{"id":"305ad674-de8a-4177-b58d-efc5e5cf5509","arxiv_id":"2511.08488","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"A new sufficient criterion certifies quantum non-Gaussianity from g2 and g3 alone: observing sqrt(g3) + 3 sqrt(g2) < 2 proves the state is not a mixture of Gaussian states; a quantum-dot source violates it by 0.174(13) vs 2.","lead":"The paper derives and tests a simple rule: if the measured second- and third-order photon correlations satisfy sqrt(g3) + 3 sqrt(g2) < 2, the light cannot be a mixture of Gaussian states, so it is quantum non-Gaussian. The rule is loss-tolerant, and the authors demonstrate it on a quantum-dot single-photon source with a margin of 0.174(13) versus the bound of 2.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Experimental certification assumes click statistics equal Glauber correlations; efficiency not quantified, so measured violation may not directly certify non-Gaussianity.","rationale":"The reader's weakest assumption correctly identifies that the experimental certification relies on click statistics matching Glauber correlations. This is the most load-bearing concern because it affects the experimental demonstration, a key part of the paper's central claim. The theoretical derivation of the inequality (18) appears correct: the pure-state bound follows from a valid polynomial inequality, and the mixed-state proof using Jensen and Cauchy-Schwarz is sound, including the trivial extension to g2≥4/9. The multi-mode extension is also plausible. The main unresolved issue is the unquantified assertion that the detection efficiency is 'sufficiently low' to match the Glauber definition. Since the paper acknowledges the potential discrepancy but does not provide quantitative support, the experimental claim remains conditional on this assumption. The proposed concrete test would settle whether the measured violation truly certifies quantum non-Gaussianity. Therefore, the reader's CONDITIONAL verdict remains appropriate, and no change is needed.","tokens_in":15501,"tokens_out":21006,"duration_ms":181603,"concrete_test":"Measure g^(2) and g^(3) for the same source under at least two different overall detection efficiencies (e.g., insert calibrated neutral density filters reducing the count rate by factors of 2 and 4) and verify that the extracted √g3 + 3√g2 remains constant within the reported statistical uncertainty (0.013). Alternatively, independently calibrate η and compute the leading-order correction to the click-based estimators for the measured count statistics; if the correction to √g3 + 3√g2 is below 0.013, the concern is resolved and the experimental demonstration stands.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central theoretical criterion (18) is derived for Glauber correlation functions g^(n) = <a†^n a^n>/<a†a>^n. The experimental demonstration extracts g^(2) and g^(3) from click counts of non-number-resolving detectors. For a detector with efficiency η, the click probability for n incident photons is 1-(1-η)^n, which is proportional to n only in the limit η→0. The paper asserts in the Experimental Validation section: 'In our experimental setup, the efficiency is sufficiently low to match the Glauber definition, which is the relevant quantity for our bound.' However, no quantitative efficiency value or error bound from the click-to-Glauber mapping is provided. If the efficiency is not sufficiently low, the measured normalized correlation functions deviate from the Glauber definitions, and the reported violation (√g3 + 3√g2 = 0.174(13)) would not directly certify quantum non-Gaussianity of the emitted state. This is load-bearing because the experimental violation is a headline result of the paper and the abstract reports it as confirmation of the criterion. The theoretical bound itself appears sound, so the concern is specifically about the experimental certification.