{"id":"295b5628-8473-4c61-a973-3e8be3d779ef","arxiv_id":"2412.00199","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A coarse-grained, noncontextual weak-measurement experiment can certify contextuality of the fine-grained pure-state experiment it is built from.","lead":"The paper designs a six-protocol experiment using weak measurements that is noncontextual, i.e. classically explainable, when the measured state is Kirkwood-Dirac positive, but that still lets Bob certify that another experiment Alice performed on the underlying pure states is contextual. This shows the classical/nonclassical boundary is not preserved under coarse-graining.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The verification step is non-constructive: Lemma 8 guarantees a bound δ > 0 without supplying it, so Bob cannot actually choose ε satisfying δ > 3d²ε; the claimed experiment is not yet executable.","rationale":"The paper's technical machinery is largely sound: the contextuality witness (Theorem 5, Supplementary Note III) is proven with explicit ε thresholds, and the noncontextual model for KD-positive states (Supplementary Note IV) is an explicit construction whose probability bounds are checked via Lemma 7. The main result, however, depends on the ability of Bob to verify δ > 3d²ε. Lemma 8 only asserts the existence of δ by compactness and hyperplane separation; it gives no value or method. This is a genuine load-bearing gap for the claim that an experimenter can verify contextuality, because without a computable δ the protocol cannot be run with any guaranteed choice of ε. The reader's weakest_assumption identified both the imported exotic-state existence and the non-constructive δ; I agree with the latter as the more immediate internal concern, while treating the former as an external citation-dependent assumption. Since the reader already issued a CONDITIONAL verdict, my analysis does not move the verdict; it sharpens the condition under which the paper would be accepted: an explicit exotic state and a computable positive lower bound on its decomposition negativity must be supplied, or an alternative argument giving Bob a usable ε must be provided.","tokens_in":18572,"tokens_out":21754,"duration_ms":217753,"concrete_test":"Take an explicit exotic KD-positive state ρ⋆ from Ref. [26] in the smallest dimension where it exists, with nondegenerate A and B satisfying P_j ≠ Π_k for all j,k. Compute the convex-hull distance δ between ρ⋆ and conv(E_pure_KD+) via the semidefinite programming formulation of that membership problem; if δ > 0, choose ε = δ/(6d²) and verify (a) the Note IV noncontextual hidden-variable model has all probabilities in [0,1] at this ε, and (b) Theorem 1 part 1 applies to the guaranteed ψ−. If no explicit exotic state with a computable δ is supplied, or if the O(ε²) truncation in Eqs. (6)–(8) prevents tomographic confirmation of N(ρ⋆) = 0, then the central verification claim remains unsupported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central scenario requires Bob to certify contextuality of Alice's postselected experiment by checking the inequality δ > 3d²ε, where δ is a lower bound on the KD-negativity of some pure state in every decomposition of the exotic state ρ⋆. The existence of such a δ is proven in Note V, Lemma 8, but the proof is purely existential: it uses the hyperplane separation theorem and the extreme value theorem to conclude that δ = N(ρ_min) > 0, without providing any algorithm, explicit bound, or dependence of δ on ρ⋆. Since Bob does not know Alice's decomposition and only reconstructs ρ⋆ tomographically, he has no way to compute δ or to fix a weak-measurement strength ε that is guaranteed to satisfy δ > 3d²ε. δ could in principle be arbitrarily small, so no predetermined choice of ε is safe. Consequently, the described protocol is not an executable verification procedure: the advertised 'construction' of a noncontextual experiment that verifies contextuality is incomplete. The imported existence of exotic KD-positive states from Ref. [26] is a further external assumption, but the immediate internal gap is the missing computable δ.