{"id":"d55574b7-9096-43a2-885b-f21f40c53db8","arxiv_id":"2411.16334","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"For coherent states in a Mach-Zehnder interferometer, postselected dark-port detection yields an amplified phase shift that matches weak-value amplification in the AAV limit, but with shot-noise (1/√N) scaling rather than Heisenberg scaling.","lead":"It shows that a postselected, dark-port measurement in a Mach-Zehnder interferometer can amplify a small optical phase shift, recovering the weak-value amplification result in the Aharonov-Albert-Vaidman limit. It clarifies when quantum postselection actually helps phase measurement and quantifies technical advantages such as tolerance to detector saturation.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Quadrature and sensitivity formulas (Eqs. 11, 12, 15) contain algebraic errors in prefactors and a missing square root; the central amplification claim survives, but the quantitative performance evaluation is unreliable.","rationale":"The paper's central construction is mathematically sound: a coherent-state input through a balanced beam splitter yields a product state, and the dark-port output is indeed a single coherent state whose phase is amplified near the dark fringe. The reader's stated weakest assumption (the product-state property) is not a vulnerability for the stated setup—it follows directly from linear optics on coherent states. The real load-bearing defect lies in the quantitative analysis: Eqs. (11) and (12) carry a 1/√2 prefactor error, and Eq. (15) is missing a square root, making the reported signal-to-noise and sensitivity values quantitatively wrong. Because the paper's technical-advantage claims (LO phase tolerance, detector-saturation avoidance) are qualitative and would survive correction, the verdict should remain CONDITIONAL rather than REJECT or ACCEPT. A direct recomputation from Eq. (5) would settle whether the reported figures in Figs. 3 and 4 are accurate.","tokens_in":10170,"tokens_out":22866,"duration_ms":190371,"concrete_test":"Recompute the dark-port quadrature directly from α_f in Eq. (5) using X_ξ = ⟨α_f|(a†e^{iξ}+ae^{-iξ})/2|α_f⟩ with ξ=π/2+λ, and propagate Gaussian quadrature noise to ~χ. Compare the resulting prefactors with Eqs. (11), (12), and (15); if the signal prefactor is √(N/2) rather than √N/2 and Eq. (15) contains √(1−sin(2θ2)cosχ), the reported SNR and sensitivity curves in Figs. 3 and 4 must be regenerated.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—that the MZI dark-port output is a single coherent state with amplified phase and shot-noise-limited precision—rests on the correct product-state transformation of a coherent state through the beam splitter. That part is sound. However, the quantitative extraction formulas are internally inconsistent. From Eq. (9) with γ=0, |α_f| = sqrt(N/2) sqrt(1 − sin(2θ2) cosχ), so the homodyne quadrature at ξ=π/2+λ is X_ξ = |α_f| sin~χ_B. Equations (11) and (12) instead put sqrt(N)/2, a factor-of-1/√2 error in the signal amplitude. More seriously, Eq. (15) gives ~R_S/N = √(2N)[1 − sin(2θ2) cosχ]|cos~χ_B|~χ_B, but error propagation from X_ξ = |α_f| sin~χ_B yields δ~χ_B = 1/(2|α_f||cos~χ_B|), hence ~R_S/N = √(2N) sqrt(1 − sin(2θ2) cosχ)|cos~χ_B|~χ_B. The missing square root changes the quantitative sensitivity and the comparison with conventional homodyne detection. These errors do not invalidate Eqs. (8) and (10), but they make the paper's quantitative conclusions unreliable until corrected.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript considers a Mach-Zehnder interferometer with an input coherent state and proposes a postselected phase-shift measurement at the dark port, where the output field remains a single coherent state whose phase is amplified. In the small-phase (AAV) limit the amplification coefficient is identified with the Aharonov-Albert-Vaidman weak value, and an exact expression for the amplified phase is given. The paper then analyzes homodyne quadrature detection of the dark-port field, including signal-to-noise ratios, averaging over repeated measurements, and technical advantages with respect to local-oscillator phase errors and photodetector saturation. The central conclusion is that the amplified phase can be extracted from a quadrature measurement with shot-noise-limited precision 1/√N, and that postselection offers technical, not fundamental, advantages in this coherent-state