{"id":"76237180-1959-4ad1-8beb-c8c3986cff79","arxiv_id":"1908.01521","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"For a β=0.1 nanolaser, intensity feedback shifts the lasing transition to lower pump values, narrows it, and produces irregular spiking, with g(2) correlations tracking the dynamics.","lead":"This paper simulates a nanoscale laser with some of its own light fed back into the cavity, and shows feedback lowers the pump needed for lasing and turns the output into irregular pulses. The results help chip designers understand stray reflections in tiny lasers and support using a simple autocorrelation measurement to diagnose them.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The large-feedback results (fext=0.3) are presented up to P/Pth=4 even though the model's own incoherent-feedback condition breaks down once g(2)(0) approaches 1 near P/Pth=2.5; this undermines the central claim unless results are restricted or the condition is quantitatively verified.","rationale":"The reader's weakest assumption is exactly the incoherent-feedback validity condition, and I agree that it is the load-bearing point. The paper is internally consistent and the stochastic simulation method is appropriate for the low-coherence regime, but the self-identified breakdown condition is reached within the stated pump range for fext=0.3. The primary claims about threshold reduction, the onset of a continuous low-coherence component, and the sufficiency of g(2) as a diagnostic are all drawn from parameter regions where the photon-number-only model may have already failed. This does not require rejection; it requires either a quantitative verification of the coherence-length condition across the full parameter grid or an explicit restriction of the conclusions to the valid region. Since the current conditional verdict already captures this uncertainty, no change to the reader's verdict is needed.","tokens_in":13827,"tokens_out":3762,"duration_ms":39132,"concrete_test":"For each (fext, P/Pth) point in Figs. 2, 3, and 7, estimate the field coherence time τc from the half-width of the rf spectrum or from the decay of g(2)(τ) in the free-running simulation at the same effective pump. Flag all points where τc ≥ 2τext (round-trip time, ≈2.3 ns for Lext=35 cm) as outside the incoherent-feedback regime. Recompute the input-output curves and zero-delay g(2) after excluding flagged points; if the fext=0.3 threshold shift and revival structure are concentrated in the excluded region, the central claim requires either a phase-resolving model (e.g., Lang-Kobayashi with noise) or a restricted pump range.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The modeling treats reinjection as an incoherent Poisson photon fraction, valid only when the external-cavity round-trip time exceeds the field coherence time. The paper itself states that 'when the Poisson regime is attained, the model is expected to break down' and warns that at upper pump values and sufficiently strong feedback 'the predictions may break down' (Sec. IV A). For fext=0.3, Fig. 3 shows g(2)(0) converging to 1 near P/Pth≈2.5, and Fig. 8 (for fext=0.1) is shown up to P/Pth=4, inside the pump range P≤4Pth used throughout. Since feedback lowers the effective threshold, the emitted field becomes more coherent at lower nominal pump than in the solitary case, so the condition 'round-trip longer than coherence length' is not automatically satisfied across the claimed parameter space. The threshold reduction, the appearance of a continuous low-coherence component at fext=0.3, and the validation of g(2) as a sufficient diagnostic all depend on dynamics in this potentially invalid region. The paper flags the caveat but does not delimit the affected results or verify the coherence-length condition quantitatively.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents a fully stochastic simulation of a β=0.1 nanolaser subject to incoherent intensity feedback, modeled by reinjecting a delayed fraction of the emitted photons into the cavity as a Poisson process. Using a photon-number recurrence scheme (Eqs. 1-4), the authors compute input-output curves, zero-delay and time-delayed second-order autocorrelations, temporal traces, and radiofrequency power spectra for feedback fractions fext = 0, 0.015, 0.1, and 0.3. They report that feedback shifts the input-output transition to lower pump values, shrinks the transition region, and at the largest feedback produces irregular spiking with a growing continuous, low-coherence component. They also show that external-cavity revival peaks in g(2)(τ) match periodic features in the rf spectra, and interpret this as validating the use of second-order autocorrelation as a sufficient diagnostic for nanolaser dynamics. The paper emphasizes the