Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

Sharp qutrit discrimination and unavoidable correlation overcounting for complementary measurements

Anonymous

Published Nov 4, 2026• 0 citations
Publisher

Abstract

Unrefereed research candidate. Two results on complementary (mutually unbiased) measurements and the complementary quantum correlation (CQC) inequality of Schneeloch, Broadbent and Howell (2014), which proposed that the classical correlations in two complementary measurement tables never exceed the quantum mutual information. 1. Sharp qutrit discrimination. For every pair of complementary bases of a qutrit and every two orthogonal states, the triangular discrimination summed over the two bases is at most 8/3 (the trivial ceiling is 4), answering Qian, arXiv:2608.14806v2, Section 11(4). Every equality case is classified; for the computational and Fourier bases they are tied to the nine Hesse SIC rays. Consequences: choosing one of the two measurements at random recovers at most two thirds of the quantum Fisher information in a two-state estimation task, and Qian's two-ray qubit-qutrit family obeys CQC with margin at least 0.1005 nats. The proof is computer-assisted: an exact rational sum-of-squares certificate (scout_qutrit/exact_sos_certificate.json, 660,609 bytes, SHA-256 825bb8108096d38a44f63f1bc91964405d189ad57b5f568530b5d8e27d598c2f) checked by a standalone exact verifier; this deposit is the certificate's citable locator. 2. Every prescribed pair fails for a qubit and a qudit of dimension n ≥ 4. For every prescribed pair of complementary measurements on each side there is a separable state violating CQC, with a uniform certified margin below −9/20,000,000 nats (previously the every-pair statement was known only for n ≥ 164). A phase-averaging extension of the known shared-label mechanism covers every prescribed pair when both dimensions are at least three; an earlier, unchecked every-pair statement for equal dimensions by E. O'Connor is credited. Open: the full qubit-qutrit CQC question and the sharp constant in dimension 5 (numerics only). The margins are tiny and structural, not experimental. The archive contains the manuscript (PDF and LaTeX), the exact certificate and verifier, negative controls, an exact rational margin checker, numerical cross-checks, working proof notes, novelty audits, the bibliography with registry records, CI and SHA-256 sums. Assurance: computer-assisted exact rational arithmetic, not a formal proof; internal AI-assisted review only (not human peer review or independent reproduction). Creator: Anonymous. Text CC0-1.0; code MIT.

DOI
10.5281/zenodo.23933730

Related papers

Separable Counterexamples to Complementary Quantum Correlations, and Why Random Search Missed Them

Lilong Qian

The complementary quantum correlations (CQC) relation bounds the sum of two classical mutual informations, obtained from local mutually unbiased measurements, by the quantum mutual information of the premeasurement state. We refute it. Separable rank-two counterexamples exist in every local dimension pair \(m\times n\) with \(m,n\ge3\), with closed-form excess at least \(1/(8m^2n^2)\) nats, and in every qubit--qudit pair \(2\times n\) with \(n\ge3\) except \(n=3,5\); every covered pair also admits full-rank separable counterexamples. We then analyse the two residual qubit--qudit dimensions. A dimension-free entropy envelope replaces the natural quadratic majorant and lowers the requirement for closing the equal-prior orthogonal two-ray family from a triangular-discrimination bound \(S\le8/3\) to \(S\le3.8265583\ldots\); the associated gate matrix has trace exactly two, so its spectral test collapses to a single eigenvalue-free scalar; and the exact identity \(X=1-4\operatorname{Var}(c)\) turns the prime-Fourier full-spark barrier into a variance bound. A \(128\)-bit interval cover then closes that family at \(2\times3\) with gap at least \(0.012021\) nats, and at \(2\times5\) an exact saturator attaining \(S=(14+2\sqrt5)/5\) refutes three competing routes. We also give a state-dependent corrected inequality that is universal, and show it is incomparable with CQC already at \(2\times2\). Finally we quantify why the original searches had essentially no power to find these states: the violating set is a sliver against the low-rank boundary, the witness lies \(7.3\) standard deviations below the Hilbert--Schmidt mean, the bases must be aligned to about nine degrees (\(\sim10^{-21}\) of frames), and extrapolating the sample minimum demands \(10^{10}\) to \(10^{18}\) samples against the \(10^7\) ever run.

