Pith. sign in

REVIEW 4 major objections 5 minor 31 references

Every minimal unextendible product basis in 3×odd bipartite dimensions is graph-equivalent to a single contextuality-based construction, linking contextual strength to bound entanglement.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · deepseek-v4-flash

2026-08-04 07:05 UTC pith:RLMFEGT2

load-bearing objection New construction and a nice KCBS/Pyramid identification, but Theorem 2's classification rests on an unproven connectivity assumption and the GenPyramid observations contradict each other. the 4 major comments →

arxiv 2510.26719 v2 pith:RLMFEGT2 submitted 2025-10-30 quant-ph

Graph theoretic quantum contextuality and unextendible Product Bases

Gurvir Singh , Arvind This is my paper
classification quant-ph MSC 81P1381P4005C50 PACS 03.67.-a
keywords quantum contextualityunextendible product basesLovász theta numbergraph equivalenceKCBS inequalitybound entanglementorthogonal representationsPaley graphs
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper builds a graph-theoretic bridge between quantum contextuality and unextendible product bases (UPBs). It first identifies the five vectors of the Pyramid UPB with the KCBS contextual vectors and constructs a one-parameter family interpolating between the Tiles and Pyramid UPBs, showing that the contextual strength of the underlying vectors tracks the linear entropy of entanglement of the associated bound-entangled states. Generalizing this, the paper proves that generalized KCBS vectors are equivalent to GenPyramid UPB vectors in prime dimensions and also exist for certain odd-perfect-square non-prime dimensions. The central new object is the GenContextual UPB, built from Lovász-optimal orthogonal representations of odd cycle graphs and their complements, and the main theorem states that every minimal UPB in C^3⊗C^{d_2} with odd d_2 is graph-equivalent to it. A sympathetic reader would care because the result unifies two previously separate phenomena and points to a quantitative tie between contextuality and entanglement.

Core claim

On the paper's own terms, the central discovery is Theorem 2: for a bipartite Hilbert space C^3⊗C^{d_2} with d_2 odd, any two minimal UPBs of cardinality n=d_2+2 are graph-equivalent, so every minimal UPB in that dimension has the same orthogonality structure as the newly constructed GenContextual UPB. The proof colors the edges of the complete orthogonality graph by which party detects orthogonality; minimality forces the party-1 factor to be a 2-regular graph (assumed connected, hence an odd n-cycle), and the party-2 factor is its complement. The paper also establishes a quantitative correspondence: within a one-parameter family of UPBs in C^3⊗C^3, the contextual strength C(C_5,{α_j(θ)}) o

What carries the argument

The load-bearing objects are orthogonal representations of graphs and the orthogonality graphs of UPBs. An orthogonal representation assigns to each vertex a vector so that adjacent vertices get orthogonal vectors; a Lovász-optimal orthogonal representation (LOOR) is one that achieves the Lovász theta number ϑ(G). The GenContextual UPB is formed by taking the LOOR of an odd cycle C_n in C^3 and the LOOR of its complement C_n in C^{n-2}, and multiplying them to get product vectors in C^3⊗C^{n-2}; Lemma 1, a graph-theoretic characterization of UPBs, verifies unextendibility. Graph equivalence of UPBs is defined by a permutation that preserves the party-wise orthogonality coloring, and Theorem

Load-bearing premise

The proof of Theorem 2 assumes without proof that the party-1 orthogonality graph of a minimal UPB in C^3⊗C^{d_2} is connected; degree counting gives only 2-regularity, which permits disjoint unions of cycles, and without connectedness the n-cycle structure and the graph-equivalence conclusion do not follow.

