{"id":"66188e75-fc8c-4c7b-919b-64fbae552356","arxiv_id":"2607.17587","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":7,"one_line_summary":"In a disordered Tavis-Cummings model, the pump-probe line shape at the dark-state energy evolves from derivative-like to absorptive as polaritons relax into dark states, and the amplitude scales as 1/N.","lead":"This paper simulates pump-probe spectra of molecular polaritons when molecule energies are disordered, focusing on the dark-state energy region. It predicts the line shape changes from a derivative-like feature into a simple absorptive peak as polaritons relax into dark states, which could act as a fingerprint of polaritonic response.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Stated rate parameters contradict the claimed 50/50 LP-to-DS/GS branching, and this branching controls the late-time line-shape sign.","rationale":"The reader identified the hand-picked Markovian rate model as the weakest assumption. My stress-test agrees with that area but sharpens it into a concrete internal inconsistency: the parameter values reported for the rate model do not produce the 50/50 branching that the text says they do. This is not a criticism of the physics consensus; the response-function framework is standard and the simplified line-shape model in Appendix A is a valid derivation. The issue is that the central temporal evolution—derivative-like at early times becoming a single positive absorptive peak at late times—depends on the relative populations of DS and GS after LP decay. If the true branching is ~99% into DS, the relative weights of GSB_DS versus the positive ESADS/ESA2DS/SEDS pathways shift, and the net sign of the late-time peak is not guaranteed by the stated equations. Since no code or data are provided, the reader cannot tell whether the figures correspond to the stated rates or to the stated branching. This is an addressable, concrete problem rather than a fatal flaw, so the appropriate verdict remains CONDITIONAL; my read therefore does not change the reader's verdict.","tokens_in":14205,"tokens_out":18603,"duration_ms":164541,"concrete_test":"Set N=10 and evaluate Eqs. 9-14 with the stated rates. Compute P_DS_total(∞)/P_LP(0) and P_GS(∞)/P_LP(0); verify they equal 0.5/0.5. Then rerun the PP calculation of Fig. 3b at T=0 and T=400 fs with the corrected equal-branching rates (k_GS=(N-1)k_DS=0.005 fs^-1, total k_LP=0.01 fs^-1) and compare the sign and shape of the late-time spectrum. If the T=400 fs signal remains a single positive peak, the inconsistency is a typo; if it becomes negative or derivative-like, the fingerprint claim depends on the erroneous rates.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The temporal line-shape transition is driven entirely by the population rate model (Eqs. 9-14). The text states (Sec. III.A) that the rates are chosen so that 'the LP relaxes ... 50% ... into the ground state and 50% ... into the manifold of DSs.' But substituting the stated values 1/k_LP=100 fs, 1/k_GS=50 fs, 1/k_DS=50/(N-1) fs into the equations does not give this branching. If Eq. 9 is read literally, the total LP decay rate is k_LP+(N-1)k_DS = 0.01 + (N-1)^2/50 fs^-1; for N=10 this is 1.63 fs^-1 (lifetime ~0.6 fs), and the asymptotic DS-manifold fraction is (N-1)k_DS/(k_LP+(N-1)k_DS) ≈ 0.994, while the GS fraction is ≈0.006 (or ≈0.012 if the numerator in Eq. 14 is meant to be k_GS). Even using the paper's stated total rate k_LP=0.01 fs^-1, equal 50/50 branching would require k_GS=(N-1)k_DS=0.005 fs^-1, i.e. 1/k_GS=200 fs and 1/k_DS=200/(N-1) fs, not the reported 50 fs and 50/(N-1) fs. Because the late-time 'absorptive' line shape is the sum of the negative GSB_DS pathway and positive ESADS/ESA2DS/SEDS pathways, the net sign and the derivative-to-absorptive transition depend on the actual branching fractions. The paper provides no code/data to disambiguate whether the figures were generated with the stated rates or with the stated 50/50 branching; either way, the model as written is internally inconsistent at exactly the point that controls the central claim.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript extends the authors' previous disorder-free Tavis-Cummings (TC) pump-probe study by adding on-site energetic disorder, which gives nominally dark states (DS) a small photonic weight so they can be probed. Using third-order response functions and double-sided Feynman diagrams, it computes the transient absorption signal near the DS energy after pumping the lower polariton. A Markovian rate-equation model for LP relaxation into the ground state and the DS manifold is used to follow the delay-time dependence. The authors report a transition from a derivative-like to an absorptive line shape with delay, analyze the dependence on disorder width, and claim that both early- and late-time DS signals scale as O(N^{-1}) for large N, comparable to the LP/UP signals.","tokens_in":14719,"tokens_out":8392,"duration_ms":74170,"significance":"If substantiated, the