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript derives a sufficient criterion for quantum non-Gaussianity based on the normalized intensity correlations g^(2) and g^(3). The central result is that every single-mode state that is a mixture of Gaussian states must satisfy sqrt(g^(3)) + 3 sqrt(g^(2)) >= 2, so observing the opposite inequality certifies quantum non-Gaussianity. The proof proceeds by deriving a quadratic bound for Gaussian pure states from closed-form moments, proving it by a polynomial inequality, and extending it to mixtures with Jensen and Cauchy-Schwarz inequalities; a multi-mode extension is given in the supplement. The authors apply the criterion to a quantum-dot single-photon source, reporting g^(2)=0.00334(4), g^(3)<=1.7e-4, and a violation of 0.174(13) < 2, with an extremely small p-value.","tokens_in":95,"tokens_out":7334,"duration_ms":105807,"significance":"If correct, this is a useful addition to the quantum non-Gaussianity toolbox: the criterion is parameter-free, uses only normalized nth-order correlations, and is therefore insensitive to attenuation and to finite detection efficiency in the idealized Glauber sense. The analytic proof is self-contained and appears sound; the multi-mode extension broadens its scope. The experimental demonstration shows a very large statistical separation, and the availability of data on Zenodo is a strength. The main caveat concerns the mapping from detector clicks to Glauber correlations, which is asserted but not quantified.","major_comments":[{"comment":"The claim that click statistics match Glauber correlations rests solely on the statement 'In our experimental setup, the efficiency is sufficiently low to match the Glauber definition.' No quantitative efficiency value, detector model, or systematic-error bound is given. For a non-number-resolving detector with efficiency η, the click probability for n incident photons is 1-(1-η)^n, which is proportional to n only in the limit η→0. Unless η is shown to be small enough (or a correction is applied and propagated), the measured coincidence counts may not equal <a†^n a^n>/<a†a>^n. Since the reported violation is a headline experimental result, this is load-bearing. Please provide an estimate or upper bound on η and a worst-case propagation to sqrt(g^(3))+3sqrt(g^(2)), or replace the heuristic statement with a calibrated click-to-Glauber analysis.","section":"Experimental validation (last paragraph)"}],"minor_comments":[{"comment":"The step from Eq. (14) to Eq. (15) is not logically equivalent when g^(2)>4/9; in that regime Eq. (15) holds trivially because 3√g^(2)>2. The text should state this explicitly so the reader does not infer that the square-root step is valid globally.","section":"Proof of the inequality for Gaussian mixed states"},{"comment":"The derivation of the boundary via Taylor expansion in r and solving for α² is heuristic. The subsequent rigorous proof is convincing, but the presentation should clearly separate the heuristic insight from the proof.","section":"Bound for Gaussian states, Eq. (10)"},{"comment":"The caption says 'The white region cannot be reached by any incoherent superposition of Gaussian states.' This is only true for g^(2)<4/9; the red region shows that mixtures can enter below the pure-state bound for g^(2)>4/9. Please add this qualification to avoid overstatement.","section":"Fig. 2 caption"},{"comment":"The phrase 'more than 100 standard deviations' is not by itself a well-defined confidence statement, especially since g^(3) is reported as an upper bound. The p-value test is the more rigorous statement; consider reporting a confidence interval for sqrt(g^(3))+3sqrt(g^(2)) instead of, or in addition to, the sigma language.","section":"Experimental validation and Conclusion"}],"recommendation":"major_revision","confidential_remarks":"The theoretical part of the manuscript is, in my reading, correct and likely publishable on its own: the derivation is parameter-free, the inequalities are used properly, and the multi-mode extension is reasonable. The experimental validation is the main weakness: the click-to-Glauber equivalence is asserted without a quantitative efficiency estimate. This is fixable in revision and does not undermine the central theoretical claim, so I recommend major revision rather than rejection. The paper fits the journal's scope; the related independent work by Filip et al. is acknowledged."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: the paper works. The inequality sqrt(g3) + 3 sqrt(g2) < 2 is a clean sufficient criterion for quantum non-Gaussianity, and the proof is solid. I checked the pure-state reduction to the polynomial inequality and the Jensen/Cauchy-Schwarz steps for mixtures; both are legitimate. The multi-mode extension via Bloch-Messiah is also reasonable, and the paper is honest about the g2 < 4/9 regime: a violation forces g2 < 4/9 automatically, so the abstract's 'any mixture' phrasing is actually defensible, though the main text's caveat is the better way to present it.