{"id":"0c20c0b1-9460-483a-901f-874e6d99e7e2","arxiv_id":"2502.05168","paper_version":3,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Frequency-dependent squeezed readout lowers the impulse detection threshold as e^{-r}, down to a damping-limited floor of Δp_SQL/√Q; losses soften the gain to e^{-r/2}.","lead":"The authors calculate how much squeezed light can improve the sensitivity of mechanical detectors to tiny momentum kicks, and show that frequency-dependent squeezing can beat the standard quantum limit. They find a fundamental floor set by mechanical damping, and optical losses weaken the squeezing benefit.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified: Eq. (27) and Eq. (32) follow from the stated model, and the flagged assumptions are explicit scope conditions.","rationale":"The reader's verdict of ACCEPT with HIGH confidence is appropriate. I traced the main derivation: the force PSD after optimal frequency-dependent squeezing (Eq. 24), the large-coupling expansion (Eq. 26), the integral leading to Eq. (27), and the large-squeezing analysis leading to Eq. (32). In the exponential regime Q >> e^{2r}, choosing g >> e^r g_*(ωm) makes the unsqueezable resonance term e^{2r} m ωm^2 / (2Q g̃^2) small relative to the squeezed backaction term e^{-2r} m ωm^2 g̃^2 / (2Q), and makes the FDT floor m γ ωm small relative to both; the resonance integral then yields Δp ≈ e^{-r} Δp_SQL. When r reaches roughly (1/2) ln Q, no g can satisfy both inequalities, and the floor and unsqueezable resonance term become O(1) corrections, producing the plateau at Δp_SQL / sqrt(Q). This is internally consistent. The reader's weakest assumption (zero-temperature FDT floor and 1D Markov f = 0) is the right place to look, but the paper explicitly states both: App. C derives the floor and App. D flags the f = 0 simplification and the 3D difference. No circularity or missing step was found. A direct numerical check of the exact PSD integral would further harden the result, but I do not see a basis for changing the verdict.","tokens_in":23050,"tokens_out":37134,"duration_ms":357824,"concrete_test":"Symbolically or numerically integrate the exact force PSD of Eq. (24) (including the mγ|ν| term) over ν for Q = 10^4 to 10^8 and r = 0 to 10, optimizing g at each r, and verify that the minimum tracks e^{-r} Δp_SQL for e^{2r} << Q and plateaus at Δp_SQL / sqrt(Q) as r approaches (1/2) ln Q; this directly tests Eqs. (27) and (32).","verdict_should_be":"UNCHANGED","load_bearing_attack":"I cannot identify a load-bearing flaw in the central argument. The e^{-r} scaling of Eq. (27) is derived in the regime e^r g_*(ωm) << g << sqrt(Q) g_*(ωm), where the unsqueezable on-resonance contribution and the mγ|ν| FDT floor are subdominant; the plateau of Eq. (32) appears when e^{2r} ~ Q, where those same terms set the floor. The claimed Q-limited floor follows directly from the zero-temperature fluctuation-dissipation noise S_FF^QN = mγν derived in App. C, and the paper explicitly sets this 'throughout.' The f = 0 Markov approximation of App. D is stated as a simplifying constraint, with the 3D forward-scattering difference noted in the appendix and the outlook. These are scope conditions, not hidden assumptions. The remaining substantive risk is external validity to real finite-temperature, 3D levitated sensors, but this does not undermine the theoretical claim as stated.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper analyzes impulse (delta-function force) sensing with harmonically suspended optomechanical detectors, using frequency-dependent squeezed readout. It derives the momentum threshold scaling Δp ~ e^{-r} in an intermediate-coupling window, a Q-limited floor Δp_min ≈ Δp_SQL/√Q for lossless measurement, and loss-modified scalings e^{-r/2} for small detection efficiency. The model is developed for a Fabry-Pérot cavity in the bad-cavity limit and a 1D dielectric slab, with the mapping between them established in App. D.","tokens_in":23186,"tokens_out":9923,"duration_ms":92554,"significance":"The e^{-r} scaling and the Q-floor are concrete, falsifiable predictions with direct relevance to levitated optomechanical sensors. The paper is careful to state scope conditions (zero-temperature FDT floor, Markov f=0 approximation), and the central analytical results are cross-checked numerically in Figs. 6 and 7(b). The connection to known limits on dissipative measurements grounds the result in the existing literature.","major_comments":[],"minor_comments":[{"comment":"Equation (3) has a typo: the right-hand side should sum over input operators Oin_j, not output operators Oout_j; as written the relation is circular.","section":"II A, Eq. (3)"},{"comment":"The derivation of the central floor Δp_min ≈ Δp_SQL/√Q is compressed into a single sentence (\"one can fully do the integration...