low-coherence transition regime and explicitly notes that the incoherent-feedback treatment is valid only while the external round-trip time exceeds the field coherence length, with breakdown expected once Poisson statistics are attained.","tokens_in":14109,"tokens_out":8461,"duration_ms":88191,"significance":"If the results hold, the paper provides a useful extension of stochastic laser modeling to nanolasers with feedback and offers practical guidance for experiments, namely that g(2)(τ) can serve as a simpler diagnostic than rf spectroscopy for identifying feedback-induced dynamics. The work is valuable for its explicit stochastic treatment of low-photon-number dynamics in the transition region and for comparing several observables from the same simulated output, which avoids any parameter fitting to a target result. The authors are also commendably explicit about the model's limitations, including the finite-sampling artifact below threshold and the expected breakdown in the Poissonian regime. The main strength is the internal consistency demonstrated among time traces, autocorrelations, and power spectra, which supports the qualitative interpretation of the simulations. The paper does not, however, provide an independent validation of g(2) as a universally sufficient tool, and its central conclusions partly rely on a parameter region where the model's own validity condition is questionable.","major_comments":[{"comment":"The paper states that 'when the Poisson regime is attained, the model is expected to break down' and that for the upper values of the pump range 'for sufficiently strong feedback, the predictions may break down,' but it does not delimit which results are affected. For fext=0.3, Fig. 3(d) shows g(2)(0) converging to 1 near P/Pth≈2.5, and Fig. 8 displays fext=0.1 dynamics up to P/Pth=4, where g(2)(0) has also reached the Poisson limit (Fig. 3(c)). Since several conclusions—threshold reduction, the appearance of a continuous component, and the 'noisy but certainly coherent emission at P=4Pth'—are drawn from these high-pump points, the authors should either restrict the analysis and claims to the region where the incoherent-reinjection assumption is justified (e.g., g(2)(0)>1) or provide a quantitative estimate of the emitted-field coherence time and verify that it remains longer than the external round-trip time across the full parameter range. As written, the presentation includes results in a region the authors themselves flag as invalid, which is load-bearing for the validity of the central claims.","section":"Sec. IV A, Figs. 3 and 8"},{"comment":"The claim that the paper 'validates the use of the second order autocorrelation as a sufficient tool for the interpretation of the dynamics' is stronger than what the evidence supports. The comparison between g(2)(τ), time traces, and rf spectra is performed on the same simulated photon-number output, so it demonstrates internal consistency of the stochastic model rather than an independent validation of g(2) as a diagnostic. The authors also concede in Sec. IV B that g(2) 'cannot distinguish between regular and irregular sequences,' which further weakens the word 'sufficient.' I recommend softening the abstract and conclusion to state that the results are 'consistent with' the usefulness of g(2) for capturing principal dynamical features, or providing an independent test (e.g., using an analytic relation or a different simulation method).","section":"Abstract and Sec. V"}],"minor_comments":[{"comment":"The sentence 'Sinj,q−d and Rinj,q−d is the fraction of back propagating photons which have probability 3×10−3 of being reinjected into the cavity' is unclear and appears inconsistent with the feedback fractions fext up to 0.3 used later. Please clarify how the feedback fraction defined in Eq. (6) is implemented in the stochastic algorithm.","section":"Sec. III"},{"comment":"When describing the delayed autocorrelation revivals at fext=0.3, the text refers to 'Fig. 5d' but the sharp revivals are shown in Fig. 6(d); the rf spectrum with strong peaks is Fig. 5(d). Please correct the cross-reference.","section":"Sec. IV C"},{"comment":"The below-threshold values of g(2)(0) are artificially close to 1 because of the finite-sampling artifact described in the text. The figure captions for Figs. 3 and 7 should explicitly note that below-threshold g(2)(0) values are not physically meaningful, so readers do not misinterpret them as indicating coherent statistics.","section":"Sec. IV A / Fig. 3 caption"},{"comment":"For reproducibility, the paper should state the simulation time step, the number of stochastic realizations, and the averaging time used for the steady-state and correlation estimates, rather than only citing Ref. [25].","section":"Sec. III"}],"recommendation":"major_revision","confidential_remarks":"The paper is within the scope of the journal and the stochastic simulation approach is appropriate for the low-photon-number regime. The main obstacle is the unresolved validity boundary: the authors disclose that the incoherent-feedback model breaks down once Poisson statistics are reached, yet they still present and interpret results in that regime. A major revision that either restricts the claims to the valid parameter region or adds a quantitative verification of the coherence-length condition would address the concern. The internal-consistency validation of g(2) is useful but should be framed more modestly."