{"id":"9aa5b81d-b725-448e-be48-8a684258e74b","arxiv_id":"2411.19818","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"The statistics of transmission and dwell time in disordered media made of resonant scatterers are set by the frequency-dependent mean free path and energy velocity, yielding large wavefront-shaping gains in transmission and stored energy.","lead":"Using computer simulations of light scattering by many tiny resonators in a waveguide, this paper maps how transmission and how long light lingers inside the material change as the light frequency approaches the resonators' natural frequency.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central claim presumes that the non-resonant dwell-time distribution Eq. (31) and the τmax–τTh scaling in Eq. (33) survive in resonant media; no derivation is given and the numerics never vary ℓ and v_E independently.","rationale":"I agree with the reader's conditional verdict and with the identification of Eq. (33) as fragile, but I would frame the load-bearing issue slightly more broadly: the entire non-resonant distribution Eq. (31), not just the τmax–τTh scaling, is imported into the resonant setting without derivation. The paper's simulations are consistent with the claim, and the analytic formulas are parameter-free, which is real evidence. However, the decisive test of sufficiency — independent variation of ℓ and v_E — is not performed. This does not justify rejection, but it does justify keeping the verdict conditional until the assumption is either derived or tested more directly. Because the reader already reached this verdict, I recommend no change.","tokens_in":18478,"tokens_out":8943,"duration_ms":82849,"concrete_test":"Use the same coupled-dipole solver to measure P(τ) in the diffusive regime for two parameter sets chosen so that L/ℓ is held fixed (adjusting Ns as δ and Q are changed) while the predicted v_E from Eq. (32) differs by at least a factor of 2. If the normalized distributions P(τ)/⟨τ⟩ do not collapse onto Eq. (31) with the τs/⟨τ⟩ ratio predicted by Eq. (33), the two-parameter characterization fails. As a complementary check, independently fit τs and ⟨τ⟩ from the simulated Q_d spectrum and compare with Eq. (33); a deviation beyond statistical error would pinpoint the unproven step.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The conclusion that P(τ) is fully characterized by ℓ(ω) and v_E(ω) rests on Eq. (31), which is carried over from non-resonant diffusive media, and on Eq. (33), which sets τs = (π/2)ℓ/v_E and ⟨τ⟩ = (π/2)L/v_E. The only justification offered is the sentence 'Assuming that the scaling of τmax with τTh must hold for resonant media.' That is not a derivation: τmax is the upper edge of the non-resonant distribution, and its relation to the Thouless time is a property of the same diffusion model that produced Eq. (31). The adapted energy velocity in Eq. (32) is also imported from Ref. [30] for 3D scattering without an explicit 2D derivation. The numerics provide support, but the tests are not decisive: Fig. 4 compares one diffusive detuning value, and Fig. 5 varies Ns at fixed resonance, so ℓ and v_E are changed simultaneously in a parameter combination fixed by the microscopic model. A shape correction or an additional dependence on, say, Q/kℓ at fixed L/ℓ would not be visible in these comparisons. Thus the strong claim that no other mesoscopic parameter enters P(τ) is not yet established.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the statistics of transmission and dwell-time matrices for light propagating through a 2D waveguide filled with point-like resonant scatterers. Using coupled-dipole simulations for up to 2×10^4 scatterers and 5760 disorder configurations, the authors show that the eigenvalue distributions P(T) and P(τ) evolve with frequency detuning, and they interpret the evolution through the frequency-dependent transport mean free path ℓ(ω) and energy velocity v_E(ω). The central claim is that these two mesoscopic parameters, together with the system length L and channel number N, fully characterize P(T) and P(τ) across quasi-ballistic, diffusive, and localized regimes. The paper derives predictions for wavefront-shaping enhancements of transmission and energy storage, including a factor ∝ Q/kℓ for high-Q resonators.","tokens_in":18767,"tokens_out":5169,"duration_ms":46098,"significance":"If the central claim holds, the paper provides a useful mesoscopic description of a broad class of resonant disordered media, connecting microscopic scattering parameters to wavefront-shaping performance. The numerical campaign is substantial: 5760 disorder configurations, up to 2×10^4 scatterers, and a numerically exact coupled-dipole treatment with proper waveguide