{"id":"7d2340d4-ea54-4a1e-9841-b48a290931d4","arxiv_id":"2501.10044","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Frequency-comb spectral interferometry achieves 0.67 nm single-shot precision at 25 microseconds and 4.5e-12 m/Hz^1/2 sensitivity, close to the detector shot-noise limit.","lead":"The team used a special laser (a frequency comb) to measure distances with sub-nanometer precision at high speed, close to the physical limit set by light noise. This could make absolute distance measurements as precise as traditional laser interferometers, useful for next-generation length standards.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The quantum-limit comparison rests on a simulation-derived conversion factor G=7e-13 m² that is only calibrated for spectrally uniform white noise, while the real per-mode RIN is demonstrably non-uniform (Fig. S5); if G is shape-dependent, the 'close to shot noise' claim is not established.","rationale":"The reader's weakest assumption correctly identifies the conversion factor G as the load-bearing step in the quantum-limit claim. The paper's direct precision result is well supported: the noise-injection simulation reproduces the measured distance-noise spectrum and Allan deviation, and the length-dependent sensitivity trend is consistent with the proposed intensity-plus-frequency-noise model. What is not independently established is that the shot-noise-limited sensitivity is 3.4-3.7×10⁻¹² m/Hz^1/2, because G is calibrated only against spectrally flat white noise and the actual comb RIN is demonstrably non-uniform. The appendix-level limitation appended to the Methods acknowledges that the uniform-white-noise model is unsuitable for real noise, which makes the reliance on that same model for the quantum-limit number particularly fragile. Since the paper is already CONDITIONAL, the stress-test does not move the verdict; rather, it sharpens the condition that should be imposed: either provide the simulation/raw data needed to re-derive G for the actual RIN spectrum, or re-state the quantum-limit comparison with an explicit uncertainty budget for G. No ad hominem or external-consensus objection is being raised; this is an internal calibration-validity concern that can be settled by a concrete numerical reproduction.","tokens_in":15475,"tokens_out":6523,"duration_ms":70656,"concrete_test":"Independently re-implement the Appendix G simulation using the measured non-uniform per-mode RIN spectrum from Fig. S5a, together with the exact 10th-order super-Gaussian window, visibility, and polynomial peak-fit routine, and compute the effective conversion factor as S_distance(f)/S_RIN(f) after accounting for the known spectral weighting. If the resulting G deviates by more than about 30% from the stated 7×10⁻¹³ m², the quoted shot-noise-limited sensitivity should be recalculated and the 'close to quantum limit' conclusion reconsidered. As a secondary check, recompute the shot-noise ASD from the stated V_CCD = 370 mV, conversion efficiency 128 nV/e⁻, V = 0.6, and bandwidth 20 kHz to resolve the 3.4×10⁻¹² vs 3.7×10⁻¹² m/Hz^1/2 discrepancy in the paper.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central quantum-limit claim is not backed by a direct measurement; it relies on the chain S_distance_shot = G * S_RIN_shot, with G = 7e-13 m² obtained from the Appendix G simulation. That simulation injects spectrally uniform white noise into an ideal interference signal with V = 0.6 and then reads out the distance PSD. However, Appendix E and Fig. S5 show that the actual EO-comb RIN is strongly non-uniform over the comb spectrum, with some spectral regions well above the shot-noise level. The subsequent 10th-order super-Gaussian window and polynomial peak fit do not weight all modes equally, so a single scalar G calibrated on flat white noise need not describe the response to the real, structured noise spectrum. The Methods section even states that the uniform-white-noise model is 'not suitable for actual distance measurements,' yet it is the only calibration used to quote the quantum-limited sensitivity. No code or raw data are provided to test the robustness of G. There is also a minor internal inconsistency in the quoted shot-noise ASD: 3.7×10⁻¹² m/Hz^1/2 in the Results versus 3.4×10⁻¹² m/Hz^1/2 in the Methods. If recalibration with the actual non-uniform spectrum changed G by more than about 25%, the measured 4.5×10⁻¹² m/Hz^1/2 would no longer be 'close' to the shot-noise limit, and the headline conclusion would need qualification. The direct precision measurement at 0.67 nm and the agreement of the noise-injection prediction with the measured noise spectrum are convincing for the reported precision itself; the vulnerable part is specifically the quantum-limit comparison.