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper connects generalized contextuality to the Kirkwood-Dirac (KD) quasiprobability distribution for a pair of nondegenerate observables A and B. It presents six measurement protocols (weak and projective) and proves, in Theorem 1, two partial results: if a state has KD-nonpositivity N(ρ)>3d²ε, the protocols are contextual (for small weak-measurement strength ε), and if ρ is KD-positive (N(ρ)=0), the protocols admit an explicitly constructed noncontextual hidden-variable model. The main claimed application is an Alice–Bob scenario: Alice sends a sequence of pure states that average to an 'exotic' KD-positive mixed state ρ⋆ (KD-positive but not a convex mixture of pure KD-positive states). Bob's experiment on ρ⋆ is noncontextual by Theorem 1, yet by Lemma 8 every pure-state decomposition of ρ⋆ contains a state with KD-nonpositivity above some δ>0; if δ>3d²ε, Bob can certify that Alice's experiment (or a postselected part of it) is contextual. The paper argues this gives a noncontextual experiment that verifies contextuality.","tokens_in":18766,"tokens_out":9477,"duration_ms":86663,"significance":"The conceptual claim—that a demonstrably noncontextual experiment can certify contextuality of another experiment—is novel and interesting, and the paper has real strengths: the noncontextual hidden-variable model in Note IV is explicit and is checked against both the correctness constraints and the noncontextuality constraints; the contextuality direction in Note III is proved with a concrete threshold ε<δ/(3d²); and the connection between KD-positivity, convex roofs, and exotic states is clearly articulated. If the central construction can be made executable, the result would be a valuable contribution to the study of generalized contextuality. However, the main verification step currently rests on a non-constructive existence statement for the bound δ, which is load-bearing; and the abstract's 'iff' claim is stronger than the proven theorems.","major_comments":[{"comment":"The verification step is non-constructive. Lemma 8 proves the existence of δ>0 via the hyperplane separation theorem and the extreme value theorem, but it provides no way to compute δ from ρ⋆ or from Bob's tomographic data. In the protocol, Bob must choose the weak-measurement strength ε before running the experiment and must ensure δ>3d²ε; with no lower bound on δ, no predetermined ε is guaranteed to satisfy this inequality, and Bob cannot certify that his chosen ε is small enough. Consequently, the advertised construction of a noncontextual experiment that verifies contextuality is not executable as stated. The authors should either supply a computable lower bound for δ (for the specific exotic state or family used) or reformulate the claim as an existence result with an unspecified parameter.","section":"Experiment and analysis; Note V, Lemma 8"},{"comment":"The abstract claims that the experiment is 'contextual iff the underlying state is not KD-positive,' but Theorem 1 proves only two partial statements: contextuality when N(ρ)>3d²ε and noncontextuality when N(ρ)=0. For states with 0<N(ρ)≤3d²ε, the theorem is silent, so the 'if' direction of the 'iff' is not established. Please revise the claim to match the proven threshold, e.g., by stating the contextuality criterion as N(ρ)>3d²ε, or prove the missing direction.","section":"Abstract; Theorem 1"}],"minor_comments":[{"comment":"In Eq. (8) of the main text, the first term should be (1/2) Tr(Π^z_k ρ), not (1/2) Tr(Π^z_j ρ); the subscript must match the kth eigenspace of B.","section":"Main text, Eq. (8)"},{"comment":"The sentence 'When ϵ=π/2, then N_{x,j} and M_{x,j} are projective' should refer to M_{y,j}, since the Y-type Kraus operator was defined just before.","section":"Main text, after Eq. (5)"},{"comment":"The text refers to 'Protocols (9)–(11)' and 'protocols (6)–(8)' where the intended referents are equations, not protocols; please rephrase to avoid confusion.","section":"Main text, equations (9)–(11)"},{"comment":"The weak-measurement unitary is written as U = I⊗P_j + Z⊗P_j^⊥, which interchanges control and target; it should be P_j⊗I + P_j^⊥⊗Z (with the system factor first). The subsequent Kraus operators are consistent with the corrected form.","section":"Supplementary Note I, Fig. 1"},{"comment":"The acknowledgments contain the typo 'EPRSC'; presumably this should be EPSRC.","section":"Acknowledgments"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is within the journal's scope and the underlying idea is publishable