setting.","tokens_in":10451,"tokens_out":12550,"duration_ms":110952,"significance":"The core derivation (Eqs. (3)-(10)) is self-contained, and the observation that the dark-port output is a single coherent state rather than a cat-like superposition is a useful conceptual clarification of the difference between interferometric postselection and qubit-coupled weak-value amplification. The paper correctly avoids claiming any quantum-enhanced precision scaling and explicitly notes the distinction from entangled qubit-oscillator models. Its potential value lies in the technical-advantage analysis (tolerance to LO phase errors and detector saturation). However, the quantitative formulas for the quadrature signal and sensitivity contain algebraic errors that must be corrected before the performance claims can be accepted.","major_comments":[{"comment":"The quadrature amplitude in Eq. (11) is too small by a factor of 1/√2. From Eq. (5) with γ=0, |α_f| = sqrt(N/2)(cosθ2 − sinθ2) in the AAV-limit amplitude, so with ξ=π/2+λ the homodyne signal is X_ξ = sqrt(N/2)(cosθ2 − sinθ2) sin~χ_A. Equation (11) instead has √N/2, which is smaller by √2. This is not a convention issue: it is inconsistent with the authors' own Eq. (5) and with Eq. (14), which appears to have been derived from the correct amplitude.","section":"Eq. (11)"},{"comment":"The same factor-of-1/√2 error appears in the beyond-AAV signal. From Eq. (5) with γ=0, |α_f| = sqrt(N/2) sqrt(1 − sin(2θ2) cosχ), so Eq. (12) should read X_ξ = sqrt(N/2) sqrt(1 − sin(2θ2) cosχ) sin~χ_B, not (√N/2) times the bracket. The printed form underestimates the signal amplitude by √2.","section":"Eq. (12)"},{"comment":"Equation (15) is missing the square root on the interference factor. Error propagation from X = |α_f| sin~χ with δX = 1/2 gives δ~χ = 1/(2|α_f||cos~χ|), hence ~R_S/N = √(2N) sqrt(1 − sin(2θ2) cosχ) |cos~χ_B| ~χ_B. The printed [1 − sin(2θ2) cosχ] without the square root overestimates the sensitivity when the bracket is small; it is also inconsistent with Eq. (14), since in the AAV limit 1 − sin(2θ2) cosχ reduces to (cosθ2 − sinθ2)^2 and the square root is needed to recover Eq. (14).","section":"Eq. (15)"},{"comment":"The sensitivity is defined as ~χ/δ~χ, i.e., for the amplified phase. For estimating the physical phase χ, the relevant quantity is χ/δχ with δχ = δ~χ/|d~χ_B/dχ|. In the AAV limit these coincide, but for the beyond-AAV examples (e.g., Fig. 2 with χ=10^{-2}) the nonlinearity of ~χ_B(χ) means the reported ~R_S/N is not the estimation sensitivity for χ. The authors should either compute the derivative with respect to χ or explicitly state that they report the SNR of the amplified phase rather than of the true phase.","section":"Eqs. (14)-(15)"}],"minor_comments":[{"comment":"The phrase 'optical coherent states (|1⟩ and |2⟩)' should read '(|α1⟩ and |α2⟩)'; the current notation is confusing and appears to be a typo.","section":"Abstract and Introduction"},{"comment":"The expression for |f⟩ has an unbalanced parenthesis: '|f ⟩ = cosθ2|1⟩ +i sinθ2e−iγ |2⟩)/√2' should omit the extra closing parenthesis before the division.","section":"Eq. (7)"},{"comment":"The statement that 'the phase shift χ is amplified as ~χ_A = A_w χ' holds only when c1+c2 is real (e.g., γ=0); for complex c1+c2 the phase of the output is not simply A_w χ. This condition should be stated explicitly.","section":"Below Eq. (7)"},{"comment":"The word 'squeezing uncertainty' is misleading: a coherent state has no squeezing, and the reduction described is simply the usual averaging over repeated measurements.","section":"Fig. 3 caption"},{"comment":"The subscript 'Byd' appears to be a typo for 'Beyond' or similar and should be standardized.","section":"Eq. (12)"},{"comment":"The notation for the inverse sine is awkward; the argument should be written as X_ξ / [|α_f| (kmax/Nsat)] with clearer bracketing.","section":"Eq. (22)"}],"recommendation":"major_revision","confidential_remarks":"The core derivation is sound and the paper is within scope for a quantum-optics or quantum-metrology journal. The algebraic errors in Eqs. (11), (12), and (15) are fixable, but they affect the quantitative performance claims, so a major revision is appropriate. I would also suggest the authors carefully reconsider the sensitivity definition for the true phase χ in the nonlinear regime, as this is currently a source of potential misinterpretation. The paper does not claim a quantum advantage in precision scaling, and its contribution is moderate but useful; after correction it should be acceptable."