{"id":"771e1c02-07ca-4976-87a7-8069a338e9a7","arxiv_id":"1908.04539","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"The AMDI-QKD protocol cannot surpass the repeaterless key rate bound when the entanglement sources are standard parametric down-conversion sources.","lead":"This paper tests whether a proposed quantum key distribution protocol, AMDI-QKD, can beat the fundamental repeaterless bound when entanglement sources are imperfect. It finds that standard parametric down-conversion sources cannot meet the required source quality, so the protocol needs much better single-photon sources to be useful.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The PDC-impossibility proof depends on Eq. (1) being an exact or upper-bound rate for realistic multi-photon sources; this status is not established, so Eq. (9) may not be a necessary condition.","rationale":"The reader's weakest assumption is exactly the transfer of the Ref. [34] security proof to the realistic device set. I agree with that identification and sharpen it: the direction of the inequality matters. The necessary-condition argument compares Eq. (8) to the repeaterless bound; for that to be a valid necessary condition, Eq. (8) must be an upper bound on the actual secret key rate. The usual role of Eq. (1) in asymptotic analyses is as an achievable-rate lower bound, and the paper never proves that it is also an upper bound (or exact) for sources of Eq. (2) with multi-photon components. Thus the central corollary is not yet supported as a no-go statement. The admitted lack of an analytical proof of pX_c=pZ_c is a secondary but related gap: it affects the numerical tighter condition and, at unit efficiency, the derivation of Eq. (8) relies on the corresponding equalities. Because this missing step underpins the main impossibility claim rather than just the quantitative optimization, I would move the reader's conditional assessment to unverdictable pending a security-level analysis or an independent upper bound. The analytical calculations of the POVM probabilities are transparent and likely correct, so the concern is about the logical status of Eq. (1) in the realistic model, not about the algebra.","tokens_in":27362,"tokens_out":13819,"duration_ms":140870,"concrete_test":"Independently compute the Devetak-Winter (or entropic-uncertainty) asymptotic key rate for the full photonic protocol with sources of Eq. (2) and PNR detectors of Eq. (5), for a representative PDC case (e.g., lambda=mu=0.1, eta_ch=0.01, eta_det=1, tau=0), without invoking Eq. (1). Compare that rate with Eq. (8) and with -log2(1-eta_ch^2): if the true rate exceeds the repeaterless bound while Eq. (8) does not, the necessary-condition proof fails; if Eq. (8) is shown to upper-bound the true rate for all p_n,q_m, the concern is resolved.","verdict_should_be":"UNVERDICTED","load_bearing_attack":"To prove the corollary, the paper must establish that for every PDC source the actual secret key rate never exceeds the repeaterless bound -log2(1-eta_ch^2). The proof instead derives Eq. (9) from Eq. (8), which comes from the key-rate formula Eq. (1) of Ref. [34]. Eq. (1) is the standard asymptotic key-rate expression for the idealized protocol; in a security proof it is normally a lower bound on the achievable rate. A necessary-condition argument, however, requires an upper bound (or an exact expression) for the true rate after ideal sources are replaced by the realistic multi-photon sources of Eq. (2) with PNR detectors. The paper does not prove that Eq. (8) is such an upper bound. This is not a formality: multiphoton contributions create postselected events in which the retained modes are not qubit pairs, so the complementarity reasoning behind h(eZ)+h(eX) may fail. The paper itself notes in Appendix B.4 that pX_c=pZ_c and pX_nc=pZ_nc are supported only by strong numerical evidence, not an analytical proof; this is a symptom of the missing security analysis for the modified device set. Without an upper-bound statement, the contradiction in the corollary shows only that a heuristic rate formula cannot beat the bound, not that PDC sources cannot.