{"id":"e1375566-39e3-4c17-8738-ea1011bea82d","arxiv_id":"2607.19242","paper_version":1,"verdict":"CONDITIONAL","confidence":"LOW","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Pumped excitons produce persistent coherent oscillations that generate exciton-field-induced Floquet sidebands in the TR-ARPES spectrum of a two-band semiconductor.","lead":"This paper simulates how a laser pulse can create excitons—bound electron-hole pairs—whose collective oscillations persist after the pulse and imprint fake 'Floquet' sidebands in time-resolved photoemission spectra. The result matters because it offers a cheap, interaction-aware method to explain and predict these ultrafast spectral fingerprints in pumped semiconductors.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"HF unitary evolution has no dephasing: post-pump sidebands are guaranteed to persist, so experimental corroboration is underdetermined","rationale":"The reader's weakest assumption correctly identifies the HF approximation as the fragile link. I sharpen this into a concrete in-principle issue: the HF dynamics defined by Eqs. (12)–(13) are unitary and admit no dephasing channel, so the post-pump persistence of coherent oscillations is built into the approximation, not independently predicted. This does not prove the physical mechanism wrong; it makes the experimental-corroboration claim conditional on coherence surviving beyond the probe delay, which the paper does not establish. Code/data absence and convergence checks are secondary to this in-principle sensitivity.","tokens_in":20797,"tokens_out":9406,"duration_ms":122525,"concrete_test":"Recompute Fig. 2(e) (A0 = 0.5 V·fs/nm, semiconducting phase) with a phenomenological dephasing rate γ = 0.05 eV added to the off-diagonal density-matrix evolution, e.g. by modifying Eq. (12) to include a Lindblad-type decay for interband coherences, and reevaluate the TR-ARPES signal at tpr = 40 fs. If the 1.35 eV sideband disappears or shifts by more than its linewidth, the quantitative sideband prediction depends on the absence of dephasing and should be reported conditionally.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central sideband claim rests on Eq. (12) with Eq. (13): the evolution P(k,t) is unitary (iℏ∂tP = ΞP with Hermitian Ξ), so any coherence generated by the pump persists indefinitely. Consequently, the post-pump oscillations of ΔΓ(t) in Fig. 3(a) and the sidebands in Fig. 2 at tpr = 40 fs are guaranteed by the mean-field unitary dynamics; there is no dephasing, recombination, or screening channel. The paper provides no estimate of a dephasing rate or a beyond-HF comparison, so the claim that these features \"corroborate\" the experiment of Ref. [82] is not quantitatively secured. This is not an internal inconsistency, but it is the least secure link between the model and the claimed physical result: with a realistic dephasing time comparable to 40 fs, the sideband maximum and its intensity shift with pump amplitude could be substantially altered or washed out.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper extends the Dynamical Projective Operatorial Approach (DPOA) to pumped excitonic systems, incorporating Coulomb interactions at the Hartree-Fock (HF) level. It derives the time evolution of the single-particle density matrix and the TR-ARPES signal, and applies the formalism to a prototypical two-dimensional two-band semiconductor. The authors compute the equilibrium phase diagram, define an equilibrium excitonic frequency via the response to a weak impulsive pump, and then show that pumping at or near this frequency produces post-pump Floquet sidebands in the TR-ARPES spectrum, which they attribute to coherent oscillations of the HF self-energy ('excitonic field'). They distinguish these from band-resonance-induced sidebands, analyze the local interaction limit, and claim corroboration of the experimental findings of Ref. [82].","tokens_in":21040,"tokens_out":5071,"duration_ms":52192,"significance":"The framework offers an efficient route to TR-ARPES simulations in interacting systems, and the compact expressions for the signal [Eqs. (20)-(22)] are a useful contribution. The separation of exciton-field-induced and band-resonance-induced sidebands is conceptually valuable. However, the physical conclusions are weakened by two structural issues: the unitary HF dynamics contain no dephasing channel, so post-pump coherence and sidebands are guaranteed by the mean-field approximation; and the excitonic frequency used as the pump resonance is defined from the same model's response, making the central 