What would settle it

Exhibit a minimal UPB in C^3⊗C^5 (n=7) whose party-1 orthogonality graph is a disjoint union of a 3-cycle and a 4-cycle; such an example would violate Theorem 2's claim that every minimal UPB is graph-equivalent to GenContextual. More directly, a reader can check the missing step: prove or disprove that the 2-regular party-1 graph of any minimal UPB must be connected.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

If this is right

  • All minimal UPBs in C^3⊗C^{d_2} (d_2 odd) share the same orthogonality graph as GenContextual UPB, making it the canonical representative for studying their distinguishability and entanglement.
  • The quantitative link between contextual strength and linear entropy of entanglement gives a new way to rank UPBs by the contextual resources of their local vectors.
  • Since GenContextual UPB is minimal, Cohen's conjecture that all minimal UPBs are indistinguishable under the topological closure of LOCC can be tested on this new family.
  • The observation that QuadRes UPB vectors are LOORs of Paley graphs identifies Paley graphs as a source of noncontextuality inequalities with high contextual ratio ϑ(G)/α(G).
  • The extensibility of GenPyramid UPBs to non-prime odd-perfect-square dimensions (p=9,25) widens the known scope of GenPyramid UPB constructions.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If the connectedness assumption in Theorem 2 can be proved from the UPB conditions alone, the graph-equivalence result would be unconditional; if a disconnected 2-regular party-1 graph exists for some minimal UPB, there would be a second, non-equivalent orthogonality class and the theorem would need modification.
  • The observed correlation between contextual strength and entanglement entropy suggests a possible monotone: UPBs whose local vectors achieve higher Lovász-optimal contextual strength may yield bound-entangled states with systematically higher entanglement, a conjecture that could be tested on the full six-parameter UPB family.
  • The Paley-graph connection points toward a broader principle: any graph whose orthogonal representation is LOOR and whose complement also admits an LOOR in the complementary dimension may generate a new UPB, so the LOOR–UPB correspondence may extend beyond cycles and Paley graphs.
  • One testable extension is to compute the separability/LOCC-distinguishability of GenContextual UPB directly; the paper's SDP verification that GenTiles2 and GenContextual UPBs are SEP-distinguishable leaves open whether GenContextual UPB resists LOCC, as conjectured.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper claims a graph-theoretic correspondence between quantum contextuality and unextendible product bases (UPBs). It identifies the Pyramid UPB vectors with KCBS vectors, proposes a one-parameter family of five vectors in C^3 whose associated product states are claimed to form UPBs in C^3⊗C^3, and reports a numerical correlation between the contextual strength C(C5,{α_j(θ)}) and the linear entropy of entanglement (LEE) of the associated bound-entangled state. It then generalizes the correspondence to odd cycle graphs: the GenPyramid UPB is related to generalized KCBS vectors, and a new class of minimal UPBs, called GenContextual UPB, is constructed from LOORs of C_n and their complements. The central structural theorem asserts that every minimal UPB in C^3⊗C^{d_2} with odd d_2 is graph-equivalent to the corresponding GenContextual UPB. In the reverse direction, the QuadRes UPB is related to LOORs of Paley graphs, and noncontextuality inequalities based on Paley graphs are suggested.

Significance. If established, the paper would provide a useful bridge between two well-studied quantum information notions, contextuality and UPBs, and would add a classification result for minimal UPBs in C^3⊗C^{d_2}. The construction of GenContextual UPBs from LOORs of cycles and complements is appealing, and the observation connecting QuadRes UPBs to Paley graphs is a promising direction. The manuscript also has the merit of stating a concrete quantitative conjecture (Table I). However, the central claims are not supported in the current version: the proposed one-parameter family does not satisfy the stated orthogonality relations, the proof of the classification theorem contains an unproved and load-bearing connectivity assumption, and the proof of the UPB criterion has an unproved spanning claim. These issues affect the main results of the paper, so the contribution is not yet acceptable.