predicted derivative-to-absorptive transition at the dark-state energy would be a useful spectroscopic fingerprint of dark-state participation in polariton relaxation, and the O(N^{-1}) scaling is relevant for experimental feasibility in large ensembles. The paper uses standard response-function formalism, gives an explicit analytic line-shape decomposition in Appendix A, and assigns pathway/scaling contributions in Fig. 2. However, the central quantitative results depend on the kinetic relaxation model and on numerical scalings that are not fully documented; the parameter inconsistency in the rate equations is a load-bearing issue that must be corrected.","major_comments":[{"comment":"The stated rate parameters do not produce the asserted 50/50 branching. With 1/k_LP=100 fs, 1/k_GS=50 fs, 1/k_DS=50/(N-1) fs, one has k_DS=(N-1)/50 fs^-1. For N=10, (N-1)k_DS=81/50=1.62 fs^-1 while k_GS=0.02 fs^-1, so about 99% of the LP decays into the DS manifold and only about 1% into the ground state, not 50/50. In addition, the text states k_LP=k_GS+(N-1)k_DS, which makes Eq. (9) double-count the DS rate, and Eq. (14) uses k_LP instead of k_GS in the numerator. Since the late-time 'absorptive' line shape is the net of positive and negative pathway classes whose weights are set by these branching fractions, the central time-evolution result as written is not reproducible. Please correct the equations/notation and either adjust the parameters to match the 50/50 statement or recompute the spectra for the actual branching.","section":"II.B and III.A, Eqs. (9)-(14)"},{"comment":"The O(N^{-1}) scaling claim for the DS line shape rests on the sentence 'we have numerically checked that mu^2_LP->DLP scales with O(N^{-1}) and mu^2_DS->2DS scales with O(N^{-2})' and on Fig. 5a, which shows no error bars, fit residuals, or convergence in N or in the number of disorder realizations. Moreover, the text states that every pathway class except SE_DS contributes O(1); it is then not explained how the net late-time 'trivial' line shape, observed to scale as O(N^{-1}), arises unless there is an O(N^{-1}) cancellation among O(1) contributions. The LP/UP argument in Appendix A is explicit, but the DS case is not. Please provide the scaling data/fits and an explicit demonstration of the net cancellation, or amend the argument.","section":"III.B, Fig. 5 and surrounding text"},{"comment":"The derivative-to-absorptive transition is computed from a Markovian parallel-decay rate-equation model with phenomenological rates and with all coherences during T neglected. The abstract states this transition as a general consequence of 'relaxation to dark states and disorder.' The claim should be explicitly qualified as a property of this kinetic model, and the sensitivity of the transition to the branching ratio and to the Markovian assumption should be discussed or tested. This is distinct from the parameter typo above; even with the intended 50/50 rates, different relaxation mechanisms could change the timing or even the sign of the late-time line shape.","section":"II.B, Eqs. (9)-(14), and Abstract"}],"minor_comments":[{"comment":"The solutions are written as dP_DSi(T) and dP_GS(T), but they should be the populations P_DSi(T) and P_GS(T); the notation is confusing.","section":"Eqs. (13)-(14)"},{"comment":"The horizontal-axis label appears as 'inverse number of molecules N 1'; it should be 'N^{-1}'.","section":"Fig. 5a"},{"comment":"No error bars or convergence checks are reported for the 1000-realization disorder average. Reporting the standard error or showing convergence with realization count would make the numerical claims more robust.","section":"III.A, disorder average"},{"comment":"The approximation in Eq. (A6) is valid for omega*delta << sigma^2; the range of validity should be stated more explicitly, especially because the derivative-like extrema are later quoted as ±sigma.","section":"Appendix A"},{"comment":"The data are available only 'upon request.' For reproducibility of the scaling and line-shape claims, consider depositing the disorder-averaging and rate-equation code.","section":"Data Availability"}],"recommendation":"major_revision","confidential_remarks":"The paper is a forward-model study with no experimental comparison; its main advance over ref. 39 is the inclusion of disorder and the resulting DS line-shape prediction. The rate-equation inconsistency and the missing numerical support for the DS scaling are the key technical obstacles. If the authors correct the equations and provide the scaling evidence, the manuscript could become publishable. Given the heavy reliance on the authors' prior work for eigenstate/pathway classification, the novelty is incremental and the paper's fit to a broad journal will depend on how convincingly the central claim is documented."