{"id":"fb660557-3e8b-4454-b7fb-380fd9b7ff3d","arxiv_id":"1908.05568","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"A non-stationary two-time correlation measurement shows that timing jitter in photon condensate formation, seeded by spontaneous emission, is the dominant fluctuation near a transient phase transition.","lead":"Researchers studied a dye-filled optical cavity in which light briefly condenses after a strong pump pulse. They found the exact time of condensation changes from shot to shot, and this timing jitter grows near the transition threshold, revealing a new kind of transient-phase-transition fluctuation.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central claim depends on the fitted non-Markovian reservoir regime, but the fit lacks uncertainty quantification and a statistical test against the Markovian limit, where the paper itself says jitter vanishes.","rationale":"After reviewing the manuscript, I find the reader's weakest assumption to be the correct load-bearing concern. The experimental observation of off-diagonal anti-correlations is the key evidence for formation jitter, but its interpretation is model-dependent: the same data could in principle arise from other non-Markovian mechanisms, and the model's explanation (jitter) requires the specific reservoir regime. The paper's own statement that jitter vanishes in the Markovian regime makes the fitted parameter values load-bearing. The authors do not provide statistical uncertainties for these fits, nor do they test whether the Markovian alternative is excluded by the one-time data. The two-time agreement is a strong point in favor of the model, but because the parameters were fit to one-time data and then used to generate the two-time prediction, it does not remove the need for regime validation. I also note a secondary tension in the free-energy section: the universality argument assumes a lossless fixed-Nex limit, which is not the fitted regime; however, this mainly affects the generality claim, not the core observation. The concrete test proposed—a likelihood-based comparison of non-Markovian versus Markovian fits, plus a trajectories check at the Markovian boundary—would settle whether the concern lands. If the Markovian boundary is excluded by the data, the paper's conditional acceptance is appropriate; if not, the central claim would be unverified.","tokens_in":16127,"tokens_out":15907,"duration_ms":146970,"concrete_test":"Re-fit the mean-field model of Eqs. (2)-(3) to the experimental one-time pulse shapes (Fig. 3) and light-yield curves (Fig. 2 inset) over a grid of (κ, Γ↓/Γ0), including the Markovian boundary Γ↓≫κ, and compute the profile likelihood or χ² surface. If the 95% confidence region for (κ, Γ↓/Γ0) includes the Markovian boundary, or if the best Markovian fit is within Δχ²≈4 of the best fit, the regime assignment is unsupported and the jitter interpretation is not secured. As a cross-check, run the quantum-trajectories model at the best Markovian-fit parameters and confirm that the off-diagonal anti-correlation lobes in g(2)(t1,t2) disappear; if they persist, the paper's stated Markovian limit would be incorrect.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central claim—that the off-diagonal anti-correlations in g(2)(t1,t2) witness formation jitter and that this jitter is a universal feature of transient phase transitions—rests on the system operating in the non-Markovian reservoir regime. In Sec. II D the authors fit κ=10^10 s^-1 and Γ↓=0.998 Γ0 to the averaged pulse shapes, and in the final paragraph of Sec. III they state that in the Markovian regime (Γ↓ dominating molecular de-excitation) both critical slowing down and timing jitter vanish. If the true reservoir were closer to Markovian, the anti-correlations would require a different explanation and the universality argument would collapse. The manuscript reports no error bars on these fitted parameters, no goodness-of-fit measure, and no model-selection comparison against the Markovian limit. The quantum-trajectories match to the two-time data is encouraging but uses the same fitted parameters, so it does not independently validate the regime choice. Compounding this, the free-energy universality argument in Sec. III assumes fixed total excitation number (κ=Γ↓=Γ↑=0), which is not the fitted regime (κ≫Γ↓); the paper flags this but does not quantify how much the fixed-Nex approximation degrades the predicted universal behavior. Thus the load-bearing premise—the non-Markovian regime—is asserted rather than statistically established.