in principle. The non-constructive bound δ in Lemma 8 is the central obstacle: if the authors can provide a constructive lower bound for a concrete exotic state, or otherwise make the verification step executable, the paper would be substantially stronger. The abstract should also be aligned with the threshold theorem. I do not see circularity or grounds for rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThe thing to know: this paper gives a genuinely new twist—a noncontextual experiment (Bob's) can certify contextuality of a finer-grained experiment (Alice's) built from the same data. The mechanism is clever: Bob sees only the exotic KD-positive mixture ρ⋆, which admits a noncontextual HV model (Note IV constructs it), but from the public data he can infer that any pure-state decomposition must contain a state with KD-negativity bounded below, and that state's postselected experiment is contextual. That is a real result, and it does not reduce to the earlier weak-value contextuality results [31,32]; the explicit HV model for KD-positive states is new.\n\nThe main-text proofs are mostly solid. The contextuality direction (Note III) gives a clean threshold: N(ρ)>3d²ε implies no noncontextual model. The noncontextual direction (Note IV) is an explicit construction with the ε<√5/5 condition. I checked the algebra in the supplementary; it works.\n\nSoft spots, in order of severity. First, the abstract's 'iff' is not what Theorem 1 proves. The theorem covers N=0 and N>3d²ε; states with 0<N≤3d²ε are unclassified at fixed ε. That is an overclaim and should be fixed. Second, the Alice-Bob verification step is non-constructive. Lemma 8 guarantees δ>0 by hyperplane separation and compactness, but does not give a way to compute δ from ρ⋆. Bob, who only knows ρ⋆ tomographically, is said to 'check δ>3d²ε'; he cannot, unless someone hands him a δ. For an existence proof of the phenomenon this is acceptable; for a claimed 'construction of an experiment' it is incomplete. The paper would be stronger with a worked example of an exotic state where δ is computable, or at least an explicit algorithm. Third, the exotic-state existence is imported from Ref. [26] without stating the dimensions or observables for which those states arise; if the known examples are only in special dimensions, the scenario's scope is narrower than the abstract suggests. That is minor if the reference is correct, which it appears to be.\n\nWho is this for? People working on generalized contextuality and quasiprobability witnesses. They will want to see the δ gap addressed, but the core idea is sound and worth refereeing. I would send it to a serious referee.\n\nBest.","headline":"Genuinely new conceptual result on KD-positivity and Spekkens contextuality, but the abstract overclaims an 'iff' and the Alice-Bob verification step rests on a non-constructive δ that Bob cannot in practice compute.","tokens_in":19335,"tokens_out":2728,"would_cite":true,"duration_ms":26137,"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":"The paper claims that a noncontextual experiment can certify another experiment's contextuality, via exotic KD-positive mixtures whose pure decompositions always contain a highly nonpositive component.","keywords":["generalized contextuality","Kirkwood-Dirac distribution","KD-positive states","exotic states","weak measurements","hidden-variable models","noncontextuality constraints","quasiprobability distributions"],"falsifier":"Compute, for a candidate exotic KD-positive state in the protocol's dimension and observable pair, the convex-roof KD-negativity—the smallest negativity among pure states in any decomposition. If that minimum is at most $3d^2\\epsilon$ for the weak-coupling strength used, Lemma 8 cannot supply the required $\\delta$ and Bob's certification step fails; exhibiting such a state would directly falsify the paper's central claim. Equivalently, writing down a noncontextual hidden-variable model for Alice's postselected data when a component has $N(\\psi_-)>3d^2\\epsilon$ would contradict the first part of Theorem 1.","tokens_in":18373,"feed_emoji":"⚛️","tokens_out":7669,"duration_ms":63057,"temperature":0.7,"pith_summary":"This paper claims that contextuality—the failure of an experiment to admit a classical hidden-variable description—can be certified by an experiment that is itself