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript examines a realistic implementation of the adaptive measurement-device-independent QKD (AMDI-QKD) protocol of Azuma et al. [34], replacing the ideal entanglement and single-photon sources with sources of the form Eq. (2), including type-II PDC sources, and replacing threshold detectors with photon-number-resolving detectors. The paper derives a simple analytical necessary condition, Eq. (9), on the photon-number statistics of the sources for beating the repeaterless bound, and uses it to prove a corollary that PDC sources cannot beat this bound with AMDI-QKD. It then presents a numerical study of a tighter necessary condition for non-unit detector efficiencies and finite switching latency, based on a full-mode calculation of the relevant probabilities in Appendix B.","tokens_in":27583,"tokens_out":10667,"duration_ms":116681,"significance":"If the corollary were rigorously established, it would be a useful no-go result: the most common entangled-photon source technology would be excluded from the AMDI-QKD approach to beating the repeaterless bound, even with ideal PNR detectors. The paper's analytical derivation of Eq. (8), the Taylor-bound argument leading to Eq. (9), and the PDC contradiction are clean and are supported by a substantial full-mode probability calculation. The main reservation is that the key-rate formula Eq. (1) is used as an exact expression for the rate with multiphoton sources, whereas in the QKD literature it is normally a lower bound for the idealized protocol; the paper does not establish the upper-bound statement that a necessary-condition argument requires. The result is therefore best viewed as a strong conditional statement about the rate formula of Ref. [34], rather than a complete impossibility proof for actual PDC-based implementations.","major_comments":[{"comment":"The proof of the Claim and the Corollary treats Eq. (1) as the exact secret key rate for the realistic sources of Eq. (2). In standard QKD security proofs, Eq. (1) is a lower bound on the achievable key rate for the idealized single-photon protocol of Ref. [34], not an upper bound. A necessary-condition argument requires an upper bound (or an exact expression) on the true rate after the ideal sources are replaced by multiphoton sources with PNR detectors. The paper does not supply such an upper bound: the full-mode calculations in Appendices B.2–B.4 compute detection probabilities, but they do not prove that the complementarity reasoning behind h(eZ)+h(eX) remains valid for postselected events in which the retained modes are not qubit pairs. Without this upper-bound statement, Eq. (8) is not established as a necessary condition, and the contradiction in the Corollary shows only that the heuristic rate formula cannot beat the bound, not that PDC sources cannot.","section":"Sec. II.B / Sec. IV.A"},{"comment":"The manuscript states that the identities pX_c=pZ_c and pX_nc=pZ_nc are supported only by strong numerical evidence and not by an analytical proof. These quantities enter the phase-error rate eX in Eq. (B5), so the numerical necessary condition in Sec. IV.B (Fig. 5) depends on an unproven identity. If the same identity is also used in Appendix B.5 to obtain the ideal-detector rate Eq. (8), then the analytical necessary condition likewise lacks a fully demonstrated derivation. Please provide an analytical proof of these identities, or explicitly restrict the numerical and analytical claims to the regime in which they are proven.","section":"Appendix B.4 / Sec. IV.B"},{"comment":"The monotonicity argument that R(ηdet=1,τ=0) is the best case is physically plausible because lossy detectors can be simulated by perfect detectors preceded by a beamsplitter, but the paper does not prove this for the specific postselected key-rate expression. Since the whole necessary condition relies on comparing the ideal-efficiency rate with the actual rate, this step should be stated as an explicit assumption or proven from the device model.","section":"Sec. IV.A, p.6"}],"minor_comments":[{"comment":"The notation q2 ≤ min(25/96 p1 q1^2, 1−q1) is ambiguous; it should read q2 ≤ min( (25/96) p1 q1^2, 1−q1 ).","section":"Eq. (9)"},{"comment":"The simplification pZ^2 ≈ 1 is used without a precise justification; please state the condition on pZ under which the simulations remain