{"id":"2891e706-6d92-46cc-bddb-0e7379dd61c1","arxiv_id":"2506.21640","paper_version":1,"verdict":"UNVERDICTED","confidence":"HIGH","novelty_score":0.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A survey of CV-QKD theory, photonic integration, and machine learning advances, with no new results.","lead":"This paper is a survey of continuous-variable quantum key distribution (CV-QKD), covering squeezed light, chip integration, machine learning, and tensor networks. It offers a broad entry point to the field but contains no new experimental or theoretical results.","discovery_kind":"review","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Practical-security premise is asserted but not verified; the conclusion's MITM argument is a concrete warning sign.","rationale":"The paper is a survey; the central claim is an external summary, not a new result. For that claim to hold, practical security must hold under finite-size, composable, realistic conditions. The survey does not establish this, and its conclusion contains a concrete security error regarding man-in-the-middle attacks. Eq. (8) also appears dimensionally wrong, further reducing reliability as a reference. These are real concerns, but they do not turn a survey with no original results into an accepted or rejected research paper; they reinforce the reader's UNVERDICTED assessment. The proposed numerical check would settle whether the practical-security premise actually holds. Credit is due for the survey's substantial citation base and for acknowledging hardware challenges such as excess noise and side channels; the broad telecom-compatibility claim is not baseless, but the security premise is insufficiently analyzed.","tokens_in":18790,"tokens_out":10231,"duration_ms":113353,"concrete_test":"Take the channel parameters of Figure 6 (50 km single-mode fiber, 500 kHz repetition, realistic excess noise) and compute the composable finite-size secret key rate using the security framework of Ref. [23] (Pirandola, PRR 2021), together with the corrected mutual information I(A:B) = (1/2) log2(1 + Vs/VN) in place of Eq. (8). If the resulting key rate is not positive for the cited protocols, the survey's unconditional 'robust security proofs' statement and its practicality claim are overstated.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central practicality claim requires that CV-QKD security proofs remain valid under realistic finite-size, composable conditions with practical hardware noise. The survey asserts this in Section 4.1 ('providing robust security proofs [28–31]') and again in Section 5.4, but it never verifies that the cited proofs cover the regimes of its own experimental examples, such as the 50-km, 500-kHz fiber setup in Figure 6. This is the load-bearing step: if refs [28,30] are asymptotic, assume collective attacks, or require ideal reconciliation, then the 'practical alternative' claim is unsupported. A concrete warning sign is the Conclusion's claim that sending homodyne measurement results over a classical channel 'partially reduce[s]' man-in-the-middle risk because the key cannot be reconstructed from them. That is wrong: an active MITM can replace or suppress those results, and QKD requires an authenticated classical channel. Thus the survey contains a security misconception at the point where it argues for practical deployment.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript is a survey of continuous-variable quantum key distribution (CV-QKD). It covers squeezed states of light, continuous-variable quantum teleportation, the basic principles and experimental implementations of CV-QKD, chip-scale integration, and feasibility challenges such as squeezed-light generation, photon loss, and security. The central claim, stated in the abstract and conclusion, is that CV-QKD is a more practical alternative to discrete-variable QKD because of its compatibility with existing telecom infrastructure, and that progress in photonic integrated circuits, machine learning, and tensor networks is enabling large-scale quantum-secure networks. The paper does not present original derivations; it reviews the literature and includes proposals for an integrated teleportation circuit and a 50-km CV-QKD experimental setup.","tokens_in":18875,"tokens_out":6068,"duration_ms":67190,"significance":"If accurate, the survey offers a useful, broad entry point to the CV-QKD literature, especially for readers interested in photonic integration and the role of machine learning. It compiles a large bibliography and highlights concrete experimental milestones. The main value is pedagogical and bibliographic rather than advancing new technical results. However, several technical and security-related inaccuracies reduce its reliability as a reference: a missing logarithm in a central formula, a mistaken claim about man-in-the-middle protection, and an unqualified assertion about the security proofs underlying the practicality claim. These need correction before the survey can serve as a dependable guide.","major_comments":[{"comment":"The claim that sending homodyne measurement results over the classical channel 'partially reduce[s]' man-in-the-middle risk because 'none of the keys could be directly reconstructed using the measurement results' is incorrect. An active man-in-the-middle can replace or suppress the classical messages; QKD requires an authenticated classical channel for tamper evidence. This is a security misconception