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know upfront. First, the paper does something genuinely new: by adding on-site disorder to the Tavis-Cummings model, the dark states acquire small transition dipoles, and the pump-probe signal at the dark-state energy becomes a direct probe of dark-state population. The predicted evolution from a derivative-like to an absorptive line shape, plus the O(1/N) scaling, is a concrete fingerprint that experimentalists could look for. That is worth taking seriously. Second, though, the kinetic model that drives the temporal evolution is internally inconsistent as written. The text says the rates are chosen so the LP decays 50% to the ground state and 50% to the dark-state manifold, but the stated values—1/k_LP=100 fs, 1/k_GS=50 fs, 1/k_DS=50/(N-1) fs—do not produce that branching. For N=10, Eq. 9 gives a total LP decay rate of about 1.63 fs⁻¹ and the DS manifold gets about 99% of the population. Even if you reinterpret k_LP as the total rate, the numbers still don't match. This matters because the sign and timing of the derivative-to-absorptive transition depend on the branching fractions.\n\nWhat the paper does well: the response-function framework is standard and the Appendix A decomposition of the line shape into trivial and derivative parts is clean. The analytic scaling argument for the LP signal is a nice check, and the simplified model explains why the DS signal survives at large N. The main physics—disorder-induced gray states and their relaxation—is clearly laid out. The extension of their earlier disorder-free work (ref. 39) is natural and the connection to experimental observables is explicit.\n\nSoft spots beyond the rate inconsistency: the disorder average is over 1000 realizations with no error bars, so we can't gauge the noise in the N-scaling. The scalings of μ²_LP→DLP and μ²_DS→2DS are only said to be 'numerically checked'—no data or derivation is shown. And the Markovian, parallel rate model is a strong assumption; coherences during the delay time are neglected without justification. These are addressable, but they need to be fixed or defended.\n\nBottom line: the core idea is plausible and the paper deserves a serious referee. It would not survive review in its current form because of the rate inconsistency, but with the parameters corrected and the numerical checks documented, it could become a solid contribution. I'd send it to review, but I'd make the kinetic model and the error bars a hard requirement for revision.","headline":"Useful extension of the group's polariton line-shape work, but the stated relaxation rates don't give the claimed 50/50 branching—the kinetic model needs fixing before the main prediction can be trusted.","tokens_in":15175,"tokens_out":9237,"would_cite":false,"duration_ms":67998,"reading_group":"maybe","serious_thinker":"no","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Adding on-site disorder to a Tavis–Cummings polariton model turns dark states into direct pump-probe probes, with a line shape that evolves from derivative-like to absorptive as the lower polariton relaxes into them.","keywords":["polaritons","dark states","Tavis-Cummings model","on-site disorder","pump-probe spectroscopy","transient absorption","line shape analysis","relaxation dynamics"],"falsifier":"Pump a disordered microcavity at the lower-polariton energy, record the pump-probe spectrum at the dark-state energy from zero to several hundred femtoseconds, and check for the predicted derivative-to-absorptive transition and its O(N^{-1}) amplitude scaling with molecule number; if the signal stays derivative-like long after the polariton decays, or if the amplitude does not fall as 1/N, the central claim would be disproved.","tokens_in":14111,"feed_emoji":"🔬","tokens_out":10895,"duration_ms":80876,"temperature":0.7,"pith_summary":"The paper asks what pump-probe signals look like at the energy of the dark molecular states in a disordered polariton system. In a clean Tavis–Cummings model, dark states have zero dipole moment and are spectroscopically silent; once random on-site energies are included, they acquire small photonic weight and a weak signal appears. The central finding is that this line shape is derivative-like right after the lower polariton is excited, then collapses into a single positive absorptive peak as the polariton relaxes into the dark-state manifold and the ground state. The authors trace this to a balance of Liouville-space pathways and show the signal scales as the inverse molecule number, the same scaling as the polariton lines, making the dark-state line shape a usable fingerprint of polaritonic response.","feed_headline":"Dark-state line shape flips to absorptive as polaritons relax","feed_subtitle":"Dark-state signal starts derivative-like, then becomes a single peak — a fingerprint of polariton relaxation.","key_machinery":"The paper's central object is the disordered Tavis–Cummings Hamiltonian (Eq. 1), with a single cavity mode, N two-level molecules, and random excitation energies drawn from a uniform distribution of width Δ. The key identity is the simplified two-Gaussian line-shape model (Appendix A): for two opposite-sign