valid.","section":"Sec. II.B"},{"comment":"The sentence claiming that 'we can actually beat the repeaterless bound in this limit' refers to the heuristic rate formula Eq. (8) under the assumption of ideal detectors and negligible dark counts; this context should be made explicit to avoid overstatement.","section":"Sec. IV.A, p.6"},{"comment":"The full-mode formulas in Appendices B.2–B.4 are extremely long and nested; given that Fig. 5 is generated numerically, the authors should provide either the numerical code or a statement about numerical precision and cross-checks of the sums.","section":"Appendix B"}],"recommendation":"major_revision","confidential_remarks":"The central issue is the unproven upper-bound status of Eq. (1) for multiphoton sources. I would not accept the corollary as an impossibility proof without an additional security argument showing that the key-rate expression bounds the actual secret key rate from above, or a full security proof for the modified device set. If the authors can supply such an argument, or alternatively reframe the claims as properties of the rate formula of Ref. [34], the paper could be suitable for publication. The self-citation of Ref. [34] is not itself a problem, but the paper leans heavily on a formula whose validity domain for realistic sources is not established."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe paper gives a clean analytic necessary condition on source photon-number statistics for AMDI-QKD to beat the repeaterless bound, and a neat proof that PDC sources violate it. If the proof stands, it is a useful negative result: the most common entangled-photon source cannot make this particular protocol beat the fundamental limit. The derivation of Eq. (9) is correct and transparent, and the full-mode formulas in the appendices are thorough. Credit is due for that.\n\nThe soft spot is structural. The impossibility argument starts from the key-rate expression of Ref. [34], which is a security-bound-derived rate for idealized devices. In QKD, such expressions are lower bounds on achievable rates. A necessary condition for beating a fundamental bound requires an upper bound on the true rate for the realistic multi-photon sources. The paper does not establish that Eq. (8) is an upper bound. This is not a formality: with multiphoton emissions, the postselected events need not correspond to qubit pairs, so the complementarity reasoning behind the error terms may not apply. The paper itself notes in Appendix B.4 that the equality pX_c=pZ_c and pX_nc=pZ_nc is supported only by numerical evidence, which is a symptom of the missing security analysis.\n\nSo the corollary as stated—“using PDC sources, it is impossible to beat the repeaterless bound”—is not proven. What is proven is that a heuristic rate formula, built from the idealized protocol's expression, cannot beat the bound. That is still interesting, but weaker. If the authors can supply an upper-bound argument (or an exact rate derivation for the realistic device set), the result would be solid. Until then, treat the impossibility as conditional.\n\nThe self-citation of Ref. [34] is not the problem; the formula is presumably correct for its intended setting. The problem is the unstated assumption about its status when the devices change. The paper is candid about its numerical simplifications, which makes it a serious piece of work even with the gap.\n\nMy recommendation: send it to peer review. A good referee can push for the missing upper bound or a more careful statement. The paper is worth engaging, but I would not cite the PDC no-go as established.","headline":"A neat necessary-condition argument with a PDC no-go corollary, but the impossibility claim rests on an unproven upper-bound status of a lower-bound key-rate formula.","tokens_in":28165,"tokens_out":3237,"would_cite":false,"duration_ms":33631,"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":"Adaptive MDI-QKD cannot beat the repeaterless bound with PDC sources.","keywords":["adaptive measurement-device-independent QKD","repeaterless bound","parametric down-conversion","photon-number-resolving detectors","quantum key distribution","full-mode analysis","necessary condition","entanglement sources"],"falsifier":"Measure the photon-number statistics of a PDC source and check whether $q_2 \\le \\min\\!