in the section that argues for practical deployment, and it should be revised to state that the classical channel must be authenticated and that measurement outcomes provide no confidentiality against active attacks.","section":"Section 6 (Conclusion)"},{"comment":"The mutual information for Gaussian-modulated coherent states is given as I(A:B) = 1/2(1 + Vs/VN). The correct expression is (1/2) log2(1 + Vs/VN). As written, the formula is dimensionally inconsistent and would give incorrect key-rate values; it also conflicts with the entropy-based definition in Eq. (7). Please correct the formula and specify the base of the logarithm.","section":"Section 4.1, Eq. (8)"},{"comment":"The survey asserts 'robust security proofs [28–31]' and later offers qualitative mitigation strategies for side-channel attacks, but it does not state whether the cited proofs are finite-size and composable, nor whether they cover the specific protocols and experimental regimes (e.g., the 50-km, 500-kHz example in Fig. 6) used to support the practicality claim. Because the abstract's 'practical alternative' claim depends on this premise, the paper should either explicitly cite composable finite-size security results (e.g., Refs. [23,32]) and state the attack model, or qualify the claim to avoid overstating the current evidence.","section":"Sections 4.1 and 5.4"}],"minor_comments":[{"comment":"Both equations contain a duplicated exponential ('exp exp'). Please remove the redundant 'exp'.","section":"Eqs. (1) and (2)"},{"comment":"The notation ⟨ξ⟩ = ψ_r(x) is incorrect; the wavefunction should be denoted ψ_r(x) or similar. Also, λ is called 'the squeezing parameter,' but the squeezing parameter is r (with λ = e^{2r}). In the momentum-space wavefunction, the variable x appears instead of p. These points should be corrected.","section":"Section 3.1, Eq. (4) and surrounding text"},{"comment":"The phrase 'shore algorithms' should read 'Shor's algorithms.'","section":"Section 6 (Conclusion)"},{"comment":"Section 4.3 states that fiber-based DV-QKD achieves 'kbps level' rates for distances up to 100 km, while the Introduction cites a 400-km DV-QKD demonstration. These statements should be reconciled by specifying the distance/rate trade-off or the type of fiber used.","section":"Section 4.3 vs. Introduction"},{"comment":"The sentence 'However, this protocol demonstrates robustness [107]...' does not identify which protocol is meant. Please clarify the referent, as the preceding text discusses photon loss generally and also mentions scalability.","section":"Section 5.3"},{"comment":"The phrase 'Quantum attacks, including collective Gaussian attacks' conflates attack classes. Collective attacks are not the most general; coherent attacks are. Please distinguish these classes, especially when discussing the scope of the security proofs cited in Section 4.1.","section":"Section 5.4"},{"comment":"For continuous-variable systems, the entropies in Eq. (7) are differential entropies, not discrete Shannon entropies. The text should say so, and the logarithm base should be specified consistently with Eq. (8).","section":"Section 4.1, Eq. (7)"}],"recommendation":"major_revision","confidential_remarks":"The survey has a useful scope and a rich bibliography, but the security discussion needs substantive revision. The man-in-the-middle misconception and the uncritical assertion of 'robust security proofs' are the most serious issues because they directly support the paper's central practicality claim. The authors would benefit from consulting with a QKD security expert during revision. The paper also has several smaller technical errors (e.g., Eq. (8), Eq. (4)) that should be corrected before publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a review, not a research paper. It reports no new protocol, measurement, or proof, and the authors don't claim otherwise. Its value is organizational: it pulls together CV-QKD basics, PIC integration, ML-assisted operation, and tensor-network ideas, with a mostly sensible reference list. The high-level claim that CV-QKD is more compatible with existing telecom infrastructure than DV-QKD is defensible and in line with the literature.\n\nWhere it gets soft: the technical accuracy is not yet referee-clean. Eq. (8) gives mutual information without the logarithm; as written it's not information-theoretically meaningful. Eq. (4) garbles the squeezed-state wavefunction (and uses notation like ⟨ξ⟩ for a function). There are repeated typos in the operator definitions (\"exp exp\") and \"shore algorithms\" in the conclusion. None of these sink the survey's purpose, but each is exactly the kind of thing a newcomer will copy.\n\nThe more substantive problem is the security discussion. Section 4.1 asserts \"robust security proofs [28–31]\" without saying which attack models, finite-size effects, or composability definitions those proofs cover. That matters because the paper's central practicality claim depends on those proofs holding in realistic conditions; the survey never checks. And the conclusion makes a concrete error: it says sending homodyne measurement results over a classical channel \"partially reduce[s]\" man-in-the-middle risk because the key can't be reconstructed from them. That's wrong. An active MITM can replace or suppress those classical results, and QKD requires an authenticated classical channel. This is not a minor slip; it's the one place the paper argues for practical deployment against the most obvious attack.