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This paper applies an existing stochastic recurrence scheme (Puccioni & Lippi) to a beta=0.1 nanolaser with intensity feedback, sweeping the feedback fraction from 0 to 0.3. The new content is the parameter-regime exploration: threshold shift to lower pump, shrinking transition region, spike regularization at weak feedback, and irregular spiking with a continuous low-coherence component at strong feedback. The authors also argue that the second-order autocorrelation captures the same features as rf spectra and time traces, validating g(2) as a sufficient diagnostic.\n\nWhat is done well: the stochastic model is appropriate for the low-photon-number regime; there is no parameter fitting to a target result; the authors openly acknowledge that finite sampling drives g(2)(0) to 1 below threshold and that the model breaks down once Poisson (coherent) statistics are attained. The qualitative comparison with the microlaser experiment (ref [23]) is reasonable.\n\nThe main soft spot, which the stress-test note correctly identifies, is that the model's validity condition is not quantitatively verified. The paper states that reinjection is incoherent only if the external-cavity round-trip exceeds the field coherence time, and later admits that 'for the upper values of this pump range, for sufficiently strong feedback, the predictions may break down.' Yet Fig. 3 shows g(2)(0) converging to 1 near P/Pth about 2.5 for fext=0.3, and Fig. 8 for fext=0.1 goes up to P/Pth=4. Since feedback lowers the effective threshold, the coherence length grows faster than nominal pump would suggest, so the incoherent-feedback assumption is not automatically satisfied across the whole claimed parameter space. The paper should either restrict the results to the region where the condition holds or verify it quantitatively (e.g., by comparing with a phase-resolved model). This is not a fatal flaw, because the core findings—threshold shift, transition shrinkage, spike regularization—are in the lower-pump region where the model is more credible, but the validation of g(2) as a sufficient tool is weakened if it relies on data in the possibly invalid regime.\n\nMinor issues: no error bars on the autocorrelation curves, and no code or data deposit, which limits reproducibility. A few figures would benefit from showing statistical spreads over multiple runs.\n\nWho is this for? Researchers working on nanolaser dynamics and feedback, especially experimentalists who use g(2) as a diagnostic. It deserves a serious referee. I would recommend conditional acceptance: ask the authors to delimit the valid parameter range, add error estimates, and make the simulation code available.","headline":"A useful but limited stochastic simulation study of incoherent feedback in a beta=0.1 nanolaser; the central caveat about model validity is stated but not fully delimited.","tokens_in":14651,"tokens_out":2758,"would_cite":false,"duration_ms":27291,"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":"For a β=0.1 nanolaser, incoherent intensity feedback lowers the lasing threshold, compresses the transition region, and at 30% feedback produces irregular spiking with a growing continuous component.","keywords":["nanolaser","intensity feedback","second-order autocorrelation","photon statistics","stochastic simulation","lasing transition","low-coherence regime","radiofrequency spectrum"],"falsifier":"Simulate the same β=0.1 parameters with a phase-resolved (coherent) feedback model: if the input-output curves or g(2)(0) differ by more than the statistical error of the stochastic runs at fext=0.3 and P/Pth≤2.5, then the incoherent assumption is load-bearing and the paper's central claim applies only inside that regime. Alternatively, shorten the external cavity until the round-trip time becomes comparable to the measured coherence time and look for coherent-feedback signatures such as a narrow spectral line or regular pulsing.","tokens_in":13665,"feed_emoji":"💡","tokens_out":5863,"duration_ms":59142,"temperature":0.7,"pith_summary":"Using a fully stochastic photon-number