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports an experimental and theoretical study of transient photon condensation in a dye-filled microcavity after pulsed excitation. The authors measure the time-resolved cavity output for a range of pump energies and introduce a two-time, non-stationary second-order correlation function g(2)(t1,t2) as a probe of the transient relaxation dynamics. They observe delayed condensate formation near threshold, interpreted as a transient analogue of critical slowing down, and diagonal correlations together with off-diagonal anti-correlations in g(2), interpreted as shot-to-shot timing jitter in the condensate formation seeded by spontaneous emission. The experimental results are compared with mean-field rate equations, a quantum regression approach, and quantum trajectory simulations using parameters fitted to the averaged pulse shapes. The authors then construct an effective free-energy landscape and argue that formation jitter is a universal feature of quenches through second-order phase transitions.","tokens_in":16422,"tokens_out":5702,"duration_ms":56128,"significance":"If substantiated, the paper introduces a genuinely useful experimental tool, the non-stationary two-time correlation function, and identifies a distinct fluctuation phenomenon in transient phase transitions, namely timing jitter in order-parameter formation. A notable strength is that the two-time data are compared with a quantum-trajectories model whose parameters were fitted to independent one-time intensity data, providing a partially independent test of the model. The free-energy argument connects the observations to a broader class of systems. However, the central interpretation and the universality claim rest on the identification of a specific non-Markovian reservoir regime, and the manuscript currently does not provide the statistical evidence needed to establish that regime. The paper is therefore significant but requires further quantitative support before the claims can be fully accepted.","major_comments":[{"comment":"The classification of the system as non-Markovian rests entirely on the fitted values kappa=10^10 s^-1 and Gamma_down=0.998 Gamma_0, yet no uncertainties, goodness-of-fit measures, or model-selection tests are reported. Because the final paragraph of Sec. III states that in the Markovian regime both critical slowing down and timing jitter vanish, the central claim depends on this fitted regime being statistically distinguishable from the Markovian limit. The authors should provide confidence intervals for (kappa, Gamma_down, alpha) and a quantitative comparison, such as a profile likelihood or an information criterion, against the Markovian alternative.","section":"Sec. II D and Sec. III (final paragraph)"},{"comment":"The effective free-energy landscape in Eq. (7) is derived after setting kappa=Gamma_down=Gamma_up=0, whereas the fitted experimental regime has kappa >> Gamma_down (kappa approximately 40 times Gamma_0). The paper acknowledges that the fixed-Nex approximation is only valid in the non-Markovian regime, but it does not quantify how the finite-loss dynamics modify the free-energy geometry or the predicted jitter. To support the universality claim, the authors should show that the Langevin dynamics on Eq. (7) reproduces the same qualitative g(2) features as the full quantum-trajectory model with the actual lossy parameters, or provide a controlled expansion in the loss-to-stimulated-emission ratio.","section":"Sec. III, Eq. (7)"},{"comment":"The experimental g(2)(t1,t2) maps are presented without error bars or confidence intervals. Since the off-diagonal anti-correlation lobes are the central witness of formation jitter, the reader cannot assess whether the deviations from g(2)=1 are statistically significant, particularly at P/Pth=1.07 where the count rate is low (Fig. 6, bottom left). The authors should include uncertainty estimates derived from finite detection counts and perform a statistical comparison between the experimental maps and the quantum-trajectory predictions.","section":"Figs. 5 and 6"}],"minor_comments":[{"comment":"The jump rate for the free-space spontaneous emission process sqrt(Gamma_down) sigma_- is written as R2=Gamma_down n, but it should be proportional to the number of excited molecules m, i.e., R2=Gamma_down m. This appears to be a typo that should be corrected.","section":"Appendix B, Eq. (B3)"},{"comment":"The statement that the slight deviation of the quantum-trajectory result from the off-diagonal experimental data 'may be attributed to the error that propagates from determining the peak time t0' is not quantified. Please propagate the uncertainty in t0 into the theoretical curve or otherwise justify that the