\n\nThe reader's take says the descriptive statements broadly match the literature, and I agree. The reference list is mostly appropriate, and the self-citations appear as background rather than as load-bearing support. The machine-learning and PIC sections are a fair reflection of current work.\n\nBottom line: this survey can orient a newcomer if read alongside a reliable textbook or review. It is not a dependable standalone reference in its current form. If it came to a journal freshly, I'd send it to review and ask for major revision—fix the equations, correct the MITM claim, and add a sentence or two distinguishing asymptotic/composable security results. The core organizational value is real, but the errors need fixing before it should be recommended to students.","headline":"A useful newcomer-oriented CV-QKD survey whose organizational value is undercut by several real technical errors, including a mistaken claim about classical-channel homodyne results and man-in-the-middle attacks.","tokens_in":19456,"tokens_out":1830,"would_cite":false,"duration_ms":21831,"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":"This survey argues that continuous-variable QKD, which encodes keys in the quadratures of coherent or squeezed light, is the near-term practical route to quantum-secure networks because it reuses telecom components and supports chip-scale…","keywords":["continuous-variable quantum key distribution","squeezed light","photonic integrated circuits","machine learning","tensor networks","measurement-device-independent QKD","homodyne detection","quantum networks"],"falsifier":"A field trial over a metropolitan fiber link in which, under fully untrusted-device assumptions, the measured excess noise drives the finite-size composable secret key rate to zero at the protocol's nominal distance and modulation would falsify the survey's central practicality claim.","tokens_in":18520,"feed_emoji":"🔐","tokens_out":4803,"duration_ms":55052,"temperature":0.7,"pith_summary":"This survey argues that continuous-variable quantum key distribution (CV-QKD), which encodes key bits in the amplitude and phase quadratures of coherent or squeezed light, is the near-term practical route to quantum-secure networks. Unlike discrete-variable QKD, which needs single-photon sources and detectors, CV-QKD works with standard telecom lasers, modulators, and homodyne receivers, and it has been demonstrated over 50 to 100 km of fiber. The paper gathers evidence that photonic integrated circuits can put these systems on a chip, that machine learning can suppress excess noise and detect eavesdropping, and that measurement-device-independent protocols remove the main detector side channels. If the survey's picture is right, large metropolitan quantum networks can be built by upgrading existing optical infrastructure rather than replacing it.","feed_headline":"CV-QKD is the practical route to quantum-secure fiber networks","feed_subtitle":"Telecom-compatible lasers, chip-scale integration, and machine learning put metropolitan quantum networks within reach.","key_machinery":"The load-bearing mechanism is the continuous-variable encoding itself: information is carried in the quadratures of coherent or squeezed states and read out by homodyne or heterodyne detection, which is exactly the hardware of coherent optical telecommunications. Security and rate are governed by the formula $K = \\beta I(A:B) - \\chi(E)$, where $\\beta$ is reconciliation efficiency, $I(A:B)$ the Alice-Bob mutual information, and $\\chi(E)$ Eve's Holevo bound computed from symplectic eigenvalues of the covariance matrix. The paper treats the squeezed state, generated by parametric down-conversion, four-wave mixing, or micro-ring resonators, as the enabling resource, and chip-scale integration of sources, modulators, and detectors as the scaling path.","core_discovery":"The paper's central claim is that CV-QKD has matured from laboratory protocol to deployable technology: it reports that fiber-based CV-QKD reached roughly 1 kbps over 80 km and beyond 100 km once excess noise was managed, that chip-based homodyne detectors and quantum entropy sources exist, and that measurement-device-independent QKD variants have run over 404 km of fiber and a 19.2 km urban free-space link. The security story it tells is that Gaussian and discrete-modulated CV-QKD now carry composable security proofs, while machine-learning-assisted estimation and tensor-network processing keep the rate formula $K = \\beta I(A:B) - \\chi(E)$ positive under realistic noise. The survey's conclusion is that these pieces fit together: CV-QKD's telecom compatibility, integrated photonics, and data-driven noise control put large-scale quantum-secure networks within practical reach.","pith_inferences":["Beyond the survey, if CV-QKD's practicality claim holds, the technology could become the default first-generation QKD for optical backhaul, because it reuses coherent transceivers already installed in telecom networks.","The survey's reliance on composable security proofs suggests the decisive next tests will be end-to-end key-rate demonstrations with fully untrusted devices, rather than loss-versus-rate curves alone.","Machine-learning-based anomaly detection may push QKD security practice toward a hybrid model that combines provable bounds with continuous hardware monitoring, an implication the paper does not develop.","Squeezed-light