quant-pharXiv:2608.14806•Aug 18, 2026•1 citation

Beyond Two Queries: A Sharp Robustness Constant and a Three-Query Qubit–Qutrit Separation for Centrally Symmetric Quantum Correlations

Vasily Psykovsky

Companion to the preprint “A Qubit–Qutrit Equivalence for Centrally Symmetric Two-Query Quantum Correlations” (doi:10.5281/zenodo.24093433), which proved that for two fixed binary measurements without shared randomness, a qutrit is never better than a qubit at reproducing centrally symmetric target tables. This note settles two questions left open there. Approximately symmetric targets (two queries). If a qutrit fits a table with worst-cell error ε₃, a sharp qubit pair fits it with error at most ε₃ + δ/2, where δ measures the asymmetry of the table. The constant is proved to be best possible. Three queries. The equivalence fails. Every qubit triple of binary measurements satisfies a simple inequality I ≤ 9, which follows from an identity on 3×3 correlation matrices. A triple of projective qutrit measurements reproduces an eight-point centrally symmetric table exactly with I ≥ 9.0505, so every qubit needs worst-cell error larger than 9.7×10⁻⁴ on it. This is a computer-assisted proof based on exact rational certificates, checked by two independent programs. The qutrit advantage is small: about 10⁻³ in worst-cell error. Numerically, the best qutrit value found is I* = 9.0507284088558548…. A candidate minimal polynomial of degree 13 is reported, confirmed to 3000 digits. Its symbolic derivation and the global optimality of I* remain open, and are labelled as numerical findings. Files: the preprint (PDF), and a reproducibility archive (ZIP) with all exact certificates, the original scripts, an independent audit, the full research record including failed and superseded runs, and a SHA-256 manifest. Quick check after unpacking: python3 -I audit/verify_certificates.py . (pure Python, about 1 s). Not peer reviewed. A large language model assisted with derivations, computations, certificate scripts and drafting; the author takes full responsibility for the content.

•Nov 11, 2026•2 citations

Antipodal Tables as a Quantum Dimension Witness: Blind Spots, a Classical-Trit Theorem and Noise Thresholds