'resonant excitation' claim partly self-referential. Quantitative predictions also lack convergence documentation. If these concerns are addressed, the paper could provide a sound mean-field demonstration, but the current experimental-corroboration claim is not quantitatively secured.","major_comments":[{"comment":"The evolution matrix P(k,t) satisfies iℏ∂tP = ΞP with Hermitian Ξ, so P is unitary and the single-particle density matrix evolves unitarily. Any coherence generated by the pump persists indefinitely; there is no dephasing, recombination, or screening channel. Therefore the post-pump oscillations in Fig. 3(a) and the sidebands at t_pr=40 fs in Figs. 2 and 4 are an inescapable consequence of the mean-field dynamics, not a nontrivial physical prediction. To claim quantitative corroboration of Ref. [82], the authors must include a dephasing/dissipation channel or provide an estimate of realistic dephasing times and show that the sidebands survive at comparable probe delays.","section":"Sec. II.C, Eqs. (12)-(13)"},{"comment":"The equilibrium excitonic frequency is defined as the maximum of |ΔΓ(ω)| obtained from the same HF model under a weak impulsive pump. The later finding that pumping at that frequency produces strong sidebands is therefore a consistency check rather than an independent prediction. The pump-amplitude-induced shift of the sideband maximum (1.20→1.35 eV) is not circular, but the claim that the effect is 'resonant with an excitonic mode' relies on a frequency extracted from the same observable. The authors should state this explicitly or define ω_exc from an independent linear-response/Bethe-Salpeter calculation.","section":"Sec. III.B and III.C"},{"comment":"The protocol for extracting ω_exc states that robustness 'has been verified' but no supporting data are shown. More importantly, no convergence study is reported for the k-grid (64×64), time step, pump/probe durations, or the Fourier-analysis window (25–55 fs with Hann windowing). Since the central quantitative claims—sideband maxima at 1.20, 1.23, and 1.35 eV and the shifts of 0.06–0.20 eV—are extracted from these numerics, the absence of a convergence analysis leaves the quantitative predictions unsupported.","section":"Sec. III.B, Fig. 1"},{"comment":"The entire analysis is performed in the HF approximation with a density-density interaction restricted to valence-conduction terms (Eqs. 7 and 13). This approximation neglects screening dynamics, correlation, and dephasing. The paper gives no estimate of the HF error or a comparison with a beyond-HF method (e.g., exact diagonalization on a small cluster or TD-GW). Such a comparison would be needed to assess whether the excitonic frequencies and their pump-induced shifts are robust physical quantities or mean-field artifacts.","section":"Sec. II.B and III"}],"minor_comments":[{"comment":"The caption states 'V=15 eV, excitonic-insulator phase' but the parenthetical text says '(V=6 eV, κ=0.2)'. This is inconsistent; the correct value appears to be V=15 eV.","section":"Fig. 5 caption"},{"comment":"The notation Ξ0(k) + Ξpu(k,t) = T_k(t) + e E_pu(t)·D_k(t) mixes momentum and Wannier bases. It would be clearer to write the right-hand side after the generalized Peierls substitution of Eq. (11), or to explicitly state the basis.","section":"Eq. (10)"},{"comment":"The text says the BZ is sampled by a uniform grid of 64×64 k-points 'including the Γ point'. If Γ is included, the grid may not be uniform in the usual Monkhorst-Pack sense; please clarify.","section":"Sec. III.A"},{"comment":"In panels (b), (d), and (f), the horizontal gray line denotes the equilibrium excitonic energy, but it is not labeled in the figure. Adding a legend or label would improve readability.","section":"Fig. 2"},{"comment":"The word 'dubbed' appears repeatedly in the text describing pump parameters and sideband types; consider using more formal terminology.","section":"Sec. III.C"}],"recommendation":"major_revision","confidential_remarks":"The paper's claim of experimental corroboration is significantly weakened by the unitary HF dynamics: post-pump sidebands are guaranteed by the absence of dephasing, so their presence does not constitute a predictive test. The self-referential definition of the excitonic frequency should also be acknowledged. I do not see grounds for rejection—the derivations are internally consistent—but the authors should be asked to add a dephasing estimate or substantially soften the experimental claim, and to document numerical convergence."