noncontextual. The bridge is the Kirkwood-Dirac (KD) quasiprobability distribution: for a six-protocol scheme of weak and projective measurements, the authors show the experiment is contextual when the state's KD-negativity exceeds $3d^2\\epsilon$, and noncontextual when the state is KD-positive. The key move is to feed the scheme an 'exotic' mixed state—one that is KD-positive as a whole, so the receiving experiment has a noncontextual model, but whose every decomposition into pure states contains a pure state with large KD-negativity. The receiving experimenter can therefore verify, from public data plus knowledge of the sending order, that the sender's postselected experiment was contextual. If the construction is right, the quantum-classical boundary can be detected from its classical side.","feed_headline":"Noncontextual experiment can certify contextuality","feed_subtitle":"A classical-looking weak-measurement setup lets Bob prove Alice's hidden ordering produced a contextual experiment.","key_machinery":"The central objects are the Kirkwood-Dirac (KD) quasiprobability distribution $Q_{j,k}(\\rho)=\\mathrm{Tr}(\\Pi_k P_j \\rho)$ and its nonpositivity $N(\\rho)=-1+\\sum_{j,k}|Q_{j,k}(\\rho)|$. The argument runs through six protocols: projective measurements of the observables whose eigenprojectors define the KD distribution, $X$- and $Y$-type weak measurements of $P_j$, and two procedures—randomly sampling an outcome and applying a dephasing-type channel—that are quantum-theoretically indistinguishable from parts of the weak-measurement protocols. Those indistinguishability relations become noncontextuality constraints that any hidden-variable model must obey. For KD-positive states the authors explicitly construct a noncontextual hidden-variable model on the ontic space $\\Lambda=\\{1,\\dots,d\\}$; for states with $N(\\rho)>3d^2\\epsilon$ they prove no such model exists. The exotic-state ingredient—a KD-positive mixed state not decomposable into pure KD-positive states—is what turns Bob's noncontextual experiment into a certificate of Alice's contextuality, because every pure decomposition must contain a component with negativity bounded away from zero.","core_discovery":"The central claim is that a receiving experimenter (Bob) can certify contextuality of a sender's (Alice's) experiment while Bob's own measurement procedure is noncontextual. Bob receives many copies of an exotic KD-positive mixture $\\rho_\\star = \\frac{1}{N}\\sum_j \\psi_j$, whose KD distribution is a genuine probability distribution, so by the second part of Theorem 1 his six protocols admit a noncontextual hidden-variable model. Yet Bob's data allow an informationally complete reconstruction of $\\rho_\\star$, and from the fact that $\\rho_\\star$ is exotic an argument based on hyperplane separation gives a $\\delta>0$ such that every pure-state decomposition of $\\rho_\\star$ contains a pure state $\\psi_-$ with negativity $N(\\psi_-)>\\delta$. When $\\delta>3d^2\\epsilon$, the first part of Theorem 1 applies to Alice's postselected trials and shows that no noncontextual hidden-variable model can describe her experiment. In short, public noncontextual data certify a contextual preparation.","pith_inferences":["A natural experimental extension would be to identify concrete observables and a concrete exotic state in low dimension for which the nonconstructive $\\delta$ of Lemma 8 can be computed; without such a bound the protocol is an existence proof rather than a runnable recipe.","The convexity of $N(\\rho)$ suggests a general design principle: any convex nonclassicality witness can be 'hidden' in a mixture even when every pure component is nonclassical, so certification tasks should target convex roofs or other decomposition-sensitive quantities rather than the witness value of the mixture.","In device-certification scenarios, the result implies that a party who sees only coarse-grained, classically simulable statistics may still certify nonclassicality about the data-generating process; one testable consequence is a contextuality-based certification scheme that does not require the certifier's own devices to be contextual."],"forward_implications":["If the construction holds, an experimenter whose own procedure admits a classical description can nevertheless certify that a colleague's more finely indexed