major comments (4)
  1. [Contextuality-UPB correspondence, Eq. (1), Fig. 1] The vectors in Eq. (1) do not realize the orthogonality graph in Fig. 1. With α0 = sin2θ|0⟩ - sinθcosθ|2⟩ + cosθ|1⟩ and α2 = cosθ|0⟩ + sinθ|2⟩, one has ⟨α0|α2⟩ = sinθ cos^2θ, which is nonzero for generic θ (e.g., θ=π/3 gives √3/8). Thus the supposed party-1 edge (0,2) is absent. Since the second factors are β_j = α_{2j mod 5}, the product states j=0 and j=2 are also not orthogonal on party 2 for the same generic θ, so Eq. (2) is not a UPB. This invalidates the claimed one-parameter UPB family, Observation 1, and the quantitative correspondence in Table I.
  2. [Proof of Theorem 2] The proof asserts without argument that 'the green factor (party-1 orthogonality graph) is 2-regular and connected'. Neither part follows from the preceding conditions. A 2-regular graph on odd n need not be connected, e.g., C3 ∪ C4 for n=7, and no argument connects the UPB condition of Lemma 1 to connectedness of the party-1 graph. If the green factor is disconnected, Lemma 2 cannot be applied and the chosen permutation mapping A's cycle onto B's cycle need not exist. Also, the proof states that 'every edge of K_n receives exactly one color', but Definition 1 allows an edge to be orthogonal on more than one party. The possibility of double-colored edges must be ruled out; otherwise the red/green partition of K_n is not justified. Since Theorem 2 is the central classification claim, this gap is load-bearing.
  3. [Proof of Theorem 1] The proof of unextendibility relies on the claims that in the LOOR of C_n in C^3 any set of three or more vectors spans C^3, and that in the LOOR of the complement in C^{n-2} any set of n−2 or more vectors spans C^{n-2}. These are asserted without proof. In general, d vectors in C^d need not be linearly independent, so it is essential to establish this property for the specific representations in Eq. (10) and Eq. (11). Without this, the union bound |W_A| + |W_B| ≤ 2 + (n−3) does not follow, and the GenContextual construction is not proven to be a UPB.
  4. [Observations 2 and 3] There is a direct internal contradiction. Observation 2 states that valid GenPyramid UPBs exist for non-prime p=9 and p=25, with a necessary condition that p be an odd perfect square. Observation 3 states that a valid UPB exists only when 2m+1 is prime. These claims cannot both hold. The manuscript must be revised to clarify the exact parameter regime (for instance, whether the non-prime examples require m≠t) and to correct the assertion in Observation 3.
minor comments (5)
  1. [Eq. (10)] The definition of θ_j says j=1,...,n, but the vectors are otherwise indexed j=0,...,n−1, and no definition is given for j=0. This should be made consistent.
  2. [Eqs. (6)-(7)] The definition of contextual strength says vectors are orthogonal 'whenever i and j are adjacent in complement graph G (i.e., nonadjacent in G)', while the text also calls C5 the 'orthogonality graph' and states that ϑ(C5)=√5 is the maximal KCBS violation. The graph-theoretic convention should be stated explicitly to avoid confusion.
  3. [Table I] The LEE values are reported without specifying the bound-entangled state (e.g., the normalized projector onto the orthogonal complement of the UPB) or the method used to compute the minimization in Eq. (5). Without this information the numerical results are not reproducible.
  4. [Observation 1] The statement 'for every nonzero value of θ the vectors |α_j⟩ provide contextual vectors' is too broad: at θ=π/2 the vector α0 vanishes, and at θ=0 one has α0=α4, so the set is not a valid five-vector representation.
  5. [Throughout] There are minor typographical and formatting issues, including 'Unextendable' in the Introduction versus 'Unextendible' in the title, and inconsistent spacing in names such as 'GenPyramidUPB' and 'GenContextualUPB'.