renormalization of the polarizability. The analytic lines used for the diffusive and localized P(T) are parameter-free and agree well with simulations. The paper also correctly identifies that the energy velocity, not just the mean free path, controls dwell-time statistics, which is a physically important point for resonator-based systems. The main weaknesses are that some load-bearing analytic relations are assumed or extrapolated from non-resonant theory rather than derived for resonant media, and the numerical tests do not vary ℓ(ω) and v_E(ω) independently, so the 'fully characterized' claim is not yet decisively established.","major_comments":[{"comment":"Equation (33) sets τ_s=(π/2)ℓ(ω)/v_E(ω) and ⟨τ⟩=(π/2)L/v_E(ω) by 'assuming that the scaling of τmax with τTh must hold for resonant media.' This is an explicit assumption rather than a derivation. The coefficients of Eq. (31), and therefore the predicted P(τ) and the enhancement (34), depend on τ_s and ⟨τ⟩. The numerical tests in Figs. 4 and 5 vary detuning or N_s with other parameters fixed, so ℓ(ω) and v_E(ω) are not varied independently; the simulations cannot discriminate between the two-parameter description and, for example, an additional explicit Q dependence at fixed L/ℓ. Please provide a derivation of Eq. (33) for resonant media, or a numerical experiment in which ℓ and v_E are varied independently (e.g., different Q at matched L/ℓ and ℓ/v_E) to support the central two-parameter claim.","section":"Sec. III.B, Eq. (33)"},{"comment":"The energy-velocity expression (32) is 'adapt[ed]' from Ref. [30] to 2D scattering without a derivation. The factor 1+2/(πQ) and the v_φ/c^2 term control the predicted broadening c/v_E≈Q/2kℓ in Eq. (34). Since the local density of states and the 2D Green's function differ from the 3D case, the adaptation is not self-evident; without a derivation or an independent microscopic check, the quantitative prediction (34) rests on an unverified transport coefficient.","section":"Sec. III.B, Eq. (32)"},{"comment":"Equation (36), the localized-regime tail of P(τ), is obtained by 'extrapolating known results for the Wigner time in non-resonant media' to resonant cases. The supporting argument based on P(Γ)~1/Γ is heuristic, and the agreement shown in Fig. 4 is for a single parameter set. Since the conclusion claims full characterization in the localized regime and the τ_max/⟨τ⟩ enhancement in the conclusion uses Eq. (36), this extrapolation should be either derived or tested over a range of Q and N_s values.","section":"Sec. III.B, Eq. (36)"}],"minor_comments":[{"comment":"The closed-channel intensity profile in Eq. (21) is an extrapolation of the results of Ref. [41] to small transmission and is only qualitatively supported; please state this explicitly in the text and in the caption of Fig. 3.","section":"Sec. II.C, Eq. (21)"},{"comment":"The quasi-ballistic prediction (30) matches simulations only 'provided a scaling factor is introduced' for residual scattering; this scaling factor is a free parameter, and the manuscript should state this clearly and report its fitted value rather than presenting the curve as parameter-free.","section":"Sec. III.B, Eq. (30)"},{"comment":"The symbol α is used both for the coefficients in Eq. (31) and for the polarizability in Eq. (5); renaming the coefficients, for example to a, b, g, would remove a source of confusion.","section":"Sec. III.B, Eq. (31)"},{"comment":"The text says Eq. (34) follows by 'neglecting the small impact of the refractive index on phase velocity'; please specify the conditions under which v_φ≈c is a controlled approximation, since v_φ appears explicitly in the intermediate expression for v_E.","section":"Sec. III.B, Eq. (34)"}],"recommendation":"major_revision","confidential_remarks":"The paper is a solid numerical study with mostly parameter-free analytic descriptions and a clearly stated central claim. The main gap is the assumed transfer of non-resonant scaling relations to resonant media, particularly Eq. (33) and the adapted Eq. (32). These gaps are fixable by derivation or by targeted simulations that vary ℓ and v_E independently. If the authors cannot close this gap, the 'fully characterized' conclusion should be softened. I recommend major revision rather than rejection because the numerical evidence is strong and the assumptions are clearly identified in the text, even though they are not yet justified."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The transmission part is a clean, parameter-free confirmation that the known bimodal law holds once you compute the resonant mean free path \\ell(\\omega). The genuinely new contribution is the dwell-time statistics: near resonance the distribution broadens, and the maximum dwell time gains a Q/(k\\ell) enhancement with a concrete prefactor. Both are worth having.