Vasily Psykovsky

Eighth note in the QMEM series on qubit-versus-qutrit antipodal tables. Preprint, version 1; not peer reviewed. This note places the results of the series in the standard prepare-and-measure framework for quantum dimension witnesses. A sender prepares one of m pairs of states, a receiver performs one of n binary measurements, and the observed correlators define a half-difference table D = ½(E⁺ − E⁻), evaluated using linear witnesses and the factorization norm γ₂. Main results Qubit bound. Every qubit strategy satisfies γ₂(D) ≤ 1, even with shared randomness and without assuming antipodal correlations. Dimension versus quantumness. Without the antipodal restriction, qutrits attain γ₂ = 2/√3, for example using four mutually unbiased bases in ℂ³, verified in exact arithmetic. Classical trits attain the same value on a 2 × 2 table. Moreover, shared randomness can transform any strategy into one with antipodal correlations while preserving its half-difference table. Thus antipodality alone does not establish a quantum advantage. Classical-trit theorem. For independent preparation and measurement devices, classical trits satisfy γ₂(D) ≤ 1 on exactly antipodal data, whereas qutrits can exceed this bound. A new exact rational certificate establishes γ₂ ≥ 1.007522, improving the previous lower bound of 1.007479. Together with the universal upper bound proved in Note 7, this gives 1.007522 ≤ κ ≤ 3/(2√2) ≈ 1.058972. Consequently, exactly antipodal correlations with γ₂(D) > 1 exclude qubits and classical trits under the stated independence assumption. Blind spots of dimension witnesses. Previous no-advantage theorems identify configurations in which γ₂(D) cannot exceed one: pairs with a common average state, balanced antipodal pairs, traceless measurements, and measurements annihilating a state whose eigenvalues are all at most ½. Noise robustness. For a classical trit with antipodality defect ε, an explicit construction attains γ₂(D)² = (4 − 3ε)/(4(1 − ε)). The conjecture that this is also the universal upper bound is supported by numerical searches but remains unproved. The witness for exactly antipodal data and its exclusion of qubits under depolarising noise are rigorous. The approximate visibility threshold of 0.993 against classical trits is conditional on the conjecture and the specified noise model. Experimental considerations. Idealised statistical estimates suggest approximately 55,000 trials per setting for a 5σ violation using the certified antipodal table, compared with a visibility threshold of 0.866 and a few hundred trials for a dimension-only test based on mutually unbiased bases. These estimates do not constitute a complete finite-statistics protocol. Device independence is an experimental assumption, not a property certified by the witness. The note builds on established research in prepare-and-measure dimension witnesses, including Gallego et al. (2010), Wehner et al. (2008), Bowles et al. (2014), and Diviánszky et al. (2023). It does not claim priority for the general dimension-witness framework. The accompanying reproducibility archive contains the exact rational certificate, verification scripts, numerical searches, preregistrations, and SHA-256 manifests.

•Nov 13, 2026•2 citations

When Is a Qutrit Exactly as Good as a Qubit? Three Antipodal Pairs: Reductions, Two Equality Families and a Conditional Classification

Vasily Psykovsky

This preprint investigates when a qutrit offers no operational advantage over a qubit in prepare-and-measure scenarios involving three antipodal pairs of quantum states, fixed binary measurements, and no shared randomness. Previous results establish that two antipodal pairs cannot distinguish qutrits from qubits, whereas four pairs can. The case of three pairs remains open. This fourth note develops structural reductions, explicit equality constructions, and a conditional classification to clarify this boundary. The main contributions are: Finite-query and extremality reductions. We prove that four binary queries suffice to detect any separation involving three antipodal pairs. We show that extreme admissible measurements have at most one eigenvalue strictly inside (−1, 1), and that, for fixed average states, potential separating configurations can be reduced to extreme state pairs. A common-average-state theorem. We prove that when all antipodal pairs share the same average density matrix, every resulting correlation table satisfies the Euclidean factorization bound γ₂ ≤ 1, regardless of the number of pairs or queries. For at most three pairs, this guarantees qubit reproducibility. Two exact equality families. We construct genuinely three-dimensional qutrit configurations that reproduce entire qubit correlation ellipsoids. The first extends spin-1 coherent-state constructions through non-unitary transformations. The second uses state-pair planes sharing a common vector and yields an explicit five-dimensional Euclidean-ball representation for any number of pairs. A conditional structural classification. For three orthogonal pure pairs satisfying an additional algebraic singularity hypothesis, we prove that every configuration belongs, up to unitary equivalence, to an embedded-qubit configuration or one of the two equality families. The proof uses the classification of complex matrix spaces of bounded rank. Reproducible computational investigations. Preregistered numerical studies (B4f and B4g) explore possible three-pair separations, equality configurations, and candidate proof certificates. No separation was found. The studies also identify limitations of fixed-weight Hilbert–Schmidt factorizations and simple Gram-matrix constructions. Open questions. The general conjecture that three antipodal pairs never separate a qutrit from a qubit remains unproved. It is also unknown whether every equality configuration of three orthogonal pure pairs satisfies the algebraic hypothesis required for the classification theorem. Numerical observations are explicitly distinguished from rigorous results. The accompanying reproducibility archive contains symbolic verification scripts, numerical experiments, preregistration records, computational results, and SHA-256 manifests. This is the fourth preprint in a series investigating qubit–qutrit equivalence, separation, and quantum dimension witnesses. Previous preprints: https://doi.org/10.5281/zenodo.24093433 https://doi.org/10.5281/zenodo.24121836 https://doi.org/10.5281/zenodo.24130557 Status: Research preprint; not peer reviewed.