deviation is within the expected error.","section":"Sec. II E, Fig. 6"},{"comment":"The caption states that the cavity cutoff is set to lambda0=595 nm and notes a difference from Sec. II D, but the main text does not specify the cutoff used for the data in Fig. 3. Please clarify which cutoff applies to which figure and whether this affects the parameter fits.","section":"Fig. 5 caption"},{"comment":"The validity conditions for the approximation in Eq. (5) are stated as '[a^dag(t1),a(t2)] approx 0 or <a^dag(t)a(t)> >> 1'. For t1=t2 the commutator is not small, so the large-photon-number condition is the operative one on the diagonal; stating this explicitly would avoid confusion.","section":"Eq. (5)"}],"recommendation":"major_revision","confidential_remarks":"The paper is well written and presents a novel experimental observable, but the central claim hinges on the fitted non-Markovian regime, for which no statistical justification is currently given. The authors should treat the uncertainty quantification and model-selection analysis as essential rather than cosmetic. I also suggest the editor ensure that the simulation code or data, if available, is checked for the apparent typo in the spontaneous emission jump rate in Appendix B."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know. First, the paper has a real experimental result: the first non-stationary two-time g(2) map for a dye-microcavity photon condensate, showing diagonal bunching and off-diagonal anti-correlations that are a clean fingerprint of shot-to-shot jitter in condensate formation. Second, the interpretation is sound but the universal claim outruns the evidence.\n\nThe experiment itself is well done. They quench a dye-filled microcavity, record pulse shapes near threshold, and see the expected slowing down and broadening. The new analysis is the two-time correlation function; the off-diagonal anti-correlations are exactly what you would expect if entire pulses form early or late. The quantum trajectories model, with parameters fixed by one-time intensity data, reproduces the two-time maps reasonably well. The quantum regression approximation captures the qualitative features but misses quantitative details, and the authors say so. That is honest.\n\nThe soft spot is the fitted non-Markovian reservoir. The authors fit κ=10^10 s^-1 and Γ↓=0.998Γ0 to averaged pulse shapes, then use those same parameters in the trajectories that match g2. In the Markovian limit they state that both critical slowing down and jitter vanish. So the whole effect depends on being in a regime that is inferred from fits without error bars or a model-selection test against the Markovian alternative. The g2 maps also have no error bars, so we cannot judge how significant the deviations are. The free-energy geometry in Sec. III is reverse-engineered from the same mean-field equations and assumes a fixed total excitation number that the fitted regime violates; they flag it, but do not quantify how much that approximation matters. The universality claim should be presented as a conjecture, not as established.\n\nCitation pattern looks fine. The result is not as new as the text sometimes implies: turn-on jitter seeded by spontaneous emission is old news in laser physics, and they cite the relevant literature. The novelty is the two-time measurement and the free-energy framing.\n\nWho it is for: people working on photon condensates, micro- and nano-lasers, and transient phase transitions. It deserves a serious referee: the experiment is careful, the model comparison is partially independent, and the two-time tool is worth having in the literature. With error bars and a stricter statement about universality, I would be comfortable with it. I would bring it to reading group.","headline":"A careful experiment-theory paper that introduces a genuinely useful two-time g(2) tool for transient photon condensation, with a plausible jitter mechanism that the universality claim outruns.","tokens_in":16968,"tokens_out":1973,"would_cite":true,"duration_ms":19239,"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":"Transient phase transitions exhibit a distinct divergent fluctuation: jitter in the time the ordered phase forms, witnessed by two-time non-stationary correlations.","keywords":["transient phase transition","photon condensation","non-stationary correlation function","formation jitter","critical slowing down","non-Markovian reservoir","effective free energy","dye-filled microcavity"],"falsifier":"Measure $g^{(2)}(t_1,t_2)$ in the same cavity after increasing the molecular fluorescence rate $\\Gamma_\\downarrow$ relative to the cavity emission rate $E$ so that $\\Gamma_\\downarrow \\gg E$, pushing the bath toward Markovian behaviour; the model predicts that the off-diagonal anti-correlation lobes disappear and the near-threshold pulse broadening is strongly reduced, even though a delayed pulse remains.","tokens_in":15932,"feed_emoji":"⏱️","tokens_out":8012,"duration_ms":71838,"temperature":0.7,"pith_summary":"This paper tries to establish that transient phase transitions—systems quenched suddenly through a critical point—show a distinct kind of diverging fluctuation: jitter in the time at which the ordered phase forms, not just in its size. The evidence comes from a dye-filled microcavity that condenses photons after a short pump pulse; by measuring the full two-time, non-stationary correlation function $g^{(2)}(t_1,t_2)$, the authors show that the whole condensate pulse forms early or late on each realization, with the effect amplified near the critical excitation energy. They trace the seed of this jitter to spontaneous emission and argue, through an effective free-energy landscape, that the mechanism is universal for quenches through second-order phase transitions. A sympathetic reader would care because this provides a general statistical tool for transient, non-stationary ordering and predicts a testable signature—off-diagonal anti-correlations in $g^{(2)}$—in lasers, nano-lasers, and colloidal growth.","feed_headline":"Formation jitter marks transient phase transitions","feed_subtitle":"Two-time correlation measurements show condensate pulses arrive early or late shot to shot near the critical point.","key_machinery":"The central objects are the two-time non-stationary second-order correlation function $g^{(2)}(t_1,t_2)$, defined as the normalized joint photon-detection probability, and the effective free-energy landscape $F(n)=-\\int_0^n \\dot n'\\,dn'$ for the photon number $n$ as order parameter. The correlation function carries the measurement: because the transient system lacks time-translation symmetry, the single-time $g^{(2)}(\\tau)$ is insufficient, and the full two-time map separates diagonal number fluctuations from off-diagonal timing-jitter correlations. The free-energy landscape carries the generalization: the relation $d\\psi/dt=-\\partial F/\\partial\\psi$ turns the microscopic rate equations into a geometry problem, and a Langevin walk over this landscape shows that a probability distribution passing through the convex, negative-curvature part of $F$ is briefly but strongly broadened by spontaneous-emission noise, which is the microscopic origin of the jitter. The landscape is coupled to the photon-number history through cavity loss, producing the early-pulse-decays-early correlations that appear as anti-correlation lobes.","core_discovery":"The paper's central claim is that a quench through a photon-condensation threshold is characterized not only by growing number fluctuations but by a qualitatively different divergent fluctuation: timing jitter in the growth of the order parameter. On each realization the condensed pulse is seeded by spontaneous emission, so the instant of condensate formation fluctuates from shot to shot; near threshold these timing fluctuations grow and the ensemble-averaged pulse broadens even though individual pulses remain sharp. The authors show that this jitter is directly witnessed by the two-time non-stationary second-order correlation function $g^{(2)}(t_1,t_2)$: strong diagonal correlations at the inflection of the mean pulse and off-diagonal anti-correlations because an early pulse makes late detections less likely. They observe this signature experimentally in a dye microcavity, reproduce it with quantum trajectories that keep correlations to all orders, and reinterpret it through the geometry of an effective free-energy landscape, concluding that formation jitter is a general feature of transient second-order phase transitions in systems whose excitation reservoirs retain memory.","pith_inferences":["If the universality claim holds, the depth of the off-diagonal anti-correlation dip could be used to extract the curvature of the effective free energy near the threshold without knowing the microscopic rates.","The same two-time diagnostic could be applied to quenches in atomic Bose-Einstein condensates or spin systems where the order parameter is not directly measurable, testing whether formation jitter is generic beyond photonic systems.","A dedicated analysis separating pulse-shape effects from timing variance could yield a scaling exponent for the jitter, since the paper notes that the two competing effects hide a clean divergence in the raw $g^{(2)}$ maps."],"forward_implications":["Any quench