\"). Please provide the explicit integral (or a supplementary appendix) so that the large-r expansion and the condition e^{2r} ~ Q can be checked; the numerical check in Fig. 6 is reassuring but does not replace the analytic derivation.","section":"III B, Eq. (32)"},{"comment":"In the sentence following Eq. (42), \"the shot noise is dominated by the losses, which may be mitigated by increasing the laser power, which in turn may be mitigated by squeezing the back-action\" is confusing; suggest rewording to clarify that increasing power raises back-action, which is then reduced by squeezing.","section":"IV B, Eq. (42)"},{"comment":"There are several typos: \"show noise\" should be \"shot noise\" in Sec. I; \"diectric\" should be \"dielectric\" in Sec. II A; \"loser power\" should be \"laser power\" in the Fig. 2 caption; \"derive derive\" appears in App. B; \"dicussions\" should be \"discussions\" in the Acknowledgements.","section":"I, II A, Fig. 2, App. B, Acknowledgements"},{"comment":"In Eq. (31), state explicitly that the plateau condition is e^{2r} ~ Q, which clarifies the break-down of the large-Q expansion.","section":"III B, Eq. (31)"}],"recommendation":"minor_revision","confidential_remarks":"The manuscript is a solid theoretical contribution that fits the journal's scope. The minor typographical issues and the compressed derivation of Eq. (32) are easily addressed in a revision; I do not see a need for new experiments or extensive rewriting."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper delivers what it promises: analytic scaling laws for the momentum threshold under frequency-dependent squeezing in harmonically suspended optomechanical sensors. The main new results are the e^{-r} reduction of the threshold in the regime e^r g_*(ωm) << g << sqrt(Q) g_*(ωm), the lossless floor Δp_SQL/√Q, and the loss-induced crossover to e^{-r/2}. These are derived from the stated model rather than fitted, and the numerics confirm the asymptotics. The derivations in the appendices are explicit and checkable; the cavity-to-slab mapping is a nice simplification that makes the results portable between two experimental geometries. I agree with the reader's assessment: the central argument holds up.\n\nThe soft spots are real but proportionate. The Q-limited floor relies on the zero-temperature fluctuation-dissipation noise S_FF = mγν, set 'throughout' in Sec. II A. That is a scope condition, and the paper is explicit about it, but it means the floor is an idealized bound: any additional environmental or feedback damping would revise it. Similarly, the f = 0 Markov approximation in App. D is stated as a simplifying constraint, and the paper notes in the outlook that 3D forward scattering carries position information. These are honest limitations rather than hidden assumptions. The only mild concern is external validity: real levitated sensors at finite temperature, with 3D scattering and partial photodetection, will not exactly follow the 1D slab curves. But the paper's own loss analysis (Eqs. 42-43) already gives the qualitative behavior in the lossy regime, so this is a matter of extrapolation, not a flaw in the theory as stated.\n\nCitation practice looks fair. The frequency-dependent squeezing machinery is properly credited to Caves, Kimble et al., and LIGO work; the new contribution is the application to impulse sensing and the scaling results. Self-citation is not a problem here because the prior work supplies the baseline threshold and the optimal-filter machinery that the paper builds on.