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe key claim of this paper is that postselection in an MZI with a coherent input leaves the dark-port output as a single coherent state—not a cat state—and that the AAV weak value emerges in the linearized limit. That claim survives reading. The exact beyond-AAV expression for the amplified phase (their Eq. 10) and the explicit demonstration that the photon-number scaling stays 1/√N are genuinely useful. This is a modest but legitimate contribution, not a breakthrough.\n\nThe derivation of α_f from the beam-splitter unitaries is correct. The identification of A_w = c1/(c1+c2) as the AAV weak value checks out. The numerical examples in Fig. 2 show the expected amplification. The technical-advantage sections on LO phase error and detector saturation are sensible; the saturation analysis with the exponential model is a nice addition, though the qualitative conclusion that postselection helps when detectors saturate is not new (see Refs. [21, 22, 26, 27]).\n\nThe soft spots are real but concentrated. The quadrature formulas (11) and (12) are missing a factor of √2: from Eq. (5), |α_f| = √(N/2)√(1−sin(2θ2)cosχ), not (√N/2) times that. And in Eq. (15), the factor [1−sin(2θ2)cosχ] should be its square root. These are not merely prefactor typos; they change the claimed sensitivity and the comparison with conventional homodyne detection. The phase-extraction ambiguity for large amplification (branching of arctan/arcsin) is not discussed, which could matter in practice. The abstract also has a confusing notation where path states and coherent states are both called |1⟩, |2⟩.\n\nNone of this undermines the central claim. The paper's quantitative conclusions cannot be trusted until the algebra is fixed, but the qualitative message—that postselection amplifies the phase while preserving coherent-state structure and shot-noise scaling—is sound.\n\nI'd send it to a standard quantum-metrology referee. It deserves a careful review, and the authors should be asked to correct Eqs. (11), (12), and (15) and to discuss the phase-unwrapping issue. For my own work, I probably wouldn't cite it in the next year, but I would bring it to a reading group to illustrate the MZI versus WVA distinction.","headline":"The central claim survives—postselected phase amplification in a coherent-state MZI yields a single coherent state with AAV weak value—but the quantitative quadrature and sensitivity formulas have factor errors that must be fixed before the paper's performance conclusions can be trusted.","tokens_in":11033,"tokens_out":5082,"would_cite":false,"duration_ms":43226,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["03.65.Ta","42.50.-p"],"model":"deepseek-v4-flash","headline":"Postselected amplification of a small phase shift in a Mach-Zehnder interferometer is carried by a weak value while the dark-port output remains a single coherent state, buying technical robustness rather than a precision-scaling advantage.","keywords":["postselected amplification","weak-value amplification","Mach-Zehnder interferometer","coherent state","phase shift estimation","homodyne quadrature","shot-noise limit","detector saturation"],"falsifier":"Send a coherent state through the interferometer with known $\\chi \\sim 10^{-2}$, set $\\gamma=0$, and scan $\\theta_2$ while performing homodyne tomography on port 3. If the reconstructed Wigner function shows negativity or interference fringes, or if the measured quadrature variance departs from the coherent-state value $1/4$, the product-state assumption fails. Conversely, the claim that precision stays at $1/\\sqrt{N}$ can be tested by checking whether the estimation error improves with photon number as $1/\\sqrt{N}$ rather than $1/N$.","tokens_in":9956,"feed_emoji":"⚛️","tokens_out":12489,"duration_ms":107997,"temperature":0.7,"pith_summary":"The paper proposes a postselected amplification scheme for measuring a small phase shift imprinted on one arm of a Mach-Zehnder interferometer through which a coherent-state laser passes. It argues that, although the two arms of the interferometer do not form a quantum superposition entangled with the optical field, the normally dark output port still carries an amplified phase shift $\\tilde{\\chi}_A = A_w \\chi$ in the Aharonov-Albert-Vaidman limit, with $A_w = c_1/(c_1+c_2)$ playing the role of a weak value. Beyond that limit the paper supplies an exact expression, Eq. (10), for