•Nov 12, 2026•2 citations

A Universal Bound for Qutrit Antipodal Tables: κ ≤ 1.059, Reference States, Common Averages and Coupled Certificates

Vasily Psykovsky

Seventh note in the QMEM series on qubit-versus-qutrit antipodal tables. Preprint, version 1; not peer reviewed. An antipodal table records the expectation values of n fixed binary measurements on m pairs of states with opposite values, without shared randomness. Its factorization norm γ₂ is a factorization-based measure of possible qutrit–qubit separation. The condition γ₂ ≤ 1 is necessary for qubit reproducibility and sufficient when the table has at most three rows or at most three columns. For larger tables, γ₂ > 1 certifies a separation, while γ₂ ≤ 1 does not exclude one. The constant κ is the supremum of γ₂ over all qutrit antipodal tables. Main results: Universal bound. Every qutrit antipodal table satisfies γ₂ < 1.058972, improving the previous bound √5/2 ≈ 1.118034. Consequently, 1.007479 ≤ κ < 1.058972. A simpler closed-form bound is γ₂ ≤ 3/(2√2) ≈ 1.060660. Reference states. If all measurements annihilate a state whose largest eigenvalue is at most s (½ ≤ s < 1), then γ₂² ≤ 3/2 − 1/(4s), for arbitrary numbers of pairs and measurements. Every mixture of the pairs' averages is annihilated by those measurements. At s = ½, the bound gives γ₂ ≤ 1, excluding any separation detectable by γ₂; this applies, for example, whenever the table contains an orthogonal pure pair. The underlying pointwise inequality is sharp. Distinguishability and overlap. The pair distinguishability c introduced in Note 6 satisfies c ≤ 1 − tr(ρ⁺ρ⁻) ≤ 4‖σ‖(1 − ‖σ‖) when ‖σ‖ ≥ ½, where σ is the pair's average and ‖σ‖ denotes its largest eigenvalue. Common averages in every dimension. A table whose pairs all share the same average satisfies γ₂ ≤ 1 in any finite dimension, for arbitrary numbers of pairs and measurements. Any excess of γ₂ above one therefore requires pairs with different averages. Coupled certificates. A general factorization bound couples the measurement constraints through the span of all pair averages and encompasses the certificates of Notes 5 and 6. On the certified separating tables of Note 3, a one-parameter family of coupled certificates produces upper bounds within approximately 0.002 of the certified lower bounds. The note is self-contained: Appendix A proves the two results from Note 6 needed for the universal bound. The accompanying reproducibility archive includes proof audits, verification scripts, numerical results, preregistration records, and SHA-256 manifests. The computational checks supplement, but do not replace, the mathematical proofs.

•Nov 13, 2026•2 citations

When Complementary Measurements Count the Same Classical Bit Twice: Counterexamples to CQC, ECQC, and Complementarity-Based Certification

Jinbo Wang, Qihang Wang, Kun Chen

Mutually unbiased measurements are commonly expected to expose independent facets of a quantum state: a correlation that is classical in one basis should disappear in a complementary basis. In higher dimensions, however, this intuition becomes particularly subtle because correlations recovered in different settings need not represent different information. To expose this loophole, we propose a two-branch classical null test: before the setting is chosen, a shared bit selects one of two orthogonal product preparations, producing a rank-two classical--classical state, and the candidate protocol then runs unchanged. Different settings can read the same bit through different outcome patterns. This two-branch classical architecture disproves the complementary-quantum correlation (CQC) conjecture in every dimension $d\geq3$. A distinct rank-two classical--classical state disproves its complete-basis extension (ECQC) at $d=7$, with an overrun that grows without bound along prime dimensions. Its qutrit CQC instance also gives classical false positives for a proposed quantum-correlation measure and a proposed one-sided semi-device-independent steering criterion, and refutes a conditional-probability conjecture. The failures identify the missing requirement: information read in different settings must be nonredundant. In experiments and applications, the same low-overhead architecture can serve as a calibration test before a multibasis score is assigned quantum meaning.