\n\nWho is this for? Anyone working on levitated optomechanical impulse detectors or quantum-limited force sensing. It gives concrete recipes for sub-SQL operation and identifies the damping-limited floor. It deserves a serious referee: the derivations need checking but they are checkable, the numerical confirmation is there, and the implications for dark matter and neutrino searches are concrete.\n\nRecommendation: send to peer review. I would engage with it, and I'd likely cite it.","headline":"A clean, self-contained derivation of e^{-r} momentum-threshold scaling for frequency-dependent squeezed readout, with a Q-limited floor; the main caveats are explicit scope conditions, not hidden flaws.","tokens_in":23758,"tokens_out":918,"would_cite":true,"duration_ms":11559,"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":"Frequency-dependent squeezed light can lower the resolvable impulse of a mechanical sensor below the standard quantum limit, with the ultimate benefit set by the oscillator's quality factor.","keywords":["squeezed light","frequency-dependent squeezing","optomechanics","impulse sensing","standard quantum limit","momentum threshold","levitated nanoparticles","quantum measurement noise"],"falsifier":"Measure the momentum threshold of a high-$Q$ suspended oscillator as a function of squeezing strength $r$ at near-unit detection efficiency and with coupling $g$ in the plateau regime $e^r g_* \\ll g \\ll \\sqrt{Q}g_*$; if the threshold does not fall as $e^{-r}$ toward a floor $\\Delta p_{\\mathrm{SQL}}/\\sqrt{Q}$, or if it falls below that floor, the central PSD or the fluctuation-dissipation assumption fails.","tokens_in":22825,"feed_emoji":"🎯","tokens_out":9568,"duration_ms":88505,"temperature":0.7,"pith_summary":"This paper asks whether squeezed light can improve the detection of weak, nearly instantaneous momentum kicks on a mechanically suspended sensor. It analyzes two equivalent geometries, a bad-cavity Fabry-Pérot resonator and a harmonically suspended dielectric slab, and argues that frequency-dependent squeezing lowers the quantum force noise at all frequencies away from resonance, whereas frequency-independent squeezing cannot. The central quantitative results are analytic scaling laws: in the strong-coupling regime the momentum threshold falls as $e^{-r}$ times the standard quantum limit, and even with perfect detection there is a lossless floor $\\Delta p_{\\mathrm{min}} \\approx \\Delta p_{\\mathrm{SQL}}/\\sqrt{Q}$ set by the mechanical quality factor. Optical losses soften the squeezing benefit to $e^{-r/2}$ at small efficiency or very strong squeezing but do not eliminate sub-SQL operation. If these scalings hold, levitated and cavity optomechanical detectors have a concrete route to quantum-enhanced impulse sensing.","feed_headline":"Frequency-dependent squeezed light beats the impulse-measurement SQL","feed_subtitle":"Minimum detectable kick falls as squeezing grows, with a lossless floor set by the oscillator's quality factor.","key_machinery":"The load-bearing object is the force power spectral density $S_{FF}(\\nu)$ of the estimator $F_E = Y^{\\mathrm{out}}/\\chi_{YF}$, whose inverse is integrated over all frequencies to form the momentum threshold. Squeezed light changes the input quadrature variances and, crucially, makes the cross-correlation $S_{XY}$ nonzero and negative over a chosen bandwidth; frequency-dependent squeezing picks the angle $\\theta_*(\\nu)$ that minimizes $S_{FF}(\\nu)$ at every frequency. The paper proves a correspondence between a bad cavity and a free-space dielectric slab, showing their output phase quadratures have the same functional form, so one calculation covers both systems. The matched-filter SNR integral is then evaluated in the large-$Q$, strong-coupling limit to yield both the exponential improvement and the quality-factor floor.","core_discovery":"The central claim is that frequency-dependent squeezed readout can push the resolvable impulse of a damped harmonic oscillator below the coherent-state SQL. Working from the input-output relation for the output phase quadrature, the paper obtains the force power spectral density minimized at each frequency by the optimal squeezing angle $\\theta_*(\\nu)$, and evaluates the momentum threshold $\\Delta p = [\\int d\\nu/(2\\pi S_{FF}(\\nu))]^{-1/2}$. In the regime $e^{r} g_*(\\omega_m) \\ll g \\ll \\sqrt{Q} g_*(\\omega_m)$, the threshold is $\\Delta p = e^{-r}[(g^2 + g_*^2(\\omega_m)e^{2r})/g^2]^{1/2} \\Delta p_{\\mathrm{SQL}} + O(\\tilde{g}^2/Q)$, so choosing $g \\gg e^r g_*$ recovers $\\Delta p \\approx e^{-r}\\Delta p_{\\mathrm{SQL}}$. The same large-$Q$ expansion breaks down when $e^{2r} \\sim Q$, and a separate large-$r$ integration gives the lossless floor $\\Delta p_{\\min} \\approx \\Delta p_{\\mathrm{SQL}}/\\sqrt{Q}$, which the authors attribute to the dissipative part $\\gamma$ of the mechanical response. With photodetection efficiency $\\eta$, the analytic scalings become $\\eta^{-1/4}e^{-r/2}\\Delta p_{\\mathrm{SQL}}$ at small $\\eta$ and $[1+(1-\\eta)e^{2r}]^{1/4}e^{-r}\\Delta p_{\\mathrm{SQL}}$ near unit efficiency.","pith_inferences":["The paper's own caveats imply that its 1D Markovian slab, which sets the forward-scattering coupling $f=0$, is optimistic for a 3D nanosphere: real detectors cannot collect all $4\\pi$ of scattered light, so an effective efficiency $\\eta<1$ will force the large-squeezing scaling toward $e^{-r/2}$ rather than $e^{-r}$.","If additional environmental or feedback damping contributes beyond the zero-temperature floor $m\\gamma\\nu$, the Q-limited floor becomes a best case; the same dissipative argument suggests the plateau height is set by the total damping rate actually present.","A sharp, testable signature of the theory is the crossover in the slope of $\\log(\\Delta p)$ versus $r$ from $-1$ to $-1/2$ as $(1-\\eta)e^{2r}$ crosses unity; measuring this crossover at fixed $\\eta$ would isolate the loss mechanism.","The broadband-integral logic should extend to other transient signals such as short force bursts or chirped waveforms, for which frequency-dependent squeezing would likewise beat frequency-independent squeezing; the paper does not evaluate those templates."],"forward_implications":["With 10 dB of frequency-dependent squeezing the plateau-regime threshold is about $e^{-1.15} \\approx 0.32$ times the SQL, so the same detector sees kicks roughly three times smaller before losses or the $Q$ floor intervene.","The lossless floor $\\Delta p_{\\min} = \\Delta p_{\\mathrm{SQL}}/\\sqrt{Q}$ makes the mechanical quality factor a direct sensitivity parameter: raising $Q$ by a factor of 4 doubles the best possible squeezing benefit.","On resonance the force noise remains pinned at the SQL, and the coupling needed to reach that point grows as $e^{2r}g_*^2$, so the power budget for squeezing is set by how far one wants to push off-resonance suppression.","If photodetection is imperfect, the benefit degrades from $e^{-r}$ to $e^{-r/2}$ at large $(1-\\eta)e^{2r}$, but sub-SQL thresholds survive, so squeezed readout remains useful even with substantial loss.","Because the bad-cavity and dielectric-slab output quadratures coincide, any experimental realization of one system inherits the optimal squeezing angle and the scaling laws of the other."],"supporting_citations":[{"why":"Establishes that squeezed input light can reduce force noise in an interferometer and introduces the quadrature-squeezing idea this paper applies to impulse sensing.","marker":"[6]"},{"why":"Supplies the general quantum-limited position-detection analysis and the frequency-integral form of the momentum threshold that the paper quotes as its SQL benchmark.","marker":"[26]"},{"why":"Gives the optimal matched filter and the SNR integral that define the momentum threshold used throughout.","marker":"[27]"},{"why":"Derives the frequency-dependent squeezing angle that minimizes the force PSD, which the paper adopts as its Eq. (23).","marker":"[29]"},{"why":"Demonstrates the frequency-dependent squeezing technology the paper assumes is experimentally available.","marker":"[8]"},{"why":"Shows that dissipative response limits the noise floor of interferometric measurements, the precedent for the Q-limited floor in Eq. (32).","marker":"[44]"},{"why":"States the fundamental Heisenberg bound on the force PSD that the paper compares against its own floor.","marker":"[45]"},{"why":"Gives the cavity-damping-limited noise floor for free masses with frequency-dependent squeezing, which the paper extends to a damped harmonic oscillator.","marker":"[46]"},{"why":"Provides the levitated-nanosphere parameters used in the simulations, including mass, mechanical frequency, damping, and