the amplified phase. The consequence that matters is that the postselected output remains a single coherent state, so the photon-number scaling of the precision stays at the shot-noise limit $1/\\sqrt{N}$ rather than improving to $1/N$. A reader should care because the scheme cleanly separates postselection-based amplification from entangled-meter metrology and identifies a practical regime in which postselection improves robustness to detector saturation and local-oscillator phase error without sacrificing shot-noise-limited sensitivity.","feed_headline":"Postselection amplifies tiny phase shifts in a laser interferometer","feed_subtitle":"Dark-port output stays a single coherent state, preserving shot-noise precision while dodging detector saturation.","key_machinery":"The load-bearing object is the product-state transformation of the coherent field through the interferometer, Eq. (3), together with the recombination formula for the dark-port amplitude $\\alpha_f$. Defining $c_1 = \\cos\\theta_2/\\sqrt{2}$ and $c_2 = -e^{i\\gamma}\\sin\\theta_2/\\sqrt{2}$ lets the paper rewrite $\\alpha_f = (c_1 e^{i\\chi}+c_2)\\alpha$ and identify $A_w = c_1/(c_1+c_2) = \\langle f|\\hat{A}|i\\rangle/\\langle f|i\\rangle$ with $\\hat{A} = |1\\rangle\\langle 1|$; this is the Aharonov-Albert-Vaidman weak value, a named quantity that can be large when $c_1+c_2$ is small. The phase is read out by homodyne field-quadrature measurement, $X_\\xi = \\langle \\alpha_f | \\hat{X}_\\xi | \\alpha_f\\rangle = \\sqrt{I_f}\\sin\\tilde{\\chi}$, with the local-oscillator phase chosen as $\\xi = \\pi/2 + \\lambda$ for maximal sensitivity. The single-coherent-state form of $|\\alpha_f\\rangle$ is what lets the amplified phase appear as a rotation of the field amplitude, rather than as an interference effect between two coherent-state components.","core_discovery":"The central discovery is that postselected amplification of a phase shift in a Mach-Zehnder interferometer works for optical coherent states even when the which-path degrees of freedom are not entangled with the optical field. The beam-splitter action is taken as $U_{BS_1}|\\alpha;0\\rangle = |\\alpha\\cos\\theta_1; -i\\alpha\\sin\\theta_1\\rangle$, a product of two coherent states, and the dark-port output is a single coherent state $|\\alpha_f\\rangle$ with $\\alpha_f = (c_1 e^{i\\chi} + c_2)\\alpha$. For small $\\chi$, this gives the amplified phase $\\tilde{\\chi}_A = A_w \\chi$ with $A_w = c_1/(c_1+c_2)$, the standard weak-value result, and beyond the small-$\\chi$ limit the exact amplified phase is $\\tilde{\\chi}_B = \\arctan\\big((\\sin\\chi\\cos\\theta_2 - \\sin\\gamma\\sin\\theta_2)/(\\cos\\chi\\cos\\theta_2 - \\cos\\gamma\\sin\\theta_2)\\big)$. Because the output is a coherent state rather than a superposition of two coherent states, the precision follows the $1/\\sqrt{N}$ shot-noise scaling, and the practical benefit of postselection is technical: the amplified phase is easier to read out in the presence of local-oscillator phase error and photodetector saturation.","pith_inferences":["This suggests a general rule: in any linear-optics interferometer whose beamsplitter leaves a product of coherent states, postselection cannot change the photon-number scaling; the entire benefit is improved tolerance to technical imperfections.","The paper does not compute Fisher information, but the single-coherent-state output implies the available information about $\\chi$ at the optimal quadrature is proportional to the photon number $N$, so postselection appears to redistribute measurement resources over repeated runs rather than create new information.","A direct experiment could verify the saturation claim: with a bright input ($N$ well above $N_{\\rm sat}$) and strong postselection, the estimation error should remain small even where the un-postselected homodyne currents would be saturated.","If the forthcoming nonlinear-coupling extension is made, the product-state output would still be a single coherent state, so any precision enhancement would have to originate in the nonlinear interaction itself rather than in postselection."],"forward_implications":["A small phase shift can be extracted from the dark-port homodyne signal as $\\tilde{\\chi}_A = A_w \\chi$, making the readout less sensitive to local-oscillator phase error than a direct phase-shift measurement.","Because the output is a single coherent state, postselection cannot push the precision beyond the shot-noise limit; the