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports a frequency-comb-based spectral interferometry system for absolute distance measurements, using a spectrally flat electro-optic comb and a high-speed spectrometer. The authors demonstrate a measurement precision of 0.67 nm at 25 us averaging time, an amplitude spectral density of 4.5e-12 m/Hz^1/2 above 1 kHz, claim this is close to the shot-noise (quantum) limit, and support this with a noise model that separates intensity-noise and frequency-noise contributions. They validate the model by injecting the measured RIN of the same comb into the interferogram and comparing the predicted and measured distance noise, and by analyzing the distance dependence of the sensitivity. Practical demonstrations include acoustic-wave-induced vibration measurements and voice recording through the interferometer.","tokens_in":15845,"tokens_out":5852,"duration_ms":55719,"significance":"If the central claim is established, the paper represents a significant advance in absolute distance metrology: it combines high update rate (40 kHz) with nanometric precision and provides a quantitative noise budget that identifies intensity noise as the dominant limit at short distances and frequency noise at longer distances. The direct Allan deviation and amplitude spectral density measurements are credible, and the noise-injection simulation reproduces the measured noise spectrum, which is a strong internal consistency check. The acoustic and voice-sensing demonstrations show practical utility. However, the 'close to the quantum limit' claim depends on a simulation-derived conversion factor and on internally inconsistent numerical values for the shot-noise floor, so the quantitative conclusion needs strengthening.","major_comments":[{"comment":"The paper quotes three different values for the shot-noise-limited amplitude spectral density: 3.2e-12 m/Hz^1/2 in the Introduction, 3.7e-12 m/Hz^1/2 in the Results (Fig. 2B and accompanying text), and 3.4e-12 m/Hz^1/2 in the Methods section. Because the central claim is that the measured 4.5e-12 m/Hz^1/2 is close to this limit, the reference value must be unique and consistent; please reconcile these numbers and clarify which value is the final shot-noise floor.","section":"Introduction / Results / Methods (shot-noise calculation)"},{"comment":"The conversion factor G = 7e-13 m^2 is obtained from a simulation that injects uniform white noise into an ideal interference signal with V = 0.6, and no uncertainty or sensitivity analysis is provided. The Methods even notes that the uniform-white-noise model 'is not suitable for actual distance measurements.' Since G is used to convert the shot-noise RIN into the shot-noise-limited distance sensitivity, an error in G propagates directly into the quantum-limit claim. Please provide a robustness analysis (e.g., G as a function of visibility, window order, bandwidth, and noise level) and justify why the uniform-noise calibration is valid for shot noise despite the non-uniform RIN observed in Appendix E.","section":"Methods ('Calculation of the shot-noise-limited distance sensitivity') and Appendix G"},{"comment":"The agreement between the measured distance noise and the prediction obtained by injecting the measured RIN of the same comb into the same interferogram is a valuable consistency check, but it shows that the measured intensity noise is the dominant contributor, not that the precision is at a fundamental quantum limit. The paper should explicitly distinguish the technical-noise floor (which includes RIN above shot noise, as in Fig. S5) from the shot-noise floor, and state how far the current technical noise sits above the shot-noise floor.","section":"Results ('High-precision and rapid distance measurements') and Discussion"}],"minor_comments":[{"comment":"The text refers to the measured distance line as 'blue' while the figure caption and legend describe it as 'black'; please make the colors consistent throughout the figure and text.","section":"Fig. 2A and main text"},{"comment":"The second term is typeset as '(4πIoVL/v)2 2⁄', which is ambiguous; the factor of 1/2 is unclear. Please typeset the equation clearly, e.g., (4πIoVL/v)^2 / 2.","section":"Equation (3)"},{"comment":"The 'Precision (1 σ)' entry for this work is 0.33 nm at τ_avg = 250 μs, while the main text reports 0.34 nm at the same averaging time; please harmonize these values.","section":"Table