{"id":"a69a5abc-0ffa-46a1-ae16-9c82408b2842","arxiv_id":"2501.09637","paper_version":1,"verdict":"REJECT","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"high","formal_verification":"none","parameter_count":0,"one_line_summary":"The paper claims to generate contextual hypergraphs in arbitrary dimensions, to correct a standard contextuality inequality, and to offer new quantum communication protocols.","lead":"A physicist proposes new ways to build and identify Kochen-Specker sets, which are collections of quantum measurements that rule out classical hidden variables, and showcases examples up to 32 dimensions. The paper also argues that a standard inequality used to detect contextuality is unreliable and sketches new uses in quantum cryptography and computation.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Claimed α-inequality violations vanish if α* is computed as the LP optimum of Def. II.9; Eq. (13) equates α* with the feasible point k/n, not the maximum.","rationale":"","tokens_in":45367,"tokens_out":6510,"duration_ms":73414,"concrete_test":"Re-run the LP in §II.C for hypergraphs 9-3 (Figure 4) and 25-15 (Figure 5) with bounds x_v ∈ [0,1] and the same edge constraints as Def. II.9, i.e. without the added lower bound x_v ≥ 1/n. If the optimum for 9-3 is 3 rather than 9/4, and for 25-15 is at least 7 rather than 25/4 = 6.25, then the asserted violations of α ≤ α* are artifacts of the added lower bound and the central claim fails. Any standard LP solver (e.g. scipy.optimize.linprog or Mathematica LinearProgramming) suffices for the check.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that α(H) ≤ α*(H) fails under quantum measurements, with Eq. (15) giving α > α* = k/n for e.g. 9-3 and 25-15. But α* in Def. II.9/II.10 is the LP maximum over x(v) ∈ [0,1] with hyperedge sums ≤ 1. For any hypergraph, every 0/1 independent set is feasible, so α ≤ α* is a theorem, not an empirical discriminator. The paper's LP calculations impose lower bounds x_v ≥ 1/n; the given LP call for 9-3 uses bounds {{1/4,1},...}, imported from Quantum Indeterminacy Postulate II.11. That lower bound is not part of Def. II.9. With it, every n-vertex hyperedge has sum at least 1, so feasibility forces every x_v = 1/n and α* equals k/n by construction. Removing the lower bound, the same 9-3 LP has optimum 3, not 2.25, and inequality (11) holds. The later α*_p = l (Eq. 16) is a restatement of the handshake identity Σ m(v) = nl, not a contextuality witness; it holds for binary and non-binary hypergraphs alike. Thus the paper's main negative claim about the α-inequality is internally inconsistent with its own definition of α*.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops hypergraph (MMPH) representations of Kochen–Specker and non-Kochen–Specker contextual sets, presents generation methods M1–M8, catalogs examples in dimensions 3–32, and proposes applications in quantum communication and computation. Its central theoretical claim is that the fractional-independence α-inequality α(H) ≤ α*(H) fails under quantum measurements, and that new 'postprocessed' statistics yield a reliable discriminator α < α*_p = l.","tokens_in":45681,"tokens_out":6581,"duration_ms":67939,"significance":"If the inequality analysis were correct, the paper would substantially change how contextuality is quantified and would supply a scalable toolkit for higher-dimensional KS sets. The paper's strengths are its computational generation methods, the explicit coordinatizations and strings in the appendix, and the reproducible catalog of MMPHs in dimensions up to 32. However, the central inequality-based claims are not sound, and the proposed new discriminator is tautological; the useful computational content does not rescue the advertised theoretical message.","major_comments":[{"comment":"The claimed violation of the α-inequality for 9-3 and 25-15 is an artifact of replacing the LP optimum in Def. II.9 by the raw-data sum k/n. In Eq. (13), α*(H) is equated with k/n, but Def. II.9 defines α* as the maximum of Σ x(v) over x(v) ∈ [0,1] with Σ_{v∈e} x(v) ≤ 1 for every hyperedge. The paper's own LP computation for 9-3 with free variables gives