\n\nThe experiment is convincing: g2 = 0.00334(4), g3 consistent with zero (upper bound 1.7e-4), and the p-value test against the Gaussian boundary is appropriately conservative. The added run with intentional laser leakage is a good sanity check that the three-fold correlations are not an artifact.\n\nThe real soft spot is experimental certification, not the theory. The measured quantities are multi-detector click coincidences, and the argument that these equal the Glauber g2 and g3 rests on a qualitative statement that the efficiency is 'sufficiently low'. No number or bound on the deviation is given. For a rigorous claim of 'unambiguous proof' from this particular dataset, a referee should ask for the efficiency or a direct estimate of the click-to-Glauber correction. Having said that, the violation is enormous (0.174 vs 2), so even a few-percent correction would not change the conclusion. This is a presentation/quantification gap, not a load-bearing flaw.\n\nThe relation to Filip's independent criteria [28,29] is acknowledged but not specified. Since those papers were posted this year, a referee should ask for a detailed comparison to establish exactly what is new. This is normal and should not delay things.\n\nWho this is for: anyone working on non-Gaussian state verification or single-photon source characterization. I'd cite it. It deserves serious peer review; send it out. The referee should request the efficiency estimate and the comparison, but the central result is sound.","headline":"A clean, well-proved g2/g3 witness for quantum non-Gaussianity with a convincing QD experiment; the main caveat is an unquantified detector-efficiency assumption in the experimental certification.","tokens_in":16252,"tokens_out":3794,"would_cite":true,"duration_ms":40384,"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":"This paper proves that for a single-mode light field, the inequality √g^(3) + 3√g^(2) < 2 certifies quantum non-Gaussianity — a state that cannot be any mixture of Gaussian states — and demonstrates the criterion on a quantum dot single-pho","keywords":["quantum non-Gaussianity","Gaussian states","photon correlations","non-classical light","single-photon source","quantum dot","continuous variables","loss-resistant witness"],"falsifier":"Find a statistical mixture of displaced-squeezed states with g^(2) < 4/9 whose g^(2) and g^(3) satisfy √g^(3) + 3√g^(2) < 2 — either by numerical search or by constructing an explicit counterexample such as a weighted mixture of a weakly squeezed displaced state with vacuum — and the paper's central bound would be refuted; the authors' proof asserts no such mixture exists.","tokens_in":15313,"feed_emoji":"⚛️","tokens_out":8549,"duration_ms":77411,"temperature":0.7,"pith_summary":"Quantum non-Gaussian states — those that cannot be written as mixtures of displaced and squeezed (Gaussian) states — are needed for quantum advantage in continuous-variable systems. This paper establishes a sufficient criterion for such states from only the normalized second- and third-order photon correlation functions g^(2) and g^(3): every mixture of Gaussian states must satisfy √g^(3) + 3√g^(2) ≥ 2 when g^(2) < 4/9, so violating that inequality certifies quantum non-Gaussianity. Because the bound involves only normalized correlations, the certification survives attenuation and finite detection efficiency, needing only longer measurement times. The authors demonstrate the test on a quantum dot single-photon source, obtaining √g^(3) + 3√g^(2) = 0.174(13), more than 100 standard deviations below the bound.","feed_headline":"One inequality with two photon correlations proves non-Gaussian light","feed_subtitle":"Certification survives attenuation and was verified on a quantum dot source at over 100 sigma.","key_machinery":"The load-bearing object is the normalized correlation combination √g^(3) + 3√g^(2), tested against the threshold 2. The authors show that for displaced-squeezed states the allowed region in the (g^(2), g^(3)) plane is bounded by g^(3) = (2 − 3√g^(2))^2; this curve is found by Taylor-expanding the moments to second order in the squeezing parameter r and taking the limit r → 0 while keeping the ratio with displacement fixed, then proved globally by an algebraic inequality. Wick's theorem (exact second-order cumulant expansion for Gaussian states) supplies the moment formulas, and the lifting from pure to mixed states uses Jensen's inequality and the