S1"},{"comment":"There is a typo: 'conversion efficient' should be 'conversion efficiency'.","section":"Methods ('Shot-noise calculation')"},{"comment":"For reproducibility of the G-factor simulation, please consider providing the simulation code or a detailed pseudocode, since the current statement only says additional data may be requested.","section":"Data and materials availability"}],"recommendation":"major_revision","confidential_remarks":"The paper is experimentally strong and the noise-injection validation is convincing, but the headline claim of being 'close to the quantum limit' is weakened by the inconsistent shot-noise ASD values (3.2, 3.4, and 3.7e-12 m/Hz^1/2) and by the lack of a sensitivity analysis for the simulation-derived conversion factor G. These are fixable with additional analysis and careful rewriting, so I recommend major revision rather than rejection. The editor may also wish to ask the authors to soften the 'quantum-limited' language in the abstract until the numerical consistency is resolved."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: the 0.67 nm precision at 25 µs with a 40 kHz update rate is the real story here, and the noise-injection model matches the data well. The \"close to the quantum limit\" label is weaker than the headline suggests.\n\nWhat's genuinely good: the measured Allan deviation, the amplitude spectral density, and the agreement between the measured distance noise and the prediction built by injecting the measured RIN back into a single interferogram. That is a convincing consistency check for intensity noise as the dominant short-distance limit. The improvement over the authors' own prior EO-comb work (6 nm at 25 µs) is about an order of magnitude, and the comparison to other ranging methods in Fig. 5 is fair. The sound-vibration demos are nice but not load-bearing.\n\nWhere I'd push back: the \"quantum-limited\" claim depends on a scalar conversion factor G = 7e-13 m², obtained from a simulation that injects spectrally uniform white noise. The paper's own Fig. S5 shows the real per-mode RIN is not flat; some spectral regions sit above shot noise. The Methods even says the uniform-white-noise model is \"not suitable for actual distance measurements,\" yet that same model is the only calibration behind the shot-noise limit. If G is shape-dependent, the measured 4.5e-12 m/Hz^1/2 being \"close to\" 3.7e-12 (or 3.4e-12, the paper quotes both) could easily shift outside the claimed margin. Also, the short-distance prediction uses the measured noise floor as an input, so the agreement at 100 mm is partly a consistency check, not an independent test of the quantum limit. No code or raw data are deposited, which makes it harder to independently verify G.\n\nNone of this breaks the direct precision result: 0.67 nm at 25 µs comes straight from Allan deviation. But the quantum-limit comparison needs a recalibration against the actual non-uniform RIN spectrum before I'd trust it without qualification.\n\nFor a reader: this paper is for people in frequency-comb ranging and length metrology. It deserves a serious referee, and I'd want the authors to either measure G directly or bound how strongly spectral shape changes it. I'd probably not cite the quantum-limit claim as it stands, but the precision numbers and the noise-origin analysis are worth keeping.","headline":"A real precision advance in EO-comb spectral interferometry, but the quantum-limit claim rests on a white-noise calibration the paper itself says is not suitable for actual measurements.","tokens_in":16393,"tokens_out":2107,"would_cite":true,"duration_ms":22828,"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":"Frequency-comb spectral interferometry reaches 0.67-nm precision at 40 kHz, close to the shot-noise limit, with intensity noise as the short-range floor.","keywords":["frequency comb","spectral interferometry","absolute distance measurement","shot-noise limit","electro-optic comb","Allan deviation","length metrology","vibration sensing"],"falsifier":"Re-run the shot-noise-limit calculation using the measured per-pixel relative intensity noise from Appendix E instead of uniform white noise, and compare the predicted distance amplitude spectral density with the measured $4.5\\times10^{-12}$ m/Hz$^{1/2}$ at 100 mm; if the prediction departs by more than about 20 percent, the closeness to the quantum limit is an artifact of the uniform-noise conversion factor. A direct experimental check is to place a mirror on a calibrated piezo stage at 100 mm and verify that the white-noise Allan deviation follows $3.2$ pm$\\cdot\\tau^{-1/2}$ as the detector thermal noise is reduced.","tokens_in":15284,"feed_emoji":"📏","tokens_out":11678,"duration_ms":105021,"temperature":0.7,"pith_summary":"This paper claims that frequency-comb-based spectral interferometry can measure absolute distance with sub-nanometer precision at high speed, and that the precision floor is set by fundamental light-source and detector noise rather than by the fringe-analysis algorithm. Using a spectrally flat electro-optic comb and a 40 kHz spectrometer, the authors report a precision of 0.67 nm at a 25 μs averaging time and a sensitivity of $4.5\\times10^{-12}$ m/Hz$^{1/2}$, close to the computed shot-noise floor. They argue that intensity noise is the fundamental limit at short distances and that frequency noise becomes gradually dominant as the target distance grows. If correct, this would make comb-based spectral interferometry a credible replacement for laser displacement interferometry in next-generation length standards, while also enabling real-time sensing of vibrations and sound.","feed_headline":"Comb length sensor: 0.67 nm at 40 kHz, near shot-noise limit","feed_subtitle":"Near the shot-noise floor, it tracks absolute distances at 40 kHz—fast enough to record sound and vibration","key_machinery":"The load-bearing object is the spectral interferogram produced by a spectrally flat electro-optic frequency comb with an 18 GHz mode spacing and about 10 THz bandwidth. The round-trip delay $\\tau_{TOF}=2L/v$ appears as a fringe period $1/\\tau_{TOF}$ in the frequency domain, so a Fourier transform followed by peak detection converts the spectrum into a distance readout. The noise analysis is carried by Eq. (3), which splits interferogram fluctuations into an intensity-noise term $(1+V^2/2)(\\Delta I_o(f_i,t))^2$ and a frequency-noise term $(4\\pi I_o V L/v)^2(\\Delta\\delta f_i(t))^2/2$. A simulation supplies the conversion factor $G=7\\times10^{-13}$ m$^2$ that maps a uniform white relative-intensity-noise power spectral density $S_{white}(f)$ into a distance power spectral density $S_{distance}(f)=G\\,S_{white}(f)$, and the shot-noise floor is estimated from the CCD electron count, giving $S_{distance,shot}=1.2\\times10^{-23}$ m$^2$/Hz, or $3.4\\times10^{-12}$ m/Hz$^{1/2}$. The measurement pipeline uses a tenth-order super-Gaussian window over a 9 THz bandwidth and polynomial fitting of the reconstructed peak.","core_discovery":"The central claim is that intensity noise, not the data-processing algorithm, is the limiting noise source of frequency-comb spectral interferometry at short distances, with frequency noise taking over at longer distances. In the authors' model, the spectral interference signal is $I(f_i,t)=I_o(f_i,t)\\{1+V\\cos(2\\pi(f_i+\\delta f_i(t))2L/v)\\}$, and its fluctuation separates into an intensity-noise term and a frequency-noise term proportional to $L$. Experimentally, at a target distance of about 100 mm, the measured distance sensitivity was $4.5\\times10^{-12}$ m/Hz$^{1/2}$ above 1 kHz, close to the quantum-limited (shot-noise-limited) value, and the Allan deviation was 0.67 nm without averaging and 0.34 nm at 250 μs; with one-second averaging in a stable environment the precision reaches about 3.2 pm. The white-noise sensitivity grows as $\\sqrt{(4.5\\times10^{-12})^2+(L\\cdot1.2\\times10^{-11})^2}$ m/Hz$^{1/2}$ for target distances from 100 mm to 1000 mm, confirming the predicted transition from intensity-noise to frequency-noise limitation.","pith_inferences":["Beyond the paper: if the shot-noise floor is real, further precision gains at short range depend on lowering the relative intensity noise of the comb and the detector in the high-noise spectral regions, not on better peak-fitting algorithms.","Beyond the paper: the same two-term intensity and frequency noise model could be used to predict the precision limits of other comb sources, such as mode-locked fiber combs or microcombs, directly from their measured relative intensity noise and frequency noise.","Beyond the paper: the demonstrated voice and sound sensing suggests the technique could serve as a traceable optomechanical microphone, converting acoustic pressure into calibrated length measurements.","Beyond the paper: the empirical conversion factor $G$ could be cross-checked against an information-theoretic bound on peak-position estimation, which would generalize the quantum-limit prediction to arbitrary comb spectra and noise color."],"forward_implications":["At a 40 kHz update rate with 0.67 nm precision without averaging, the method can track fast dynamic motion such as acoustic-wave-induced