α* = 3 and α = 3, so Eq. (11) holds; the computation yielding 9/4 imposes the additional lower bounds x(v) ≥ 1/4, which are not part of Def. II.9. Since any 0/1 independent set is feasible for the LP, α ≤ α* is a theorem for every hypergraph, and a measurement protocol cannot invalidate it unless it changes the meaning of α*.","section":"II.C, Defs. II.9–II.10 and Eq. (13)"},{"comment":"The postulate that an unknown pure state exits through each port of a gate with equal probability 1/n is not a consequence of quantum mechanics. For a pure state |ψ⟩ measured in an orthonormal basis {|v_i⟩}, the probability of outcome i is |⟨v_i|ψ⟩|^2; this is 1/n only for special states and bases, and equals 1 for |ψ⟩ = |v_1⟩ in that basis and 0 for the other ports. The postulate is load-bearing: it is used to justify Eq. (13), the claimed α* values such as 9/4 and 25/4, and the assertion that α* computations are of linear complexity. Without it, the claimed violations of Eq. (11) do not follow.","section":"II.C, Quantum Indeterminacy Postulate II.11"},{"comment":"The postprocessed quantum fractional independence number α*_p is defined in Eq. (14) as Σ_v m(v)/n = l. Hence the α*_p-inequality α < α*_p = l in Eq. (16) is not an independent contextuality witness; it is the handshake identity of Lemma II.4 together with the definition of α*_p. The paper acknowledges that Eq. (16) is 'another form of the v-inequality' (Eq. (8)), but the v-inequality itself, as stated in Lemma II.6, is not proved: the argument that 'the maximal number of hyperedges that contain 1 must be smaller than l' concerns hyperedge counts, while HI_cM in Eq. (8) is a vertex count. Thus the proposed 'reliable discriminator' is either tautological or unsupported.","section":"II.C, Def. II.13, Eq. (14), Theorem II.14(b), Eq. (16)"}],"minor_comments":[{"comment":"The phrase 'the α-inequality (Equations (11,16))' appears to be a typo; Eq. (16) is the α*_p-inequality, so the reference should probably be to Eqs. (11) and (15) or simply to Eq. (11).","section":"II.C, sentence after Eq. (16)"},{"comment":"The caption states 'α = 7 > α* = 25/4 = 6.25'; this is an instance of the artifact described in major comment 1, since 25/4 is the value of the uniform feasible point under the lower-bound restriction, not the LP optimum of Def. II.9.","section":"Figure 5 caption"},{"comment":"The table labels are inconsistent; the text refers to 'Table II D 1' and 'Tables II D 1–IV', which should be unified to stable table numbers.","section":"Tables I–IV and surrounding text"},{"comment":"The distribution of critical KS MMPHs would be much easier to read as a formatted table; the current line '9 (number of hyperedges, l) 18 (number of vertices, k) (1) (number of MMPHs)' is ambiguous.","section":"Appendix A.2"},{"comment":"The claim that Eve's probability of correctly guessing all states is less than 3 × 10^-17 for the 32-dim, 11-hyperedge MMPH is stated without derivation; if retained, it needs a calculation or a precise citation.","section":"II.E.1"}],"recommendation":"reject","confidential_remarks":"The manuscript's technical core is unsound because of the mislabeling of α* and the unsupported Quantum Indeterminacy Postulate; this is not a matter of style. The computational catalog and generation methods might be salvageable as a separate contribution, but the advertised inequality results and the reliability claims would have to be removed or replaced, which is beyond a minor revision. I would also note that the paper's arguments rely extensively on the author's previous work; while that is not in itself problematic, the new claims should be checked against the standard definition of fractional independence number."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know. The useful core is the generation work: concrete KS and non-KS MMPHs in dimensions up to 32, explicit strings and coordinatizations, and the M8 dimensional upscaling method that plausibly does not scale in complexity with dimension. Those are real, reproducible contributions—the programs are public and the examples are new. The paper's advertised conceptual result, that the alpha-inequality alpha(H) ≤ alpha*(H) fails for quantum measurements, does not survive contact with the paper's own definitions.