Circularity Check

0 steps flagged

No significant circularity: the paper's constructions and quantitative comparisons use independent external inputs and free parameters; the noted proof gaps in Theorem 2 are correctness risks, not circular reductions.

full rationale

Walked the derivation chain. The theta-family in Eqs. (1)-(2) is a subset of the six-parameter UPB family from [7]; theta is a free parameter, not fitted. The contextual strength C(C5,{alpha_j(theta)}) in Eq. (6) and LEE in Eq. (5) are evaluated independently from the same vectors/state; Table I is a numerical correlation, not a prediction forced by construction. Observations 1 and 3 are direct coordinate identifications between Eqs. (3)/(4) and (8)/(10); no target quantity is inserted into the construction. Theorem 1 checks the UPB conditions of [4] for a construction using LOORs from [23]; all inputs are external and the conclusion is not assumed. Theorem 2 contains unsupported assertions that the party-1 orthogonality graph is '2-regular and connected' and that every edge receives exactly one color; these are proof gaps/correctness risks, not circularity, because the premise does not define the graph as a cycle and no fitted parameter or self-citation makes the conclusion identical to the input. Observations 2 and 3 state contradictory conditions about prime/non-prime p (9 and 25 vs 'only when 2m+1 is prime'); this is an internal consistency problem, not circularity. No load-bearing self-citations: refs [4], [7], [23] are independent prior work by other groups. Therefore the circularity score is 0.

Axiom & Free-Parameter Ledger

2 free parameters · 4 axioms · 0 invented entities

No new physical entities are introduced; the free parameters are family parameters in known UPB constructions. The main hidden burden is the connectivity assumption in Theorem 2, which is an ad hoc paper-specific premise.

free parameters (2)
  • θ = cos^{-1}((√5−1)/2) (Pyramid), 3π/4 (Tiles), π/3, π/6, π/12 in Table I
    Sweeps the one-parameter UPB family in Eq. (1); it is a genuine free parameter used to compare contextual strength with LEE.
  • (m, t) / p = (2,2) giving p=5, (4,3) giving p=9, (12,10) giving p=25
    Discrete parameters in the GenPyramid construction, chosen by hand to obtain UPBs; the claimed necessary condition 'p odd perfect square' is not proved.
axioms (4)
  • domain assumption Lemma 1 of Shi et al.: a set of product vectors forms a UPB iff the union of the local orthogonality graphs is complete and no tuple of unsaturated subsets covers all vertices.
    Used without proof in the proof of Theorem 1; it is an external characterization result.
  • domain assumption The explicit vectors in Eq. (11) form a LOOR of the complement graph C_n in dimension n−2 (from Cabello et al. [23]).
    Load-bearing for the GenContextual UPB construction; the paper does not re-derive the optimality.
  • ad hoc to paper The party-1 orthogonality graph of any minimal UPB in C3 ⊗ C^{d_2} is connected.
    Asserted in the proof of Theorem 2 but not proved; required to apply Lemma 2 and conclude the graph is an n-cycle.
  • domain assumption ϑ(G) gives the maximal quantum violation of the corresponding noncontextuality inequality, and ϑ(G) > α(G) marks a quantum contextual graph.
    Used throughout to identify contextual strength and to connect contextuality to UPBs via CSW graph theory.

pith-pipeline@v1.3.0-alltime-deepseek · 9039 in / 18336 out tokens · 160207 ms · 2026-08-04T07:05:58.875371+00:00 · methodology

0 comments
read the original abstract

Unextendible product bases(UPBs) are central to the study of local distinguishability of orthogonal product states. While their connection to quantum nonlocality via Bell inequalities is well established, their link to quantum contextuality remains largely unexplored. We establish a graph theoretic connection between contextuality and UPBs. First, an equivalence between Klyachko-Can-Binicio\u{g}lu-Shumovsky (KCBS) vectors and the Pyramid UPB is shown and then by constructing a one parameter family of UPB vectors, a quantitative connection between `contextuality strength' and bound entanglement of states associated with the corresponding UPB is demonstrated. This equivalence is extended to generalized KCBS vectors and the GenPyramid UPB. A new class of minimal UPBs in $\mathbb{C}^3 \otimes \mathbb{C}^n$ is constructed using Lov\'asz-optimal orthogonal representations (LOORs) of cycle graphs and their complements which we term the GenContextual UPB. Any minimal UPB in this dimension is shown to be graph-equivalent to the GenContextual UPB. We briefly discuss the distinguishability properties of GenContextual UPB. In the reverse direction, we observe that the constituent vectors of the QuadRes UPB are LOORs of Paley graphs. The structural properties of these graphs make them suitable candidates for constructing noncontextuality inequalities, thereby establishing a bidirectional connection between quantum contextuality and UPBs.