\\left(\\frac{25}{96}\\, p_1 q_1^2,\\, 1-q_1\\right)$; since PDC statistics force the transformed inequality to require a value above $36/25$ while the maximum possible left-hand side is $4/27$, any PDC source satisfying the condition, or any PDC-based AMDI-QKD experiment that exceeds the repeaterless bound under the paper's stated assumptions, would disprove the corollary.","tokens_in":27116,"feed_emoji":"🔐","tokens_out":5820,"duration_ms":59573,"temperature":0.7,"pith_summary":"This paper examines whether the all-optical, adaptive measurement-device-independent QKD (AMDI-QKD) protocol can still surpass the fundamental repeaterless rate-loss bound when its idealized entanglement sources are replaced by realistic ones that occasionally emit multiple photon pairs. Using a full-mode model with photon-number-resolving detectors, the authors derive a necessary condition on the photon-number statistics of the sources for beating the bound. The central corollary is that parametric down-conversion (PDC) sources can never satisfy this condition, so an AMDI-QKD implementation built from them is limited to at most the direct-transmission scaling set by the repeaterless bound. This matters because PDC is the standard source technology, and the result shows that the protocol's advertised square-root rate advantage is fragile exactly where experiments would need it.","feed_headline":"Adaptive MDI-QKD cannot beat the repeaterless bound with PDC sources","feed_subtitle":"The protocol's square-root rate advantage vanishes unless entanglement sources emit almost no multi-photon pairs.","key_machinery":"The load-bearing object is a full-mode model of the protocol in which every rate is reduced to Fock-basis click probabilities for the QND, $Z/X$-measurement, and Bell-state-measurement modules. The central identity is the secret-key rate formula $R = p_Z^2\\, p_s\\, p_{\\mathrm{BSM}}\\,[1 - f\\,h(e_Z) - h(e_X)]$, where $p_s$ appears linearly rather than quadratically because BSMs are performed only on signals that survive the channel. The analytical condition comes from comparing the ideal-efficiency rate $R[\\eta_{\\mathrm{det}}=1,\\tau=0] = \\frac{3 p_1 q_1^2 \\eta_{\\mathrm{ch}}^2}{8 q_2 + 4(3q_1 - 2q_2)\\eta_{\\mathrm{ch}}}$ with the first term $1.44\\,\\eta_{\\mathrm{ch}}^2$ of the Taylor expansion of the repeaterless bound.","core_discovery":"For entanglement sources with photon-number statistics $p_n$ and $q_m$, beating the repeaterless bound requires $q_2 \\le \\min\\!\\left(\\frac{25}{96}\\, p_1 q_1^2,\\, 1-q_1\\right)$ when detectors are ideal. With the statistics of type-II PDC sources, $p_n = \\frac{(n+1)\\lambda^n}{(1+\\lambda)^{n+2}}$ and $q_m = \\frac{(m+1)\\mu^m}{(1+\\mu)^{m+2}}$, this necessary condition reduces to an impossible inequality: PDC statistics can never satisfy the requirement, so PDC sources cannot be used to beat the repeaterless bound. The result holds even with photon-number-resolving detectors, and finite detector efficiency only tightens the required source quality.","pith_inferences":["Beyond the paper's claims: the necessary condition reads as a source-design criterion, suggesting that only sources with strongly sub-Poissonian multi-pair emission, such as deterministic single-photon or quantum-dot-like sources, could plausibly let AMDI-QKD surpass the repeaterless bound.","Beyond the paper's claims: the same full-mode counting method could be applied to other adaptive or memory-assisted QKD proposals to quantify how multi-pair source components degrade their advertised rate-distance scaling.","Beyond the paper's claims: Eq. (9) could be used directly as a source-characterization test: measure $p_1$, $q_1$, and $q_2$ of a candidate source and check the inequality before attempting a full protocol implementation."],"forward_implications":["A PDC-based AMDI-QKD setup, even with ideal photon-number-resolving detection and no feedforward loss, has secret-key rate scaling at most $O(\\eta_{\\mathrm{ch}})$, so it cannot deliver the square-root improvement in rate over distance.","Any entanglement source intended for AMDI-QKD must satisfy $q_2 \\le \\min\\!