\n\nThe stress-test note is right. Definition II.9 defines alpha* as the LP maximum. For every hypergraph, any independent 0/1 set is feasible, so alpha ≤ alpha* is a theorem, and it holds for the 9-3 example: without the extra lower bound x_v ≥ 1/4, the LP optimum is 3, not 2.25. The paper obtains the lower bound from the Quantum Indeterminacy Postulate, but that postulate is not part of Definition II.9, is not derived from quantum mechanics, and is false for arbitrary states. The claimed violation is an artifact of computing a constrained feasible point and calling it the optimum. The postprocessed alpha*_p = l is likewise a restatement of the handshake identity and the v-inequality, not a new contextuality witness. The e- and v-inequalities are close to definitions; calling them noncontextuality inequalities is defensible only in a narrow sense.\n\nWhere the paper is on solid ground: M8, the new examples, and the empirical observation that minimal non-KS hyperedges stay small with dimension. The applications section is a sketch, not a protocol analysis; the Eve-probability estimate is not a security proof. The discussion of state-independent operator inequalities contains a legitimate concern about operator products in [61], though it is not developed.\n\nWho this is for: specialists who work on KS-set generation and MMPHs will want the data and the method. The paper should not be taken as establishing a failure of the alpha-inequality. In revision, Section II.C would need to be rewritten or removed; the generation results stand alone.\n\nRecommendation: send to peer review rather than desk reject, because the constructive parts are concrete and citable and a good referee can separate them from the flawed inequality claim. But expect a major revision, and the inequality section should not survive in its current form.","headline":"Useful new KS-set data and a plausible upscaling method, but the paper's central inequality claim is a mislabeled LP feasible point, not a violation.","tokens_in":46238,"tokens_out":2566,"would_cite":false,"duration_ms":30348,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["81P13","05C65","90C05"],"pacs":["03.65.Ta","03.67.Dd"],"model":"deepseek-v4-flash","headline":"The standard α ≤ α* inequality fails as a contextuality test under quantum measurement statistics; the postprocessed inequality α < α*_p = l replaces it, and contextual hypergraphs can be generated in any dimension up to 32.","keywords":["quantum contextuality","Kochen–Specker sets","non-Kochen–Specker sets","MMP hypergraphs","fractional independence number","contextuality inequalities","dimensional upscaling","higher-dimensional Hilbert spaces"],"falsifier":"Measure the exit statistics of a Stern–Gerlach-type device for a definite preparation: send spin-1 atoms prepared in the $|S_z = 0\\rangle$ (or any fixed) state through the apparatus and count detections at the three output ports; the rates are 0, 1, and 0 (up to experimental error), not 1/3 each, directly contradicting the postulate that drives $\\alpha^* = k/n$. Equivalently, compute the fractional-independence LP for the 25-15 hypergraph using actually measured per-vertex probabilities of its given coordinatization rather than uniform $1/n$, and check whether $\\alpha \\le \\alpha^*$ holds; if it does, the claimed violation is an artifact of the uniformity assumption, not a property of quantum measurements.","tokens_in":45161,"feed_emoji":"⚛️","tokens_out":18904,"duration_ms":160857,"temperature":0.7,"pith_summary":"This paper is a bid to reset the standard way of certifying quantum contextuality and to make contextual state sets available in arbitrary dimensions. Its central claim is that the widely used inequality $\\alpha(H) \\le \\alpha^*(H)$ — comparing the independence number of a contextuality hypergraph with its fractional independence number — is not a reliable test for contextuality once actual quantum measurement statistics are taken into account: under the paper's Quantum Indeterminacy Postulate, each of the $n$ vertices of a measured