Figures

Figures reproduced from arXiv: 2510.26719 by Arvind, Gurvir Singh.

Figure 1
Figure 1. Figure 1: Using these five vectors, we can construct a one pa￾rameter family of UPB’s (a subset of the six parameter family introduced in [7]) in C 3 ⊗ C 3 given by the vectors: |αj ⟩ ⊗ |βj ⟩ = |αj ⟩ ⊗ |α2j mod 5⟩ (2) This one parameter family of UPB’s parameterized by θ is local unitarily equivalent to Tiles UPB for θ = 3π/4 and to the Pyramid UPB for θ = cos−1 ( √ 5 − 1)/2, the two canonical representatives of the… view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

31 extracted references · 2 linked inside Pith

  1. [1]

    Cabello, S

    A. Cabello, S. Severini, and A. Winter, (non-)contextuality of physical theories as an axiom (2010), arXiv:1010.2163 [quant- ph]

    Pith/arXiv arXiv 2010

  2. [2]

    , WN ), where each Wm is an unsaturated set inGm, we haveSN m=1 Wm ̸= V

    For everyN-tuple(W 1, W2, . . . , WN ), where each Wm is an unsaturated set inGm, we haveSN m=1 Wm ̸= V. In our case, the orthogonality graphs are defined as follows: GA =C n (the cycle graph), associated with the local vectors {|uj⟩} ⊂C3, andG B = C n (the complement of the cycle graph), associated with{|v j⟩} ⊂Cn−2. The first condition is met asC n ∪ C ...

  3. [3]

    Cabello, S

    A. Cabello, S. Severini, and A. Winter, Phys. Rev. Lett.112, 040401 (2014)

    2014

  4. [4]

    Lov ´asz, M

    L. Lov ´asz, M. Saks, and A. Schrijver, Linear Algebra and its Applications114-115, 439 (1989), special Issue Dedicated to Alan J. Hoffman

    1989

  5. [5]

    F. Shi, G. Bai, X. Zhang, Q. Zhao, and G. Chiribella, Phys. Rev. Res.5, 033144 (2023)

    2023

  6. [6]

    C. H. Bennett, D. P. DiVincenzo, T. Mor, P. W. Shor, J. A. Smolin, and B. M. Terhal, Physical Review Letters82, 6 5385–5388 (1999)

    1999

  7. [7]

    C. H. Bennett, D. P. DiVincenzo, C. A. Fuchs, T. Mor, E. Rains, P. W. Shor, J. A. Smolin, and W. K. Wootters, Phys. Rev. A59, 1070 (1999)

    1999

  8. [8]

    D. P. DiVincenzo, T. Mor, P. W. Shor, J. A. Smolin, and B. M. Terhal, Communications in Mathematical Physics238, 379–410 (2003)

    2003

  9. [9]

    S. M. Cohen, Phys. Rev. A77, 012304 (2008)

    2008

  10. [10]

    S. M. Cohen, Phys. Rev. A105, 022207 (2022)

    2022

  11. [11]

    S. M. Cohen, Phys. Rev. A107, 012401 (2023)

    2023

  12. [12]

    Bandyopadhyay, A

    S. Bandyopadhyay, A. Cosentino, N. Johnston, V . Russo, J. Wa- trous, and N. Yu, IEEE Transactions on Information Theory61, 3593–3604 (2015)