experiment is contextual.","KD-nonpositivity becomes a quantitative, faithful witness for this family of experiments: a threshold $N(\\rho)>3d^2\\epsilon$ guarantees contextuality, while KD-positivity guarantees a noncontextual model.","Exotic KD-positive states acquire an operational role: their special convex structure lets noncontextual bulk data reveal contextuality in the fine-grained decomposition.","The result sharpens the analogy with entanglement: just as entangled states can admit local hidden-variable models, a contextual experiment can be verified from procedures that are individually noncontextual.","Because Bob's conclusion uses only public protocol choices and outcomes, the verification protocol works even though Bob does not know the ordering of Alice's states; that ordering information is the resource that 'unmixes' the mixture."],"supporting_citations":[{"why":"Establishes the existence of exotic KD-positive states—mixed KD-positive states not decomposable into pure KD-positive states—that the Alice-Bob scenario requires.","marker":"[26]"},{"why":"The weak-value contextuality results that Theorem 1's first part rephrases in terms of KD-nonpositivity.","marker":"[31, 32]"},{"why":"Supplies the KD-distribution formalism, including the invertibility used to reconstruct $\\rho_\\star$ from the measured KD distribution.","marker":"[20]"},{"why":"Defines generalized contextuality and the noncontextual hidden-variable model framework that the paper's constraints and construction use.","marker":"[11]"},{"why":"Establishes that $N(\\rho)=0$ exactly for KD-positive states, the identification underlying the witness.","marker":"[22]"},{"why":"Shows that for almost all observables KD-positive states are just mixtures of basis states, so exotic states require specially chosen $A$ and $B$.","marker":"[25]"},{"why":"Introduces the convex roof of KD-nonpositivity that the discussion uses to quantify the pure-state-level nonpositivity of mixed states.","marker":"[40]"}],"fun_headline_variants":["Noncontextual test proves contextuality","Weak-measurement setup certifies contextuality","Contextuality confirmed by noncontextual data","Certifying contextuality without contextual measurements"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The construction depends on the existence of 'exotic' mixed states that are KD-positive—their Kirkwood-Dirac distribution is an ordinary probability distribution—but cannot be written as mixtures of pure KD-positive states for the observables used in the protocols; this existence result is imported from earlier literature and not re-proved here.","fun_headline_variants_meta":{"raw":{"variants":["Noncontextual test proves contextuality","Weak-measurement setup certifies contextuality","Contextuality confirmed by noncontextual data","Certifying contextuality without contextual measurements"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00045,"raw_usage":{"total_tokens":2226,"prompt_tokens":863,"completion_tokens":1363,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":479,"completion_tokens_details":{"reasoning_tokens":1309}},"tokens_in":479,"tokens_out":1363,"duration_ms":9224,"temperature":1.0,"reasoning_tokens":1309,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T05:38:35.643601+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute, for a candidate exotic KD-positive state in the protocol's dimension and observable pair, the convex-roof KD-negativity—the smallest negativity among pure states in any decomposition. If that minimum is at most $3d^2\\epsilon$ for the weak-coupling strength used, Lemma 8 cannot supply the required $\\delta$ and Bob's certification step fails; exhibiting such a state would directly falsify the paper's central claim. Equivalently, writing down a noncontextual hidden-variable model for Alice's postselected data when a component has $N(\\psi_-)>3d^2\\epsilon$ would contradict the first part of Theorem 1.","supporting_citations":[{"cited_title":"Lupu-Gladstein, Y","cited_arxiv_id":null,"evidence_quote":"Supplies the KD-distribution formalism, including the invertibility used to reconstruct $\\rho_\\star$ from the measured KD distribution."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes that $N(\\rho)=0$ exactly for KD-positive states, the identification underlying the witness."}],"review_version":1}