simulation, this paper asks what incoherent intensity feedback does to a β=0.1 nanolaser in the transition between spontaneous emission and lasing. It finds that reinjecting a fraction fext of the emitted photons into the cavity lowers the effective threshold and compresses the input-output curve, and at fext=0.3 the output becomes frequent irregular spikes sitting on a continuous, low-coherence background. The zero-delay second-order autocorrelation g(2)(0) captures this as a faster decay toward the Poisson value, while the delayed g(2)(τ) shows revival peaks at the external-cavity round-trip time. The paper concludes that autocorrelation measurements alone can characterize feedback dynamics in nanolasers.","feed_headline":"Feedback lowers nanolaser threshold and triggers irregular spikes","feed_subtitle":"Stochastic simulations show 30% feedback shifts the lasing transition to lower pump and confirms autocorrelation-based diagnostics.","key_machinery":"The engine is a recursive, fully stochastic photon-number simulator in which every process (pumping, spontaneous and stimulated emission, mirror losses) is drawn as a Poisson count, extended by a feedback term that reinjects a Poisson-selected fraction of the photons emitted one round-trip earlier. This is the central object because it keeps the intrinsic shot noise and non-Gaussian fluctuations that the paper argues are essential in the β=0.1 transition region. The second key diagnostic is the second-order autocorrelation function g(2)(τ), computed from the simulated time traces and used to judge the onset of coherence, the presence of external-cavity periodicity, and the emergence of a continuous component in the output.","core_discovery":"The central claim is that, for a β=0.1 nanolaser pumped through the lasing transition, incoherent optical feedback acts as an effective reduction of the lasing threshold and a narrowing of the transition region: as fext grows from 0 to 0.3, the input-output response shifts to lower pump values, the differential gain above threshold increases, and the pump value where g(2)(0) reaches the Poissonian limit moves from above 4 Pth to about 2.5 Pth. At the largest feedback studied, the temporal output develops a strongly spiking, irregular dynamics with a nonzero continuous background, and the delayed autocorrelation develops sharp revivals at the external-cavity round trip with internal structure at the relaxation-oscillation frequency. The paper further claims that these signatures are enough to identify the dynamics: the delayed g(2)(τ) reproduces, in a simpler way, the information contained in the rf power spectrum.","pith_inferences":["If the incoherent-feedback picture holds, the threshold shift seen here should also appear as a lowered pump for the onset of stimulated-emission bunching in measured g(2)(0) of any high-β device with a long external cavity; this is a direct consequence but not tested here.","A natural next step is to scan fext between 0.3 and 1: the model suggests a boundary where the continuous component becomes quasi-cw while spike statistics persist, and this boundary should appear as a qualitative change in the rf spectral slope.","The model's stated breakdown at coherence implies that at pump values below the current 4Pth but with strong feedback, one might observe a crossover to coherent-feedback effects (linewidth collapse, deterministic chaos) that the current photon-number recurrence cannot describe; identifying that crossover experimentally would delineate the validity domain.","Because g(2)(τ) cannot distinguish regular from irregular spike sequences, a practical extension is to combine g(2)(τ) with a spike-interval histogram in experiments; the histogram would provide the regularity information that autocorrelation misses."],"forward_implications":["In a β=0.1 nanolaser, raising the incoherent feedback fraction to 0.3 lowers the pump at which the autocorrelation reaches the Poissonian limit from above 4Pth to about 2.5Pth.","Large feedback (fext=0.3) changes the output at threshold from rare large spikes to frequent irregular spikes sitting on a continuous, low-coherence background.","The delayed autocorrelation g(2)(τ) shows external-cavity revivals with relaxation-oscillation side structure, matching the rf spectrum, so g(2)(τ) is a sufficient experimental diagnostic.","Compared with a mesoscale laser, the β=0.1 nanolaser needs a feedback level at least an order of magnitude larger to produce qualitatively similar dynamics, implying nanolasers tolerate more parasitic optical feedback without requiring optical isolation."],"supporting_citations":[{"why":"Defines the fully stochastic recurrence relations that are the basis of all simulations.","marker":"[25]"},{"why":"Provides the experimental microcavity