\\left(\\frac{25}{96}\\, p_1 q_1^2,\\, 1-q_1\\right)$; violating this necessary condition excludes beating the repeaterless bound.","Lower detector efficiency makes the required source quality stricter, and the simulations show the tolerable two-photon probability drops by roughly an order of magnitude per efficiency step in the studied range.","The Hadamard gates placed before the final Bell-state measurement suppress a specific class of errors coming from two-photon components of the QND sources, but that suppression does not rescue PDC photon-number statistics."],"supporting_citations":[{"why":"Defines the AMDI-QKD protocol and supplies the secret-key rate formula that the paper evaluates under realistic entanglement sources.","marker":"[34]"},{"why":"Gives the repeaterless bound $-\\log_2(1-\\eta_{\\mathrm{ch}}^2)$ that the protocol must beat and whose Taylor expansion anchors the necessary condition.","marker":"[5]"},{"why":"Defines the click patterns for a successful Bell-state measurement that both the QND and BSM modules reuse.","marker":"[29]"},{"why":"Provides the type-II parametric down-conversion photon-number statistics used to instantiate the PDC corollary.","marker":"[38]"},{"why":"Reports the analogous failure for ensemble-based memory-assisted MDI-QKD, against which the paper claims a stronger result because it assumes photon-number-resolving detectors.","marker":"[37]"}],"fun_headline_variants":["PDC sources can't beat repeaterless bound in adaptive MDI-QKD","Impossible: PDC sources fail to beat repeaterless QKD bound","Adaptive MDI-QKD: PDC sources never beat repeaterless limit","No beating repeaterless bound with PDC in adaptive MDI-QKD","PDC stats rule out beating repeaterless bound in MDI-QKD"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The secret-key rate formula from the idealized protocol remains valid when the ideal sources are replaced by the realistic photon-number mixture of Eq. (2); if that security proof does not extend, the necessary condition would not apply to the actual protocol.","fun_headline_variants_meta":{"raw":{"variants":["PDC sources can't beat repeaterless bound in adaptive MDI-QKD","Impossible: PDC sources fail to beat repeaterless QKD bound","Adaptive MDI-QKD: PDC sources never beat repeaterless limit","No beating repeaterless bound with PDC in adaptive MDI-QKD","PDC stats rule out beating repeaterless bound in MDI-QKD"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000851,"raw_usage":{"total_tokens":3642,"prompt_tokens":832,"completion_tokens":2810,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":448,"completion_tokens_details":{"reasoning_tokens":2707}},"tokens_in":448,"tokens_out":2810,"duration_ms":18133,"temperature":1.0,"reasoning_tokens":2707,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:39:48.107579+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the photon-number statistics of a PDC source and check whether $q_2 \\le \\min\\!\\left(\\frac{25}{96}\\, p_1 q_1^2,\\, 1-q_1\\right)$; since PDC statistics force the transformed inequality to require a value above $36/25$ while the maximum possible left-hand side is $4/27$, any PDC source satisfying the condition, or any PDC-based AMDI-QKD experiment that exceeds the repeaterless bound under the paper's stated assumptions, would disprove the corollary.","supporting_citations":[{"cited_title":"Azuma, K","cited_arxiv_id":null,"evidence_quote":"Defines the AMDI-QKD protocol and supplies the secret-key rate formula that the paper evaluates under realistic entanglement sources."},{"cited_title":"Pirandola, R","cited_arxiv_id":null,"evidence_quote":"Gives the repeaterless bound $-\\log_2(1-\\eta_{\\mathrm{ch}}^2)$ that the protocol must beat and whose Taylor expansion anchors the necessary condition."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the click patterns for a successful Bell-state measurement that both the QND and BSM modules reuse."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the type-II parametric down-conversion photon-number statistics used to instantiate the PDC corollary."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Reports the analogous failure for ensemble-based memory-assisted MDI-QKD, against which the paper claims a stronger result because it assumes photon-number-resolving detectors."}],"review_version":1}