sources in micro-ring resonators could eventually make chip-scale CV-QKD inexpensive enough for subscriber premises, provided packaging and insertion losses are solved."],"forward_implications":["CV-QKD can be deployed over existing telecom fiber using commercial lasers and homodyne receivers, without single-photon detectors, making near-term metropolitan quantum networks feasible.","Photonic integrated circuits, especially with squeezed-light sources and modular heterogeneous integration, can shrink CV-QKD transceivers to chip scale.","Machine-learning-assisted noise estimation and parameter optimization can raise secret key rates and extend transmission distance by reducing excess noise.","Measurement-device-independent QKD variants remove detector side channels and have been demonstrated over 404 km of fiber and a 19.2 km urban free-space link, supporting CV-QKD security.","Tensor networks provide efficient tools for analyzing quantum correlations in complex networks, improving the robustness and efficiency of key distribution protocols."],"supporting_citations":[{"why":"Provides the surveyed experimental CV-QKD system picture, including the 50-km fiber setup and the emphasis on laser stability and excess noise.","marker":"[22]"},{"why":"Supplies composable-security results and practical key rates that anchor the survey's claim that CV-QKD is deployable in wired networks.","marker":"[23]"},{"why":"Demonstrates long-distance CV-QKD with entangled states, supporting the compatibility-with-telecom-infrastructure argument.","marker":"[24]"},{"why":"Supplies the survey of machine-learning-assisted CV-QKD that underlies the paper's claims about data-driven noise estimation and security.","marker":"[35]"},{"why":"Introduces measurement-device-independent QKD, the principal protocol-level fix the survey cites for detector side channels.","marker":"[83]"},{"why":"Reports the experimental long-distance CV-QKD demonstration at roughly 1 kbps over 80 km, a central data point for practicality.","marker":"[90]"},{"why":"Shows that controlling excess noise extends CV-QKD beyond 100 km, the key distance claim in the survey.","marker":"[91]"},{"why":"Provides the chip-based homodyne detector result used to argue that CV-QKD components can be integrated on-chip.","marker":"[92]"},{"why":"Demonstrates a programmable nanophotonic chip generating squeezed light in multiple modes, the main evidence for chip-scale squeezing.","marker":"[106]"},{"why":"Reports measurement-device-independent QKD over 404 km of fiber, supporting the survey's long-distance security claims.","marker":"[15]"}],"fun_headline_variants":["Chip-scale CV-QKD makes quantum-secure networks practical","ML and photonics boost CV-QKD performance for real deployments","CV-QKD survey: 404 km fiber, chip detectors, and machine learning","Practical CV-QKD: from telecom to integrated photonics and ML","CV-QKD matures with chip detectors and ML-driven noise control"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The survey's practicality claim rests on the assumption that the cited security proofs for Gaussian and discrete-modulated CV-QKD remain valid under realistic finite-size, composable conditions and with practical hardware noise.","fun_headline_variants_meta":{"raw":{"variants":["Chip-scale CV-QKD makes quantum-secure networks practical","ML and photonics boost CV-QKD performance for real deployments","CV-QKD survey: 404 km fiber, chip detectors, and machine learning","Practical CV-QKD: from telecom to integrated photonics and ML","CV-QKD matures with chip detectors and ML-driven noise control"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000311,"raw_usage":{"total_tokens":1764,"prompt_tokens":931,"completion_tokens":833,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":547,"completion_tokens_details":{"reasoning_tokens":740}},"tokens_in":547,"tokens_out":833,"duration_ms":9120,"temperature":1.0,"reasoning_tokens":740,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T22:40:57.360214+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A field trial over a metropolitan fiber link in which, under fully untrusted-device assumptions, the measured excess noise drives the finite-size composable secret key rate to zero at the protocol's nominal distance and modulation would falsify the survey's central practicality claim.","supporting_citations":[{"cited_title":"Experimental demonstration of long-distance continuous-variable quantum key distribution","cited_arxiv_id":null,"evidence_quote":"Reports the experimental long-distance CV-QKD demonstration at roughly 1 kbps over 80 km, a central data point for practicality."},{"cited_title":"Long-distance continuous-variable quantum key distribution by controlling excess noise","cited_arxiv_id":null,"evidence_quote":"Shows that controlling excess noise extends CV-QKD beyond 100 km, the key distance claim in the survey."},{"cited_title":"A homodyne detector integrated onto a photonic chip for measuring quantum states and generating random numbers","cited_arxiv_id":null,"evidence_quote":"Provides the chip-based homodyne detector result used to argue that CV-QKD components can be integrated on-chip."},{"cited_title":"Quantum circuits with many photons on a programmable nanophotonic chip","cited_arxiv_id":null,"evidence_quote":"Demonstrates a programmable nanophotonic chip generating squeezed light in multiple modes, the main evidence for chip-scale squeezing."}],"review_version":1}