quant-pharXiv:2608.03828•Aug 5, 2026•3 citations
Unrefereed research candidate. Two results on complementary (mutually unbiased) measurements and the complementary quantum correlation (CQC) inequality of Schneeloch, Broadbent and Howell (2014), which proposed that the classical correlations in two complementary measurement tables never exceed the quantum mutual information. 1. Sharp qutrit discrimination. For every pair of complementary bases of a qutrit and every two orthogonal states, the triangular discrimination summed over the two bases is at most 8/3 (the trivial ceiling is 4), answering Qian, arXiv:2608.14806v2, Section 11(4). Every equality case is classified; for the computational and Fourier bases they are tied to the nine Hesse SIC rays. Consequences: choosing one of the two measurements at random recovers at most two thirds of the quantum Fisher information in a two-state estimation task, and Qian's two-ray qubit-qutrit family obeys CQC with margin at least 0.1005 nats. The proof is computer-assisted: an exact rational sum-of-squares certificate (scout_qutrit/exact_sos_certificate.json, 660,609 bytes, SHA-256 825bb8108096d38a44f63f1bc91964405d189ad57b5f568530b5d8e27d598c2f) checked by a standalone exact verifier; this deposit is the certificate's citable locator. 2. Every prescribed pair fails for a qubit and a qudit of dimension n ≥ 4. For every prescribed pair of complementary measurements on each side there is a separable state violating CQC, with a uniform certified margin below −9/20,000,000 nats (previously the every-pair statement was known only for n ≥ 164). A phase-averaging extension of the known shared-label mechanism covers every prescribed pair when both dimensions are at least three; an earlier, unchecked every-pair statement for equal dimensions by E. O'Connor is credited. Open: the full qubit-qutrit CQC question and the sharp constant in dimension 5 (numerics only). The margins are tiny and structural, not experimental. The archive contains the manuscript (PDF and LaTeX), the exact certificate and verifier, negative controls, an exact rational margin checker, numerical cross-checks, working proof notes, novelty audits, the bibliography with registry records, CI and SHA-256 sums. Assurance: computer-assisted exact rational arithmetic, not a formal proof; internal AI-assisted review only (not human peer review or independent reproduction). Creator: Anonymous. Text CC0-1.0; code MIT.
The complementary quantum correlations (CQC) relation bounds the sum of two classical mutual informations, obtained from local mutually unbiased measurements, by the quantum mutual information of the premeasurement state. We refute it. Separable rank-two counterexamples exist in every local dimension pair m×nm\times nm×n with m,n≥3m,n\ge3m,n≥3, with closed-form excess at least 1/(8m2n2)1/(8m^2n^2)1/(8m2n2) nats, and in every qubit--qudit pair 2×n2\times n2×n with n≥3n\ge3n≥3 except n=3,5n=3,5n=3,5; every covered pair also admits full-rank separable counterexamples. We then analyse the two residual qubit--qudit dimensions. A dimension-free entropy envelope replaces the natural quadratic majorant and lowers the requirement for closing the equal-prior orthogonal two-ray family from a triangular-discrimination bound S≤8/3S\le8/3S≤8/3 to S≤3.8265583…S\le3.8265583\ldotsS≤3.8265583…; the associated gate matrix has trace exactly two, so its spectral test collapses to a single eigenvalue-free scalar; and the exact identity X=1−4Var⁡(c)X=1-4\operatorname{Var}(c)X=1−4Var(c) turns the prime-Fourier full-spark barrier into a variance bound. A 128128128-bit interval cover then closes that family at 2×32\times32×3 with gap at least 0.0120210.0120210.012021 nats, and at 2×52\times52×5 an exact saturator attaining S=(14+25)/5S=(14+2\sqrt5)/5S=(14+25​)/5 refutes three competing routes. We also give a state-dependent corrected inequality that is universal, and show it is incomparable with CQC already at 2×22\times22×2. Finally we quantify why the original searches had essentially no power to find these states: the violating set is a sliver against the low-rank boundary, the witness lies 7.37.37.3 standard deviations below the Hilbert--Schmidt mean, the bases must be aligned to about nine degrees (∼10−21\sim10^{-21}∼10−21 of frames), and extrapolating the sample minimum demands 101010^{10}1010 to 101810^{18}1018 samples against the 10710^7107 ever run.