through a second-order phase transition in a system whose excitation reservoir keeps memory of the photons should show the same qualitative signature: pulse broadening from formation jitter and anti-correlation lobes in $g^{(2)}(t_1,t_2)$.","The two-time correlation function becomes a practical diagnostic for transient critical phenomena, letting experimenters separate number fluctuations from formation-time fluctuations in a single measurement.","Micro- and nano-laser turn-on experiments, previously analyzed with single-time correlations, should be re-examined with two-time statistics to isolate formation jitter.","Colloidal nanoparticle growth, described by nucleation on a free-energy landscape, is predicted to exhibit an analogous formation-time jitter."],"supporting_citations":[{"why":"Supplies the microscopic master equation for photon-molecule dynamics from which the mean-field and correlation equations are derived.","marker":"[11-13]"},{"why":"Establishes the thermalized detailed-balance regime in dye microcavities that justifies drawing parallels to photon Bose-Einstein condensation.","marker":"[14]"},{"why":"Provides the geometric entropy-surface picture of a probability bubble evolving after a quench, the basis for interpreting fluctuations through free-energy geometry.","marker":"[2]"},{"why":"Defines an effective free energy for non-equilibrium systems whose gradient gives the order-parameter dynamics, used to build the landscape.","marker":"[5]"},{"why":"Gives the quantum regression theorem used to compute two-time correlations in a semi-analytic approximation.","marker":"[30]"},{"why":"Supplies the quantum trajectories or Monte Carlo wavefunction method that reproduces the measured $g^{(2)}(t_1,t_2)$ with correlations to all orders.","marker":"[31-34]"},{"why":"Provides the spontaneous-emission noise correlation used in the Langevin description of the free-energy walk.","marker":"[37,38]"}],"fun_headline_variants":["Timing jitter reveals transient phase transition universality","Photon condensation: formation jitter near threshold","Jitter in condensate birth: a critical slowdown signature","Transient condensation pulses arrive with random timing","Shot-to-shot jitter in photon condensate formation"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The result depends on the molecular excitation bath being in the fitted non-Markovian regime, where stimulated emission de-excites molecules faster than ordinary fluorescence; the paper states that in the Markovian limit both critical slowing down and timing jitter vanish.","fun_headline_variants_meta":{"raw":{"variants":["Timing jitter reveals transient phase transition universality","Photon condensation: formation jitter near threshold","Jitter in condensate birth: a critical slowdown signature","Transient condensation pulses arrive with random timing","Shot-to-shot jitter in photon condensate formation"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000302,"raw_usage":{"total_tokens":1802,"prompt_tokens":1071,"completion_tokens":731,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":687,"completion_tokens_details":{"reasoning_tokens":671}},"tokens_in":687,"tokens_out":731,"duration_ms":6672,"temperature":1.0,"reasoning_tokens":671,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:10:35.656535+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure $g^{(2)}(t_1,t_2)$ in the same cavity after increasing the molecular fluorescence rate $\\Gamma_\\downarrow$ relative to the cavity emission rate $E$ so that $\\Gamma_\\downarrow \\gg E$, pushing the bath toward Markovian behaviour; the model predicts that the off-diagonal anti-correlation lobes disappear and the near-threshold pulse broadening is strongly reduced, even though a delayed pulse remains.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes the thermalized detailed-balance regime in dye microcavities that justifies drawing parallels to photon Bose-Einstein condensation."},{"cited_title":"Jaynes, Frontiers of Nonequilibrium Statistical Physics (Springer, New York, 1986), pp","cited_arxiv_id":null,"evidence_quote":"Provides the geometric entropy-surface picture of a probability bubble evolving after a quench, the basis for interpreting fluctuations through free-energy geometry."},{"cited_title":"Chipot and A","cited_arxiv_id":null,"evidence_quote":"Defines an effective free energy for non-equilibrium systems whose gradient gives the order-parameter dynamics, used to build the landscape."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the quantum regression theorem used to compute two-time correlations in a semi-analytic approximation."}],"review_version":1}