hyperedge is detected with probability $1/n$, so the relevant sum is $k/n$, which can exceed $\\alpha$ (the 4-dim 25-15 example has $\\alpha = 7$ while $\\alpha^* = 25/4 = 6.25$). The paper replaces it with the postprocessed inequality $\\alpha < \\alpha^*_p$, where $\\alpha^*_p$ equals the number of hyperedges $l$, a consequence of the hypergraph handshake lemma, and argues that this, together with the v- and e-inequalities, reliably separates contextual from noncontextual hypergraphs. It also claims to generate Kochen–Specker and non-Kochen–Specker hypergraphs — MMPHs, connected hypergraphs whose hyperedges contain at most $n$ mutually orthogonal vectors — in any dimension by methods whose complexity does not grow with dimension, with new examples up to dimension 32, and sketches applications in quantum key distribution, oblivious communication, Hadamard-matrix construction, and stabilizer operations. If right, the paper reshapes which inequalities are cited as contextuality certificates and supplies scalable higher-dimensional state sets, the raw material for proposed quantum communication and computation advantages.","feed_headline":"Fails under quantum data: the α ≤ α* contextuality test","feed_subtitle":"Equal 1/n port probabilities flip the bound; hyperedge count l becomes the reliable test, up to dimension 32.","key_machinery":"The carrying object is the MMP hypergraph (McKay–Megill–Pavičić hypergraph): a connected hypergraph of dimension $n$ in which each hyperedge contains at most $n$ vertices, vertices represent quantum states or vectors, and hyperedges represent sets of mutually orthogonal vectors serving as measurement contexts. Two numbers do the work in the inequality argument: the independence number $\\alpha(H)$ — the largest set of vertices no two of which share a hyperedge, equal to the paper's maximal classical vertex index — and the fractional independence number $\\alpha^*(H)$, the optimum of a linear program maximizing the sum of vertex weights subject to each hyperedge summing to at most 1. The paper's new object is the postprocessed fractional independence number $\\alpha^*_p$, in which each vertex contributes its multiplicity $m(v)$ divided by $n$; the vertex-hyperedge lemma (a hypergraph handshake lemma) forces $\\sum_v m(v)/n = l$, the number of hyperedges, and this identity converts the unreliable bound $\\alpha \\le \\alpha^*$ into the reliable bound $\\alpha < \\alpha^*_p = l$. The generation machinery is a family of eight methods M1–M8, most prominently M1 (exhaustive enumeration of mutually orthogonal $n$-tuples from simple vector components) and M8 (dimensional upscaling by concatenating lower-dimensional KS sets with zero padding, whose complexity does not scale with dimension). The Quantum Indeterminacy Postulate — uniform exit probability $1/n$ per port — is the bridge from linear programming to quantum measurement statistics.","core_discovery":"The paper's discovery, stated on its own terms, is that quantum measurement statistics break the $\\alpha$-inequality and that a postprocessed statistic restores a reliable test. Under the Quantum Indeterminacy Postulate II.11 — an unknown pure state exits any gate (hyperedge) with equal probability $1/n$ through each of its $n$ ports — the fractional independence number $\\alpha^*(H)$ of an $n$-dim $k$-$l$ MMPH becomes $k/n$ under raw data statistics. The paper exhibits MMPHs for which $k/n$ is strictly smaller than the independence number $\\alpha$: the 4-dim non-KS NBMMPH 25-15 with $\\alpha = 7 > \\alpha^* = 25/4 = 6.25$, and even the noncontextual BMMPH 9-3 with $\\alpha = 3 > \\alpha^* = 9/4 = 2.25$ — so the inequality $\\alpha \\le \\alpha^*$, long asserted as a noncontextuality inequality, is not a reliable discriminator of contextuality. The replacement is the postprocessed quantum fractional independence number $\\alpha^*_p$, which counts each vertex with its multiplicity: summing $m(v)/n$ over vertices gives exactly $l$ by the vertex-hyperedge lemma, so the inequality $\\alpha < \\alpha^*_p = l$ is a form of the v-inequality and, the paper argues, a genuine noncontextuality inequality. On the generation side, the