\n\nThe numerics are substantial — 5760 disorder configurations, up to 2\\times10^4 scatterers — and the diffusive curves in Figs. 2 and 4 use no fitted parameters. The coupled-dipole formalism with the waveguide-dressed polarizability is careful, and the paper is honest about what is assumed versus derived. It explicitly calls the bimodal P(T) \"unsurprising\" once \\ell(\\omega) is known, and it flags the extrapolated closed-channel profile in Eq. (21) and the scaling factor in the quasi-ballistic P(\\tau).\n\nThe soft spots are the ones the stress-test names. Equation (33) sets \\tau_s and \\langle\\tau\\rangle by \"assuming\" the \\tau_{\\rm max}–\\tau_{\\rm Th} scaling carries over to resonant media. That is the load-bearing step for the dwell-time predictions, and it is asserted, not derived from the microscopic model. Equation (32) likewise adapts the 3D energy-velocity expression to 2D without derivation. And the simulations do not independently vary \\ell(\\omega) and v_E(\\omega): Fig. 4 compares a single diffusive detuning, and Fig. 5 changes N_s at fixed resonance, so both parameters move together in a way fixed by the model. The data are consistent with the two-parameter ansatz, but they do not rule out a shape correction or an extra dependence on Q/k\\ell at fixed L/\\ell. That makes the \"fully characterized\" conclusion in the abstract somewhat stronger than the evidence. This is a moderate concern, not a fatal one: if v_E or the \\tau_s relation shifts, the enhancement factors change quantitatively, but the qualitative predictions and the parameter-free agreement in the tested regimes stand.\n\nThis paper deserves a serious referee. The wavefront-shaping and nanophotonics community will use the predicted L/\\ell and Q/k\\ell enhancement targets, and the authors should be asked to derive or better test Eq. (33) and to release code or data. I would bring it to the reading group.","headline":"Solid extension of mesoscopic wave transport to all-resonant media; transmission part is clean and parameter-free, dwell-time part rests on an assumed scaling that the numerics do not independently test.","tokens_in":19308,"tokens_out":4433,"would_cite":true,"duration_ms":38205,"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":"The paper claims that the transmission and dwell-time eigenvalue distributions of resonant disordered media are fully characterized by two mesoscopic parameters—the transport mean free path ℓ(ω) and the energy velocity vE(ω)—plus the…","keywords":["resonant disorder","transmission eigenvalue distribution","dwell-time operator","wavefront shaping","energy velocity","Anderson localization","mean free path","multiple scattering"],"falsifier":"Measure the largest dwell-time eigenvalue τmax in a two-dimensional waveguide at resonance (δ=0) with fixed length L, width W, and optical thickness kℓ(ω) (compensating any change in scatterer number Ns), and vary the resonator quality factor Q over two orders of magnitude; the paper predicts τmax/τb ≃ (2π/9)(L/ℓ(ω))(1 + Q/2kℓ(ω)), so a slope in Q/(2kℓ(ω)) that is not close to 1 would falsify the energy-velocity scaling and the dwell-time distribution built on it.","tokens_in":18255,"feed_emoji":"💡","tokens_out":9552,"duration_ms":70251,"temperature":0.7,"pith_summary":"This paper studies how resonant scatterers—particles that ring at a specific frequency—change the statistics of light transmitted through a disordered medium. It claims that the entire eigenvalue distributions of the transmission matrix and of the dwell-time (Wigner-Smith) operator are set by just two frequency-dependent mesoscopic parameters, the transport mean free path ℓ(ω) and the energy velocity vE(ω), together with the sample length L and the number of propagation channels N. If true, wavefront shaping becomes quantitatively predictable: the largest transmission eigenchannel transmits a factor ∝ L/ℓ(ω) more than a plane wave in the diffusive regime, rising to ∝ N $e^{{2L/ξ}}$ in the localized regime, and the longest-lived dwell-time eigenchannels store energy with an extra factor ∝ Q/kℓ(ω) when high-Q resonators replace low-Q ones. The paper supports these claims with coupled-dipole simulations in a two-dimensional waveguide across quasi-ballistic, diffusive, and localized transport regimes.","feed_headline":"Two parameters capture light transport and storage in resonant media","feed_subtitle":"Mean free path and energy velocity set which wavefronts transmit and how long light lingers.","key_machinery":"The central object is the matrix pair formed by the flux-normalized transmission matrix t(ω) and the dwell-time operator Qd(ω) = Q(ω)+Qi(ω)+Qe(ω), where