    2015

  13. [13]

    Kochen and E

    S. Kochen and E. P. Specker, Journal of Mathematics and Me- chanics17, 59 (1967)

    1967

  14. [14]

    Cabello, S

    A. Cabello, S. Filipp, H. Rauch, and Y . Hasegawa, Phys. Rev. Lett.100, 130404 (2008)

    2008

  15. [15]

    Cabello, Phys

    A. Cabello, Phys. Rev. Lett.101, 210401 (2008)

    2008

  16. [16]

    Badzia ¸ g, I

    P. Badzia ¸ g, I. Bengtsson, A. Cabello, and I. Pitowsky, Phys. Rev. Lett.103, 050401 (2009)

    2009

  17. [17]

    A. A. Klyachko, M. A. Can, S. Binicio ˘glu, and A. S. Shu- movsky, Phys. Rev. Lett.101, 020403 (2008)

    2008

  18. [18]

    Augusiak, J

    R. Augusiak, J. Stasi ´nska, C. Hadley, J. K. Korbicz, M. Lewen- stein, and A. Ac´ın, Phys. Rev. Lett.107, 070401 (2011)

    2011

  19. [19]

    Augusiak, T

    R. Augusiak, T. Fritz, M. Kotowski, M. Kotowski, M. Pawłowski, M. Lewenstein, and A. Ac ´ın, Phys. Rev. A85, 042113 (2012)

    2012

  20. [20]

    Fritz, A

    T. Fritz, A. Sainz, R. Augusiak, J. B. Brask, R. Chaves, A. Leverrier, and A. Ac ´ın, Nature Communications4, 10.1038/ncomms3263 (2013)

    doi:10.1038/ncomms3263 2013

  21. [21]

    A. B. Sainz, T. Fritz, R. Augusiak, J. B. Brask, R. Chaves, A. Leverrier, and A. Ac´ın, Phys. Rev. A89, 032117 (2014)

    2014

  22. [22]

    T ´oth, T

    G. T ´oth, T. Moroder, and O. G ¨uhne, Phys. Rev. Lett.114, 160501 (2015)

    2015

  23. [23]

    Lov ´asz, IEEE Transactions on Information theory25, 1 (1979)

    L. Lov ´asz, IEEE Transactions on Information theory25, 1 (1979)

    1979

  24. [24]

    Cabello, L

    A. Cabello, L. E. Danielsen, A. J. L ´opez-Tarrida, and J. R. Por- tillo, Phys. Rev. A88, 032104 (2013)

    2013

  25. [25]

    Alon and L

    N. Alon and L. Lov´asz, Journal of Combinatorial Theory, Series A95, 169 (2001)

    2001

  26. [26]

    Lov ´asz and K

    L. Lov ´asz and K. Vesztergombi, Paul Erdos and his Mathemat- ics2(2009)

    2009

  27. [27]

    Broere, D

    I. Broere, D. D ¨oman, and J. N. Ridley, Quaestiones Mathemat- icae11, 91 (1988)

    1988

  28. [28]

    Magsino, D

    M. Magsino, D. G. Mixon, and H. Parshall, Linear pro- gramming bounds for cliques in paley graphs (2019), arXiv:1907.05971 [math.CO]

    Pith/arXiv arXiv 2019

  29. [29]

    Cabello, Phys

    A. Cabello, Phys. Rev. Lett.110, 060402 (2013)

    2013

  30. [30]

    Marques, J

    B. Marques, J. Ahrens, M. Nawareg, A. Cabello, and M. Bourennane, Phys. Rev. Lett.113, 250403 (2014)

    2014

  31. [31]

    Y .-H. Yang, F. Gao, G.-B. Xu, H.-J. Zuo, Z.-C. Zhang, and Q.- Y . Wen, Scientific Reports5, 11963 (2015)

    2015