Cauchy-Schwarz inequality. The measured quan","core_discovery":"The central claim is that the combination √g^(3) + 3√g^(2) < 2 is an unambiguous proof of quantum non-Gaussianity for a single-mode field, and by the supplementary multi-mode argument for Gaussian multi-mode fields as well. The proof proceeds by computing the second and third normally ordered moments of displaced-squeezed states with Wick's theorem, showing that Gaussian pure states lie on or above the curve g^(3) = (2 − 3√g^(2))^2, with the boundary reached in the limit of vanishing squeezing and vanishing displacement. Jensen's and Cauchy-Schwarz inequalities then extend the inequality to incoherent mixtures, provided g^(2) < 4/9, which is exactly the regime where the square root is meanin","pith_inferences":["Because the criterion is sufficient but not necessary, many non-Gaussian states will evade it; complementary witnesses or higher-order correlations will still be needed to certify the full class.","The g^(2) < 4/9 restriction is the true operational window: for larger g^(2) even mixtures of Gaussian states can dip below the pure-state curve, so experimental claims must report g^(2) alongside the combination.","The same ratio-symmetric structure suggests a family of higher-order witnesses (e.g., involving g^(4)) that could certify a larger set of non-Gaussian states, a direction the paper itself flags.","Applying the criterion at high detection efficiency would require number-resolving detectors or a careful calibration of click statistics to Glauber correlations, since non-number-resolving counters can otherwise mimic a violation."],"forward_implications":["Any experiment that records √g^(3) + 3√g^(2) < 2 with g^(2) < 4/9 gets a direct certificate of quantum non-Gaussianity, with no need for state tomography or a Wigner-function reconstruction.","The certification is inherently attenuation-resistant: losses only rescale acquisition time, so the same bound applies behind beam splitters, fibers, or low-efficiency detectors.","The test requires only a three-detector Hanbury Brown–Twiss setup, making it broadly applicable to single-photon sources, heralded states, and other non-Gaussian light sources.","The quantum dot demonstration reaches a combination value 0.174(13), more than 100σ below the bound, and the Gaussian-null p-value is 4·10^(−4793).","In the same framework, linear tangent versions of the bound yield simpler inequalities such as g^(3) + 3g^(2) < 1, and an additional criterion based on mean photon number and g^(2) can certify Fock states up to at least n = 1000."],"fun_headline_variants":["Two photon correlations prove non-Gaussian light","√g³ + 3√g² < 2 certifies quantum non-Gaussianity","New inequality exposes non-Gaussian quantum light","Quantum dot light violates classical bound at 100σ","Photon correlation test confirms quantum non-Gaussianity"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The certification holds if the measured coincidence counts equal the Glauber normally ordered correlation functions g^(2) and g^(3) of the field mode, which the paper assumes by operating at sufficiently low detection efficiency; at high efficiency with non-number-resolving detectors, click statistics can deviate from Glauber correlations.","fun_headline_variants_meta":{"raw":{"variants":["Two photon correlations prove non-Gaussian light","√g³ + 3√g² < 2 certifies quantum non-Gaussianity","New inequality exposes non-Gaussian quantum light","Quantum dot light violates classical bound at 100σ","Photon correlation test confirms quantum non-Gaussianity"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000188,"raw_usage":{"total_tokens":1168,"prompt_tokens":743,"completion_tokens":425,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":487,"completion_tokens_details":{"reasoning_tokens":344}},"tokens_in":487,"tokens_out":425,"duration_ms":4700,"temperature":1.0,"reasoning_tokens":344,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-03T22:49:20.635202+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Find a statistical mixture of displaced-squeezed states with g^(2) < 4/9 whose g^(2) and g^(3) satisfy √g^(3) + 3√g^(2) < 2 — either by numerical search or by constructing an explicit counterexample such as a weighted mixture of a weakly squeezed displaced state with vacuum — and the paper's central bound would be refuted; the authors' proof asserts no such mixture exists.","supporting_citations":[],"review_version":1}