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The part worth keeping is the method application: DPOA extended to a self-consistent HF two-band model, with the TR-ARPES signal reduced to a Q-matrix convolution. That is genuinely cheap and lets the authors separate exciton-field sidebands from band-resonance sidebands by looking at the post-pump regime where only the HF self-energy oscillates. The local-interaction limit is a nice control: because the self-energy becomes k-independent but the sideband stays strongest near Γ, the small-detuning argument does real work. The Mexican-hat flattening at higher pump amplitude is also consistent with the cited experiment.\n\nThe soft spots are what you would expect from a purely mean-field, parameter-heavy study. Most important: the time evolution is unitary, so any coherence generated by the pump persists forever. The post-pump oscillations in Fig. 3 and the sidebands at t_pr = 40 fs are guaranteed by the absence of dephasing, recombination, or screening. Calling this a quantitative corroboration of Ref. [82] is therefore an overreach. A realistic dephasing time on the order of tens of femtoseconds could move or wash out the sideband intensity maximum. The paper needs an estimate of the HF error or a beyond-HF test before that claim is secure.\n\nSecond, there are no convergence studies, no benchmarks against the TD-aGW results, no code, no data. For a method paper that emphasizes efficiency, a runtime comparison and a k-grid/timestep convergence check would substantially raise confidence. The parameter choices are numerous and the quantitative claims (1.20 eV shifting to 1.35 eV, 0.06–0.20 eV shifts in the ΔΓ peak) rest entirely on unshown numerical output.\n\nOne apparent criticism is not a real flaw: using a weak impulsive pump to find the equilibrium excitonic frequency and then pumping at that frequency is standard resonance determination, not circularity. The paper explicitly notes that for weak pumps the extracted frequency is an intrinsic system property. That part is fine.\n\nMinor: Fig. 5's caption says V = 6 eV while the text and Fig. 1(b) indicate V = 15 eV; a typo, but one that should be caught.\n\nThe physics is plausible, the method is clearly presented, and the sideband classification will be useful to people simulating pump-probe spectra of excitonic systems. It deserves a serious referee, but with the expectation that the authors add dephasing estimates, convergence checks, and temper the experimental-corrobation claim.","headline":"DPOA applied to HF-interacting pumped semiconductors is a useful methodological extension with clean sideband classification, but the experimental corroboration is softer than claimed because the mean-field dynamics have no dephasing and the numerics ship no convergence checks.","tokens_in":21535,"tokens_out":1367,"would_cite":true,"duration_ms":18015,"reading_group":"maybe","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 after resonant photoexcitation, coherent oscillation of the excitonic order parameter acts as an internal field that imprints Floquet sidebands on the electron spectrum, and that these sidebands persist after the pump","keywords":["time-resolved ARPES","excitons","Floquet sidebands","Hartree-Fock","pump-probe","excitonic insulator","two-dimensional semiconductor","ultrafast dynamics"],"falsifier":"If a beyond-mean-field simulation of this model that includes dephasing shows the ΔΓ(t) oscillation damping within tens of femtoseconds, the sidebands would vanish; likewise, a TR-ARPES experiment on a monolayer transition-metal dichalcogenide pumped at the 1s exciton resonance that sees no post-pump sideband would contradict the claim.","tokens_in":20732,"feed_emoji":"⚛️","tokens_out":5294,"duration_ms":63225,"temperature":0.7,"pith_summary":"The paper aims to show that after an intense pump pulse resonantly excites an excitonic mode in a semiconductor, the excitonic order parameter keeps oscillating and acts as its own time-periodic field, producing Floquet sidebands in time-resolved photoemission even after the laser is off. The authors compute this in a two-dimensional two-band model with Coulomb interactions at the Hartree-Fock level, using an operator-based real-time method. They distinguish these exciton-field sidebands from conventional laser-driven Floquet sidebands and from sidebands caused by residual interband coherences. If true, this gives a clean fingerprint of coherent excitonic dynamics and a computationally light route to a signal previously seen in expensive simulations and in experiment.","feed_headline":"Pumped excitons leave Floquet sidebands that outlive the laser","feed_subtitle":"A