Q(ω) is the Wigner-Smith operator built from frequency derivatives of t and r, Qi(ω) accounts for interference between incident and reflected fields, and Qe(ω) captures scattering into evanescent waveguide channels. These matrices are computed from coupled dipole equations with a dressed polarizability that enforces flux conservation, so that t†t+r†r = 1. The analytic engine is the injection of the detuning-dependent mean free path ℓ(ω) and the energy velocity vE(ω) into the known non-resonant distribution formulas: the bimodal P(T) of Eq. (15) and the dwell-time distribution of Eq. (31), with the scattering time and mean dwell time set by τs = (π/2)ℓ(ω)/vE(ω) and ⟨τ⟩ = (π/2)L/vE(ω) (Eq. (33)). The energy velocity expression of Eq. (32) is what produces the additional Q/kℓ(ω) enhancement of the longest dwell times.","core_discovery":"The central discovery is that resonant disorder does not require a new mesoscopic theory: inserting the frequency-dependent mean free path ℓ(ω) and energy velocity vE(ω) into the standard non-resonant formulas for the transmission and dwell-time eigenvalue distributions reproduces the numerically computed statistics at every detuning, from quasi-ballistic through diffusive to localized transport. Near resonance, ℓ(ω) drops, which makes the transmission eigenvalue distribution switch from monomodal to bimodal and back as the detuning is varied, while the dwell-time distribution broadens by orders of magnitude because vE(ω) slows the diffusive transport of energy. The same two parameters predict the wavefront-shaping enhancements: a transmission gain ∝ L/ℓ(ω) in the diffusive regime, a gain ∝ N $e^{{2L/ξ}}$ in the localized regime, and a dwell-time gain that acquires an additional factor ∝ Q/2kℓ(ω) when high-Q resonators are used, with the localized dwell-time distribution exhibiting a universal 1/τ² tail.","pith_inferences":["If the two-parameter reduction holds, other mesoscopic observables in resonant media—conductance fluctuations, intensity correlation functions, focusing contrast—should be obtainable by substituting ℓ(ω) and vE(ω) into existing non-resonant random-matrix formulas, a step the paper does not take.","Because the field-intensity profiles of the longest-lived dwell-time eigenstates are nearly independent of Q, experiments that measure only light intensity will miss most of the energy-storage enhancement; probing the material excitation (e.g., via fluorescence or absorption) would test the Q-scaling directly.","The predicted 1/τ² localized tail, if generic in any dimension, suggests wavefront shaping in resonant localized media could serve as a controllable source of very long-lived excitations, potentially relevant for slow-light or memory applications—an extension the paper leaves implicit.","A direct numerical test of Eq. (33) would be to extract τs and ⟨τ⟩ from time-dependent coupled-dipole simulations (e.g., pulse propagation) and compare them with πℓ(ω)/2vE(ω); a disagreement would require a resonant-specific correction to the dwell-time distribution."],"forward_implications":["If the two-parameter characterization is correct, P(T) and P(τ) for any resonant disordered medium can be predicted from ℓ(ω), vE(ω), L, and N without simulating the full microscopic scattering problem.","Wavefront shaping on the largest transmission eigenchannel yields transmission gains ∝ L/ℓ(ω) in the diffusive regime and ∝ N e^{2L/ξ} in the localized regime, where ξ = (π/2)Nℓ(ω).","The longest-lived dwell-time eigenchannel stores energy for a time ∝ L/ℓ(ω) longer than a plane wave, and replacing low-Q with high-Q resonators adds a factor ∝ Q/kℓ(ω) in both diffusive and localized regimes.","Near resonance, the transmission eigenvalue distribution passes from monomodal to bimodal and back as the detuning is varied, marking the quasi-ballistic-to-diffusive-to-localized crossover.","In the localized regime the dwell-time distribution has an unbounded 1/τ² tail, so the largest accessible dwell time grows with the number of disorder realizations probed, ∝ N²Nrℓ(ω)/L."],"supporting_citations":[{"why":"Supplies the decomposition of the dwell-time operator into Wigner-Smith, interference, and evanescent contributions, and the non-resonant dwell-time distribution (Eq. 31) that the paper adapts to resonant media.","marker":"[15]"},{"why":"Supplies the energy velocity expression for resonant multiple scattering that the paper specializes to 2D as Eq. (32), the key frequency-dependent parameter.","marker":"[30]"},{"why":"Provides the bimodal transmission eigenvalue distribution P(T) and the localization length ξ = πNℓ(ω)/2 used for the diffusive and localized predictions.","marker":"[33]"},{"why":"Supplies the diffusion-equation result ⟨T⟩ = πℓ(ω)/(2L + πℓ(ω)) and the mesoscopic framework for mean transmission.","marker":"[31]"},{"why":"Gives the Dorokhov open-channel picture underlying the bimodal P(T) and the counting of open channels g = N⟨T⟩.","marker":"[36]"},{"why":"Provides the transmission eigenchannel intensity profiles (Eqs. 20–21) used to interpret the eigenstate propagation in the diffusive regime.","marker":"[41]"},{"why":"Supplies Wigner time-delay statistics for one-dimensional disordered systems used to derive the localized dwell-time tail (Eq. 36).","marker":"[46–48]"},{"why":"Provides the scattering approach to Anderson localization whose prediction for P(τ) the localized power-law tail matches, up to a prefactor.","marker":"[49]"},{"why":"Supplies the Fisher-Lee relations used to construct the flux-normalized transmission and reflection matrices from the Green's function.","marker":"[32]"},{"why":"Supplies the electromagnetic energy density expression (Eq. 23) that defines the dwell-time operator via total stored energy.","marker":"[44]"}],"fun_headline_variants":["Resonant media: two parameters rule light transport","Wavefront shaping boosts light and storage in resonant media","Resonance shifts transmission statistics and dwell times","Mean free path and energy velocity predict resonant light transport"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the non-resonant relation between the longest dwell time and the diffusion (Thouless) time, transplanted to resonant media via Eq. (33) with the energy velocity from Eq. (32), remains exact—if the true energy velocity differs from that expression, the predicted Q/kℓ(ω) broadening of dwell times does not follow.","fun_headline_variants_meta":{"raw":{"variants":["Resonant media: two parameters rule light transport","Wavefront shaping boosts light and storage in resonant media","Resonance shifts transmission statistics and dwell times","Mean free path and energy velocity predict resonant light transport"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000229,"raw_usage":{"total_tokens":1532,"prompt_tokens":1052,"completion_tokens":480,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":668,"completion_tokens_details":{"reasoning_tokens":418}},"tokens_in":668,"tokens_out":480,"duration_ms":4744,"temperature":1.0,"reasoning_tokens":418,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T05:47:20.801275+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the largest dwell-time eigenvalue τmax in a two-dimensional waveguide at resonance (δ=0) with fixed length L, width W, and optical thickness kℓ(ω) (compensating any change in scatterer number Ns), and vary the resonator quality factor Q over two orders of magnitude; the paper predicts τmax/τb ≃ (2π/9)(L/ℓ(ω))(1 + Q/2kℓ(ω)), so a slope in Q/(2kℓ(ω)) that is not close to 1 would falsify the energy-velocity scaling and the dwell-time distribution built on it.","supporting_citations":[{"cited_title":"Durand, S","cited_arxiv_id":null,"evidence_quote":"Supplies the decomposition of the dwell-time operator into Wigner-Smith, interference, and evanescent contributions, and the non-resonant dwell-time distribution (Eq. 31) that the paper adapts to resonant media."},{"cited_title":"Lagendijk and B","cited_arxiv_id":null,"evidence_quote":"Supplies the energy velocity expression for resonant multiple scattering that the paper specializes to 2D as Eq. (32), the key frequency-dependent parameter."},{"cited_title":"Akkermans and G","cited_arxiv_id":null,"evidence_quote":"Supplies the diffusion-equation result ⟨T⟩ = πℓ(ω)/(2L + πℓ(ω)) and the mesoscopic framework for mean transmission."},{"cited_title":"Dorokhov, On the coexistence of localized and ex- tended electronic states in the metallic phase, Solid State Communications 51, 381 (1984)","cited_arxiv_id":null,"evidence_quote":"Gives the Dorokhov open-channel picture underlying the bimodal P(T) and the counting of open channels g = N⟨T⟩."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the transmission eigenchannel intensity profiles (Eqs. 20–21) used to interpret the eigenstate propagation in the diffusive regime."},{"cited_title":"Ossipov, Scattering approach to anderson localization, Phys","cited_arxiv_id":null,"evidence_quote":"Provides the scattering approach to Anderson localization whose prediction for P(τ) the localized power-law tail matches, up to a prefactor."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the Fisher-Lee relations used to construct the flux-normalized transmission and reflection matrices from the Green's function."},{"cited_title":"Landau, L","cited_arxiv_id":null,"evidence_quote":"Supplies the electromagnetic energy density expression (Eq. 23) that defines the dwell-time operator via total stored energy."}],"review_version":1}