paper claims that by combining lower-dimensional KS MMPHs padded with zeros in new dimensions (method M8) and by filtering masters built from simple vector components (method M1), one obtains KS and non-KS MMPHs in all dimensions with minimal critical size fluctuating between 8 and 16 hyperedges, and it presents new critical examples in dimensions 27 and 32.","pith_inferences":["The claimed failure of the $\\alpha$-inequality is driven entirely by the uniformity postulate; for prepared (known) states the exit probabilities are $|\\langle v|\\psi\\rangle|^2$ and need not be $1/n$, so the statement that quantum measurements break $\\alpha \\le \\alpha^*$ is only as general as the postulate's domain of unknown states.","A direct experimental test would measure detection statistics for the 9-3 BMMPH and the 25-15 non-KS MMPH under definite preparations: if $\\alpha \\le \\alpha^*$ is restored under state-dependent probabilities, the discriminator claim becomes a statement about ensemble preparation, not about contextuality per se.","The paper concedes that its maximal classical vertex indices come from randomized searches without backtracking, so slightly larger $\\alpha$ values might appear with more runs; because larger $\\alpha$ only widens the claimed violations ($\\alpha > k/n$), this admitted uncertainty actually points in the direction of the paper's conclusion.","The large-alphabet protocol's advantage is argued against an eavesdropper who knows only the uniform exit statistics; an extended adversarial model in which Eve knows Alice's actual prepared states may erase the claimed advantage, since the protocol's secrecy appears to rest on the uniformity postulate rather than on contextuality itself."],"forward_implications":["The $\\alpha$-inequality $\\alpha \\le \\alpha^*$ should not be used to certify contextuality: under the paper's measurement statistics it is violated not only by contextual NBMMPHs but also by the noncontextual BMMPH 9-3, so a violation of it proves nothing about contextuality.","The reliable discriminators become the v-inequality, the e-inequality $l_{cM} < l$, and the postprocessed inequality $\\alpha < \\alpha^*_p = l$; all contextual MMPHs fail the 0-1 assignment rules that these inequalities encode.","Contextual sets can be produced in any dimension by dimensional upscaling, with minimal critical size fluctuating in a narrow band (8–9 hyperedges for non-KS, 9–16 for KS) rather than growing with the dimension.","Higher-dimensional MMPHs supply concrete resource states: the 32-dim 144-11 set gives an eavesdropper-guessing probability below $3 \\times 10^{-17}$ in the proposed large-alphabet protocol, and star-like KS hypergraphs determine the elements of S-Hadamard matrices in any even dimension.","Stabilizer-operation graphs can be translated into non-KS MMPHs and reduced to smaller critical KS MMPHs, offering a route to simpler stabilizer circuits and error-correction codes."],"supporting_citations":[{"why":"Supplies the MMPH formalism, coordinatization methods, the earlier v- and e-inequality results, and the alpha/alpha* tables the present argument builds on.","marker":"[3]"},{"why":"The graph-theoretic source asserting the alpha-inequality alpha <= alpha* as a noncontextuality inequality; the claim the paper sets out to overturn.","marker":"[66]"},{"why":"Supplies the port-exit probability picture used to ground the Quantum Indeterminacy Postulate of uniform 1/n exit probability.","marker":"[67]"},{"why":"Defines the independence number and the fractional-independence linear program whose definitions underpin the inequality analysis.","marker":"[48]"},{"why":"Introduces the dimensional-upscaling method M8 whose complexity does not scale with dimension, and supplies the 27-dim 141-16 master.","marker":"[47]"},{"why":"Establishes automated M1 generation of KS masters from simple vector components, including the golden-ratio 3-dim and 9-dim classes.","marker":"[50]"},{"why":"Provides the 4-, 6-, 8-, 16-, and 32-dim masters, including 144-11, from which new non-KS subhypergraphs are derived.","marker":"[55]"},{"why":"Connects contextuality to stabilizer operations via the 