temperature.","marker":"[15]"}],"fun_headline_variants":["Squeezed light beats impulse-measurement SQL","Frequency-dependent squeezing outperforms in impulse sensing","Impulse sensitivity boosted by squeezed light, capped by Q","Squeezed readout reduces kick detection limit below SQL","Optimal squeezing yields impulse gains, with a fundamental floor"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The results assume the only mechanical noise is the zero-temperature fluctuation-dissipation floor $S_{FF}=m\\gamma\\nu$ and that, in the slab model, the forward-scattered light carries no position information ($f=0$); both are idealizations, and extra damping or 3D scattering would raise the floor and change the scaling.","fun_headline_variants_meta":{"raw":{"variants":["Squeezed light beats impulse-measurement SQL","Frequency-dependent squeezing outperforms in impulse sensing","Impulse sensitivity boosted by squeezed light, capped by Q","Squeezed readout reduces kick detection limit below SQL","Optimal squeezing yields impulse gains, with a fundamental floor"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000612,"raw_usage":{"total_tokens":2850,"prompt_tokens":952,"completion_tokens":1898,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":568,"completion_tokens_details":{"reasoning_tokens":1821}},"tokens_in":568,"tokens_out":1898,"duration_ms":14253,"temperature":1.0,"reasoning_tokens":1821,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-08T20:01:31.646379+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the momentum threshold of a high-$Q$ suspended oscillator as a function of squeezing strength $r$ at near-unit detection efficiency and with coupling $g$ in the plateau regime $e^r g_* \\ll g \\ll \\sqrt{Q}g_*$; if the threshold does not fall as $e^{-r}$ toward a floor $\\Delta p_{\\mathrm{SQL}}/\\sqrt{Q}$, or if it falls below that floor, the central PSD or the fluctuation-dissipation assumption fails.","supporting_citations":[{"cited_title":"Quantum-mechanical noise in an interferometer,","cited_arxiv_id":null,"evidence_quote":"Establishes that squeezed input light can reduce force noise in an interferometer and introduces the quadrature-squeezing idea this paper applies to impulse sensing."},{"cited_title":"Quantum-limited position detection and amplification: A linear response perspective,","cited_arxiv_id":null,"evidence_quote":"Supplies the general quantum-limited position-detection analysis and the frequency-integral form of the momentum threshold that the paper quotes as its SQL benchmark."},{"cited_title":"Backaction-evading impulse measurement with mechanical quantum sensors,","cited_arxiv_id":null,"evidence_quote":"Gives the optimal matched filter and the SNR integral that define the momentum threshold used throughout."},{"cited_title":"Conversion of conventional gravitational-wave interferometers into quantum nondemolition interferometers by modifying their input and/or output optics,","cited_arxiv_id":null,"evidence_quote":"Derives the frequency-dependent squeezing angle that minimizes the force PSD, which the paper adopts as its Eq. (23)."},{"cited_title":"Frequency-dependent squeezing for advanced ligo,","cited_arxiv_id":null,"evidence_quote":"Demonstrates the frequency-dependent squeezing technology the paper assumes is experimentally available."},{"cited_title":"Quantum limits in interferometric measurements,","cited_arxiv_id":null,"evidence_quote":"Shows that dissipative response limits the noise floor of interferometric measurements, the precedent for the Q-limited floor in Eq. (32)."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"States the fundamental Heisenberg bound on the force PSD that the paper compares against its own floor."},{"cited_title":"Quantum measurement theory in gravitational-wave detectors,","cited_arxiv_id":null,"evidence_quote":"Gives the cavity-damping-limited noise floor for free masses with frequency-dependent squeezing, which the paper extends to a damped harmonic oscillator."},{"cited_title":"Real-time optimal quantum control of mechanical motion at room temperature,","cited_arxiv_id":null,"evidence_quote":"Provides the levitated-nanosphere parameters used in the simulations, including mass, mechanical frequency, damping, and temperature."}],"review_version":1}