Companion to the preprint “A Qubit–Qutrit Equivalence for Centrally Symmetric Two-Query Quantum Correlations” (doi:10.5281/zenodo.24093433), which proved that for two fixed binary measurements without shared randomness, a qutrit is never better than a qubit at reproducing centrally symmetric target tables. This note settles two questions left open there. Approximately symmetric targets (two queries). If a qutrit fits a table with worst-cell error ε₃, a sharp qubit pair fits it with error at most ε₃ + δ/2, where δ measures the asymmetry of the table. The constant is proved to be best possible. Three queries. The equivalence fails. Every qubit triple of binary measurements satisfies a simple inequality I ≤ 9, which follows from an identity on 3×3 correlation matrices. A triple of projective qutrit measurements reproduces an eight-point centrally symmetric table exactly with I ≥ 9.0505, so every qubit needs worst-cell error larger than 9.7×10⁻⁴ on it. This is a computer-assisted proof based on exact rational certificates, checked by two independent programs. The qutrit advantage is small: about 10⁻³ in worst-cell error. Numerically, the best qutrit value found is I* = 9.0507284088558548…. A candidate minimal polynomial of degree 13 is reported, confirmed to 3000 digits. Its symbolic derivation and the global optimality of I* remain open, and are labelled as numerical findings. Files: the preprint (PDF), and a reproducibility archive (ZIP) with all exact certificates, the original scripts, an independent audit, the full research record including failed and superseded runs, and a SHA-256 manifest. Quick check after unpacking: python3 -I audit/verify_certificates.py . (pure Python, about 1 s). Not peer reviewed. A large language model assisted with derivations, computations, certificate scripts and drafting; the author takes full responsibility for the content.
Eighth note in the QMEM series on qubit-versus-qutrit antipodal tables. Preprint, version 1; not peer reviewed. This note places the results of the series in the standard prepare-and-measure framework for quantum dimension witnesses. A sender prepares one of m pairs of states, a receiver performs one of n binary measurements, and the observed correlators define a half-difference table D = ½(E⁺ − E⁻), evaluated using linear witnesses and the factorization norm γ₂. Main results Qubit bound. Every qubit strategy satisfies γ₂(D) ≤ 1, even with shared randomness and without assuming antipodal correlations. Dimension versus quantumness. Without the antipodal restriction, qutrits attain γ₂ = 2/√3, for example using four mutually unbiased bases in ℂ³, verified in exact arithmetic. Classical trits attain the same value on a 2 × 2 table. Moreover, shared randomness can transform any strategy into one with antipodal correlations while preserving its half-difference table. Thus antipodality alone does not establish a quantum advantage. Classical-trit theorem. For independent preparation and measurement devices, classical trits satisfy γ₂(D) ≤ 1 on exactly antipodal data, whereas qutrits can exceed this bound. A new exact rational certificate establishes γ₂ ≥ 1.007522, improving the previous lower bound of 1.007479. Together with the universal upper bound proved in Note 7, this gives 1.007522 ≤ κ ≤ 3/(2√2) ≈ 1.058972. Consequently, exactly antipodal correlations with γ₂(D) > 1 exclude qubits and classical trits under the stated independence assumption. Blind spots of dimension witnesses. Previous no-advantage theorems identify configurations in which γ₂(D) cannot exceed one: pairs with a common average state, balanced antipodal pairs, traceless measurements, and measurements annihilating a state whose eigenvalues are all at most ½. Noise robustness. For a classical trit with antipodality defect ε, an explicit construction attains γ₂(D)² = (4 − 3ε)/(4(1 − ε)). The conjecture that this is also the universal