vibration and laser eavesdropping in real time.","In the white-noise-limited regime the Allan deviation follows $3.2$ pm$\\cdot\\tau^{-1/2}$, so one-second averaging reaches roughly 3.2 pm in a stable environment, comparable to laser displacement interferometry without accumulated displacement.","Because the distance comes from the spectral fringe period, the measurement is absolute and free of the $2\\pi$ ambiguity that constrains single-wavelength displacement interferometers.","Combining this interferometer with AMCW ranging for coarse initialization gives absolute distance measurements beyond the 4.2 mm non-ambiguity range of the comb interferometer itself.","At long distances the frequency-noise contribution scales linearly with $L$, so reaching the quantum limit at long range will require reducing the seed-laser frequency noise rather than only improving the detector."],"supporting_citations":[{"why":"Supplies the programmable spectral shaping and super-Gaussian window used for the nanometric precision, and provides the previous precision benchmark this work improves.","marker":"(35)"},{"why":"Introduces the comb-mode-resolved spectral-domain interferometer with an electro-optic comb, the platform on which the quantum-limited precision claim is made.","marker":"(33)"},{"why":"Supplies the AMCW ranging method used to determine the coarse initial distance, resolving the 4.2 mm non-ambiguity range of the spectral interferometer.","marker":"(41)"},{"why":"Provides the chart used to convert noise power spectral densities into Allan deviations and to estimate the integrated frequency noise at a 25 μs measurement time.","marker":"(42)"},{"why":"Supports the claim that frequency noise becomes the dominant precision limit as target distance increases, the long-distance half of the paper's central hypothesis.","marker":"(17)"},{"why":"Supports the assumption that a frequency comb locked to a frequency standard has low enough frequency uncertainty for intensity noise to dominate at short distances.","marker":"(36)"},{"why":"Serves as a state-of-the-art dual-comb ranging comparison in the precision-versus-update-rate benchmark that positions the reported performance.","marker":"(14)"},{"why":"Provides the faster MHz-acquisition-rate comparison method in the benchmark against which the paper's 40 kHz update rate and sub-nm precision are judged.","marker":"(23)"}],"fun_headline_variants":["Comb interferometry nears quantum limit for length","Sub-nm length precision close to shot-noise limit","Fast comb length sensor: 0.67 nm at 40 kHz","Quantum-limited distance measurement with frequency combs"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The shot-noise-limited sensitivity is computed with a simulation that assumes uniform white noise across the spectrum and a linear conversion factor $G=7\\times10^{-13}$ m$^2$; the paper itself notes in Methods that this uniform-noise model is not suitable for actual distance measurements because real intensity noise varies from comb mode to comb mode. If that conversion factor misrepresents the real non-uniform noise, the claim that the measured $4.5\\times10^{-12}$ m/Hz$^{1/2}$ is close to the quantum limit is not established.","fun_headline_variants_meta":{"raw":{"variants":["Comb interferometry nears quantum limit for length","Sub-nm length precision close to shot-noise limit","Fast comb length sensor: 0.67 nm at 40 kHz","Quantum-limited distance measurement with frequency combs"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000315,"raw_usage":{"total_tokens":1795,"prompt_tokens":964,"completion_tokens":831,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":580,"completion_tokens_details":{"reasoning_tokens":767}},"tokens_in":580,"tokens_out":831,"duration_ms":8109,"temperature":1.0,"reasoning_tokens":767,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T19:23:45.812968+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Re-run the shot-noise-limit calculation using the measured per-pixel relative intensity noise from Appendix E instead of uniform white noise, and compare the predicted distance amplitude spectral density with the measured $4.5\\times10^{-12}$ m/Hz$^{1/2}$ at 100 mm; if the prediction departs by more than about 20 percent, the closeness to the quantum limit is an artifact of the uniform-noise conversion factor. A direct experimental check is to place a mirror on a calibrated piezo stage at 100 mm and verify that the white-noise Allan deviation follows $3.2$ pm$\\cdot\\tau^{-1/2}$ as the detector thermal noise is reduced.","supporting_citations":[],"review_version":1}