Hartree-Fock simulation shows the oscillating exciton field alone can dress bands—no laser light needed for the post-pump signal.","key_machinery":"The engine is the time-dependent Hartree-Fock self-energy, specifically its off-diagonal element ΔΓ(t), which serves as the excitonic order parameter. Within the authors' operator-based real-time evolution, the pump dresses the bands while it is on; once it is off, ΔΓ(t) continues to oscillate at the excitonic frequency, acting as an internal oscillating field that generates Floquet sidebands. The key variable is the Fourier peak |ΔΓ(ω)|, whose frequency and amplitude encode the pump-induced shift of the excitonic resonance and the strength of the sideband.","core_discovery":"The central claim is that the oscillating off-diagonal Hartree-Fock self-energy—the 'excitonic field'—is sufficient to produce the Floquet sidebands seen in the post-pump TR-ARPES spectrum. In the semiconducting phase, pumping near the equilibrium excitonic resonance leaves the order parameter oscillating at a pump-intensity-dependent frequency, and this oscillation dresses the valence band with a parallel sideband that is strongest at the zone center. In the excitonic-insulator phase, the pump partially melts the equilibrium order and the same mechanism yields sidebands parallel to both occupied bands. The authors also identify a second, distinct class of sidebands arising from residual coh","pith_inferences":["The pump-intensity dependence of the excitonic frequency suggests a mean-field nonlinearity that could be used to extract the electron-hole interaction strength from a series of TR-ARPES measurements at different fluences.","If dephasing were added, the post-pump sidebands would likely persist only for a time set by the inverse exciton linewidth; the paper's prediction of near-infinite coherence is its most exposed point, so a time-resolved measurement of sideband lifetime is a natural next test.","The same oscillating-order-parameter mechanism should apply to other collective modes (spin waves, charge order) pumped resonantly, suggesting TR-ARPES as a general probe of coherent order-parameter dynamics."],"forward_implications":["TR-ARPES at positive pump-probe delays can distinguish exciton-mediated Floquet dressing from laser-field dressing, because only the former survives after the pulse.","The excitonic frequency is pump-intensity tunable: stronger pulses shift the effective resonance, so sideband positions encode pump fluence.","The Mexican-hat renormalization of the valence band emerges as a natural consequence of the oscillating mean field, offering a measurable band-structure signature.","In the excitonic-insulator phase, TR-ARPES sidebands running parallel to occupied bands provide a probe of pump-induced melting and recovery of excitonic order.","The same framework extends to local-interaction models, indicating the phenomenology does not rely on the long-range character of the Coulomb potential."],"fun_headline_variants":["Exciton field alone creates Floquet sidebands after pump","Pumped excitons leave self-dressed Floquet bands","Excitonic field, not laser, sustains Floquet sidebands","Oscillating exciton order dresses bands with sidebands"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The post-pump coherence that drives the sidebands rests entirely on the Hartree-Fock mean-field approximation, which neglects dephasing, screening dynamics, and correlation-induced decay.","fun_headline_variants_meta":{"raw":{"variants":["Exciton field alone creates Floquet sidebands after pump","Pumped excitons leave self-dressed Floquet bands","Excitonic field, not laser, sustains Floquet sidebands","Oscillating exciton order dresses bands with sidebands"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000532,"raw_usage":{"total_tokens":2419,"prompt_tokens":787,"completion_tokens":1632,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":531,"completion_tokens_details":{"reasoning_tokens":1558}},"tokens_in":531,"tokens_out":1632,"duration_ms":11835,"temperature":1.0,"reasoning_tokens":1558,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T12:57:21.001087+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"If a beyond-mean-field simulation of this model that includes dephasing shows the ΔΓ(t) oscillation damping within tens of femtoseconds, the sidebands would vanish; likewise, a TR-ARPES experiment on a monolayer transition-metal dichalcogenide pumped at the 1s exciton resonance that sees no post-pump sideband would contradict the claim.","supporting_citations":[],"review_version":1}