scheme is a technical-robustness tool, not a way to beat standard quantum scaling.","The exact beyond-AAV formula Eq. (10) lets an experimenter calibrate the true phase $\\chi$ from the measured amplified phase, extending the scheme to larger phase shifts.","Averaging over $M$ repeated pulses reduces the estimation uncertainty by $1/\\sqrt{M}$, compensating for the photon loss caused by postselection and making the scheme practical for static parameter estimation.","The same interferometric setup can be used for measuring relativistic-gravity-induced phase shifts with the added benefit of postselection-based robustness."],"supporting_citations":[{"why":"Defines the weak value and the Aharonov-Albert-Vaidman limit that the paper reproduces as its amplified phase $\\tilde{\\chi}_A = A_w\\chi$.","marker":"[9,10]"},{"why":"Provides the entangled qubit-oscillator weak-value model with coherent states, the comparison point for the paper's product-state MZI output.","marker":"[16]"},{"why":"Shows the photon-number scaling enhancement toward $1/N$ available with nonlinear coupling and postselection in the entangled model, which the paper argues does not occur here.","marker":"[17]"},{"why":"Introduces the Mach-Zehnder phase-shift measurement for relativistic gravity parameters that this work extends with postselection.","marker":"[6–8]"},{"why":"Supplies the homodyne field-quadrature measurement method used to extract the amplified phase from the dark port.","marker":"[18]"},{"why":"Motivates the technical-advantage view that weak-value and postselected schemes can be robust to technical noise, which the paper applies to phase error and detector saturation.","marker":"[21,22]"},{"why":"Demonstrates weak-value amplification advantages under detector saturation, the effect the paper analyzes for quadrature measurement.","marker":"[26,27]"}],"fun_headline_variants":["Postselected phase amp works without which-path entanglement","Dark-port coherent state amplifies phase, dodges detector saturation","Interferometer postselection boosts phase readout, preserves precision","Phase shift amplified by postselection, output stays coherent","Postselected Mach-Zehnder: amplified phase, shot-noise retained"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The derivation assumes that after the first beam splitter the two interferometer arms are in a product state of two coherent states, so the dark-port output is one coherent state; if the input field were non-classical or the beam splitter entangled the path modes, the output would be a superposition and the phase-extraction and scaling analysis would break down.","fun_headline_variants_meta":{"raw":{"variants":["Postselected phase amp works without which-path entanglement","Dark-port coherent state amplifies phase, dodges detector saturation","Interferometer postselection boosts phase readout, preserves precision","Phase shift amplified by postselection, output stays coherent","Postselected Mach-Zehnder: amplified phase, shot-noise retained"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000211,"raw_usage":{"total_tokens":1459,"prompt_tokens":1034,"completion_tokens":425,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":650,"completion_tokens_details":{"reasoning_tokens":343}},"tokens_in":650,"tokens_out":425,"duration_ms":143884,"temperature":1.0,"reasoning_tokens":343,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T13:14:46.169880+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Send a coherent state through the interferometer with known $\\chi \\sim 10^{-2}$, set $\\gamma=0$, and scan $\\theta_2$ while performing homodyne tomography on port 3. If the reconstructed Wigner function shows negativity or interference fringes, or if the measured quadrature variance departs from the coherent-state value $1/4$, the product-state assumption fails. Conversely, the claim that precision stays at $1/\\sqrt{N}$ can be tested by checking whether the estimation error improves with photon number as $1/\\sqrt{N}$ rather than $1/N$.","supporting_citations":[{"cited_title":"Zhang, A","cited_arxiv_id":null,"evidence_quote":"Shows the photon-number scaling enhancement toward $1/N$ available with nonlinear coupling and postselection in the entangled model, which the paper argues does not occur here."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the homodyne field-quadrature measurement method used to extract the amplified phase from the dark port."}],"review_version":1}