contributions of amplitudes A and B at energies separated by δ ≪ σ, the spectrum equals a Gaussian of amplitude A−B plus a derivative term (Bδ/σ²)ω times a Gaussian. This decomposition carries the argument: it explains why the DS line shape is derivative-like when the offset δ is finite (early times) and absorptive when δ→0 (late times), and it shows that both contributions scale as O(N^{","core_discovery":"Within a Tavis–Cummings model extended by on-site molecular disorder, the pump–probe spectrum at the dark-state (DS) energy is not featureless. Disorder gives the DSs small photonic weights, so transitions to and from them acquire small dipole moments. The paper finds that right after the lower polariton (LP) is pumped, the DS line shape is derivative-like, produced by partial cancellation of a positive ESA_DLP pathway and a negative GSB_DS pathway whose transition energies differ slightly. As the LP population relaxes into the DSs and the ground state, the ESA_DLP contribution vanishes and the remaining pathways all emit at nearly the same energy, so the line shape collapses into a single p","pith_inferences":["A decisive experiment would compare the same molecules coupled and uncoupled to a cavity: the bare molecular sample should show an absorptive line at the molecular energy at all delays, while the cavity-coupled sample should show the derivative-to-absorptive evolution at the dark-state energy.","If the Markovian assumption is relaxed to allow coherent mixing during the delay time, the derivative-like shape could persist longer than predicted, making the transition time a direct test of the kinetic model.","The two-Gaussian line-shape model suggests that at very large disorder the derivative extrema would merge, so the crossover disorder strength could be used to extract the homogeneous line width.","Because the DS signal is several orders of magnitude weaker than the polariton signals but scales identically with N, techniques that enhance signal-to-noise, such as higher-order spectroscopies or heterodyne detection, might make the fingerprint practical."],"forward_implications":["A derivative-like line shape at the dark-state energy in a fresh pump-probe spectrum marks the response as polaritonic, whereas a purely absorptive peak there can indicate molecular dark states after relaxation.","The time constant of the derivative-to-absorptive switch reflects the rate at which the lower polariton transfers population to the dark-state manifold, giving a direct observable for that relaxation.","Because the dark-state signal scales as N^{-1} just like the lower and upper polariton signals, it remains observable relative to them at experimentally realistic molecule numbers rather than vanishing faster.","The separation between the two extrema of the derivative line shape provides a measure of the disorder-induced width of the dark-state manifold when inhomogeneous broadening dominates.","Observing a purely absorptive dark-state line shape at all delay times would indicate either negligible disorder-induced DS transitions or no relaxation into the DS manifold, helping to constrain relaxation regimes."],"fun_headline_variants":["Disorder flips dark-state line shape to absorptive","Dark-state probe shows polariton relaxation signature","Pump-probe dark states: disorder yields absorptive peak","Disorder makes dark states visible in pump-probe","Dark-state line shape evolves with disorder and relaxation"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The predicted time-evolution rests on a parallel, Markovian rate-equation model in which the lower polariton decays 50% into the dark-state manifold and 50% into the ground state, and all coherences during the delay time are neglected.","fun_headline_variants_meta":{"raw":{"variants":["Disorder flips dark-state line shape to absorptive","Dark-state probe shows polariton relaxation signature","Pump-probe dark states: disorder yields absorptive peak","Disorder makes dark states visible in pump-probe","Dark-state line shape evolves with disorder and relaxation"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000153,"raw_usage":{"total_tokens":1056,"prompt_tokens":771,"completion_tokens":285,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":515,"completion_tokens_details":{"reasoning_tokens":209}},"tokens_in":515,"tokens_out":285,"duration_ms":3424,"temperature":1.0,"reasoning_tokens":209,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T17:33:11.521552+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Pump a disordered microcavity at the lower-polariton energy, record the pump-probe spectrum at the dark-state energy from zero to several hundred femtoseconds, and check for the predicted derivative-to-absorptive transition and its O(N^{-1}) amplitude scaling with molecule number; if the signal stays derivative-like long after the polariton decays, or if the amplitude does not fall as 1/N, the central claim would be disproved.","supporting_citations":[],"review_version":1}