"},{"why":"The authors' earlier demonstration that a single giant atom in a coupled resonator waveguide can host a BIC.","marker":"[56]"},{"why":"Establishes the single-giant-atom BIC condition (atom size $N = 4m+2$) against which the two-atom braided configurations are classified.","marker":"[57]"}],"fun_headline_variants":["Bound states in waveguide control atom dynamics","Two BICs give Rabi oscillations, one gives partial decay","Waveguide BICs stop dissipation, enable Rabi or fractional states","Giant atoms in waveguide: BICs dictate oscillation vs decay"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole classification rests on the assumption that the simpler time-local equation the authors derive for the two atoms captures everything the exact memory-laden delay equations do, and the paper checks this agreement only at one weak-coupling value, $g = 0.1\\xi$.","fun_headline_variants_meta":{"raw":{"variants":["Bound states in waveguide control atom dynamics","Two BICs give Rabi oscillations, one gives partial decay","Waveguide BICs stop dissipation, enable Rabi or fractional states","Giant atoms in waveguide: BICs dictate oscillation vs decay"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000249,"raw_usage":{"total_tokens":1522,"prompt_tokens":891,"completion_tokens":631,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":507,"completion_tokens_details":{"reasoning_tokens":564}},"tokens_in":507,"tokens_out":631,"duration_ms":6000,"temperature":1.0,"reasoning_tokens":564,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T15:35:14.416585+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Solve the exact integro-differential equations (B7)–(B8) directly, without the reduction to the time-local equation, at couplings $g/\\xi \\ge 0.3$; if the two-BIC case shows decaying oscillation amplitude or the one-BIC plateau drifts from $|\\langle \\psi(0)|E_{\\mathrm{BIC}}\\rangle\\langle E_{\\mathrm{BIC}}|\\psi(0)\\rangle|^2$, the BIC classification is an artifact of the weak-coupling reduction. A complementary experiment: in a superconducting coupled-resonator array with inter-site hopping $\\xi/(2\\pi) = 50$ MHz, prepare one pair of giant atoms in the $N=6$ braided geometry and one in the $N=8$, $\\Delta=2$ geometry, and measure the long-time resonator populations; the prediction is persistent shuttling of the photon in the first case and a frozen two-site pattern in the second.","supporting_citations":[{"cited_title":"Lalumi` ere, B","cited_arxiv_id":null,"evidence_quote":"Supplies the non-Markovian time-delay treatment of giant atoms underlying the derivation of the dynamical equations."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the notion of fractional population used to characterize the single-BIC steady state."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes the single-giant-atom BIC condition (atom size $N = 4m+2$) against which the two-atom braided configurations are classified."}],"review_version":1}