30-108 NBMMPH whose reduction is analyzed in the applications section.","marker":"[29]"},{"why":"The theorem linking S-Hadamard matrices to star-like KS hypergraphs that the paper's hypergraph generation inverts.","marker":"[84]"}],"fun_headline_variants":["Quantum contextuality test fails; postprocessed statistic holds to dim 32","The α ≤ α* test fails; new statistic holds to dim 32","Contextuality test broken; use postprocessed statistic, dim 32","Quantum data breaks α ≤ α*; postprocessed statistic works to dim 32"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The Quantum Indeterminacy Postulate — that an unknown pure state exits through each port of a measurement gate with equal probability $1/n$ — is the load-bearing premise, and it is not derived from quantum mechanics; for a prepared state the exit probabilities follow the state's amplitudes, so the equality $\\alpha^* = k/n$, and hence the claimed violation of the $\\alpha$-inequality, holds only under this assumed uniformity.","fun_headline_variants_meta":{"raw":{"variants":["Quantum contextuality test fails; postprocessed statistic holds to dim 32","The α ≤ α* test fails; new statistic holds to dim 32","Contextuality test broken; use postprocessed statistic, dim 32","Quantum data breaks α ≤ α*; postprocessed statistic works to dim 32"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001471,"raw_usage":{"total_tokens":5982,"prompt_tokens":1081,"completion_tokens":4901,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":697,"completion_tokens_details":{"reasoning_tokens":4820}},"tokens_in":697,"tokens_out":4901,"duration_ms":30491,"temperature":1.0,"reasoning_tokens":4820,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T19:48:47.807170+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the exit statistics of a Stern–Gerlach-type device for a definite preparation: send spin-1 atoms prepared in the $|S_z = 0\\rangle$ (or any fixed) state through the apparatus and count detections at the three output ports; the rates are 0, 1, and 0 (up to experimental error), not 1/3 each, directly contradicting the postulate that drives $\\alpha^* = k/n$. Equivalently, compute the fractional-independence LP for the 25-15 hypergraph using actually measured per-vertex probabilities of its given coordinatization rather than uniform $1/n$, and check whether $\\alpha \\le \\alpha^*$ holds; if it does, the claimed violation is an artifact of the uniformity assumption, not a property of quantum measurements.","supporting_citations":[{"cited_title":"Quantum Contextuality","cited_arxiv_id":null,"evidence_quote":"Supplies the MMPH formalism, coordinatization methods, the earlier v- and e-inequality results, and the alpha/alpha* tables the present argument builds on."},{"cited_title":"The Feynman Lectures on Physics; Volume III","cited_arxiv_id":null,"evidence_quote":"Supplies the port-exit probability picture used to ground the Quantum Indeterminacy Postulate of uniform 1/n exit probability."},{"cited_title":"Generation Kochen-Specker Contextual Sets in Higher Dimensions by Dimen- sional Upscaling Whose Complexity Does not Scale with Dimension and Their Application","cited_arxiv_id":null,"evidence_quote":"Introduces the dimensional-upscaling method M8 whose complexity does not scale with dimension, and supplies the 27-dim 141-16 master."},{"cited_title":"Automated Generation of Ar bitrarily Many Kochen-Specker and Other Contextual Sets in Odd Dimensional Hilbert Spaces","cited_arxiv_id":null,"evidence_quote":"Establishes automated M1 generation of KS masters from simple vector components, including the golden-ratio 3-dim and 9-dim classes."},{"cited_title":"Arbitrarily Exhaustive Hypergraph Gene ration of 4-, 6-, 8-, 16- , and 32-Dimensional Quantum Contextual Sets","cited_arxiv_id":null,"evidence_quote":"Provides the 4-, 6-, 8-, 16-, and 32-dim masters, including 144-11, from which new non-KS subhypergraphs are derived."},{"cited_title":"Kochen-Specker sets and Hadamard matrice s","cited_arxiv_id":null,"evidence_quote":"The theorem linking S-Hadamard matrices to star-like KS hypergraphs that the paper's hypergraph generation inverts."}],"review_version":1}