results and the same incoherent feedback modeling that this nanolaser study extends.","marker":"[23]"},{"why":"Establishes the second-order autocorrelation as the experimental probe of nanolaser photon statistics.","marker":"[1]"},{"why":"Supplies the interpretation that g(2)(τ) tracks the delayed buildup of coherence after threshold.","marker":"[42]"},{"why":"Defines the normalized threshold pump Pth used throughout the paper.","marker":"[43]"},{"why":"Gives the estimate of cw emission onset near 5 Pth that justifies the incoherent-feedback assumption.","marker":"[47]"},{"why":"Documents irregular spiking in mesoscale lasers, the baseline behavior that feedback modifies.","marker":"[48]"},{"why":"Supports the idea that feedback displaces the laser response toward lower pump values.","marker":"[44]"}],"fun_headline_variants":["Feedback lowers nanolaser threshold, narrows transition","High feedback makes nanolaser output spike irregularly","Autocorrelation captures feedback dynamics in nanolasers","Feedback sharpens nanolaser transition and gives g(2) revivals"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole picture rests on the assumption that the reflected light is reinjected as pure noise-like photon counts with no phase or wave interference, which the paper says is valid only when the external-cavity round trip is longer than the coherence length and the emission is not yet coherent; if that condition fails, the predicted threshold shift and correlation revivals could cease to apply.","fun_headline_variants_meta":{"raw":{"variants":["Feedback lowers nanolaser threshold, narrows transition","High feedback makes nanolaser output spike irregularly","Autocorrelation captures feedback dynamics in nanolasers","Feedback sharpens nanolaser transition and gives g(2) revivals"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000731,"raw_usage":{"total_tokens":3233,"prompt_tokens":867,"completion_tokens":2366,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":483,"completion_tokens_details":{"reasoning_tokens":2298}},"tokens_in":483,"tokens_out":2366,"duration_ms":17389,"temperature":1.0,"reasoning_tokens":2298,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T15:09:44.968036+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Simulate the same β=0.1 parameters with a phase-resolved (coherent) feedback model: if the input-output curves or g(2)(0) differ by more than the statistical error of the stochastic runs at fext=0.3 and P/Pth≤2.5, then the incoherent assumption is load-bearing and the paper's central claim applies only inside that regime. Alternatively, shorten the external cavity until the round-trip time becomes comparable to the measured coherence time and look for coherent-feedback signatures such as a narrow spectral line or regular pulsing.","supporting_citations":[{"cited_title":"Stochastic simulator for modelling the transition to lasing,","cited_arxiv_id":null,"evidence_quote":"Defines the fully stochastic recurrence relations that are the basis of all simulations."},{"cited_title":"Dynamics of a micro-VCSEL operated in the threshold region under low-level optical feedback,","cited_arxiv_id":null,"evidence_quote":"Provides the experimental microcavity results and the same incoherent feedback modeling that this nanolaser study extends."},{"cited_title":"Direct observation of correlations between individual photon emission events of a microcavity laser,","cited_arxiv_id":null,"evidence_quote":"Establishes the second-order autocorrelation as the experimental probe of nanolaser photon statistics."},{"cited_title":"Delayed formation of coherence in the emission dynamics of high- Q nanolasers ,","cited_arxiv_id":null,"evidence_quote":"Supplies the interpretation that g(2)(τ) tracks the delayed buildup of coherence after threshold."},{"cited_title":"Photon statistics of a cavity-QED laser: a comment on the laser-phase- transition analogy,","cited_arxiv_id":null,"evidence_quote":"Defines the normalized threshold pump Pth used throughout the paper."},{"cited_title":"Onset of lasing in small devices: the identification of the first threshold through autocorrelation resonance,","cited_arxiv_id":null,"evidence_quote":"Gives the estimate of cw emission onset near 5 Pth that justifies the incoherent-feedback assumption."},{"cited_title":"Dynamical Buildup of Lasing in Mesoscale Devices,","cited_arxiv_id":null,"evidence_quote":"Documents irregular spiking in mesoscale lasers, the baseline behavior that feedback modifies."},{"cited_title":"Low- frequency fluctuations and polarization dynamics in vertical-cavity surface-emitting lasers with isotropic feedback,","cited_arxiv_id":null,"evidence_quote":"Supports the idea that feedback displaces the laser response toward lower pump values."}],"review_version":1}