upper bound is supported by numerical searches but remains unproved. The witness for exactly antipodal data and its exclusion of qubits under depolarising noise are rigorous. The approximate visibility threshold of 0.993 against classical trits is conditional on the conjecture and the specified noise model. Experimental considerations. Idealised statistical estimates suggest approximately 55,000 trials per setting for a 5σ violation using the certified antipodal table, compared with a visibility threshold of 0.866 and a few hundred trials for a dimension-only test based on mutually unbiased bases. These estimates do not constitute a complete finite-statistics protocol. Device independence is an experimental assumption, not a property certified by the witness. The note builds on established research in prepare-and-measure dimension witnesses, including Gallego et al. (2010), Wehner et al. (2008), Bowles et al. (2014), and Diviánszky et al. (2023). It does not claim priority for the general dimension-witness framework. The accompanying reproducibility archive contains the exact rational certificate, verification scripts, numerical searches, preregistrations, and SHA-256 manifests.
This preprint investigates when a qutrit offers no operational advantage over a qubit in prepare-and-measure scenarios involving three antipodal pairs of quantum states, fixed binary measurements, and no shared randomness. Previous results establish that two antipodal pairs cannot distinguish qutrits from qubits, whereas four pairs can. The case of three pairs remains open. This fourth note develops structural reductions, explicit equality constructions, and a conditional classification to clarify this boundary. The main contributions are: Finite-query and extremality reductions. We prove that four binary queries suffice to detect any separation involving three antipodal pairs. We show that extreme admissible measurements have at most one eigenvalue strictly inside (−1, 1), and that, for fixed average states, potential separating configurations can be reduced to extreme state pairs. A common-average-state theorem. We prove that when all antipodal pairs share the same average density matrix, every resulting correlation table satisfies the Euclidean factorization bound γ₂ ≤ 1, regardless of the number of pairs or queries. For at most three pairs, this guarantees qubit reproducibility. Two exact equality families. We construct genuinely three-dimensional qutrit configurations that reproduce entire qubit correlation ellipsoids. The first extends spin-1 coherent-state constructions through non-unitary transformations. The second uses state-pair planes sharing a common vector and yields an explicit five-dimensional Euclidean-ball representation for any number of pairs. A conditional structural classification. For three orthogonal pure pairs satisfying an additional algebraic singularity hypothesis, we prove that every configuration belongs, up to unitary equivalence, to an embedded-qubit configuration or one of the two equality families. The proof uses the classification of complex matrix spaces of bounded rank. Reproducible computational investigations. Preregistered numerical studies (B4f and B4g) explore possible three-pair separations, equality configurations, and candidate proof certificates. No separation was found. The studies also identify limitations of fixed-weight Hilbert–Schmidt factorizations and simple Gram-matrix constructions. Open questions. The general conjecture that three antipodal pairs never separate a qutrit from a qubit remains unproved. It is also unknown whether every equality configuration of three orthogonal pure pairs satisfies the algebraic hypothesis required for the classification theorem. Numerical observations are explicitly distinguished from rigorous results. The accompanying reproducibility archive contains symbolic verification scripts, numerical experiments, preregistration records, computational results, and SHA-256 manifests. This is the fourth preprint in a series investigating qubit–qutrit equivalence, separation, and quantum dimension witnesses. Previous preprints: https://doi.org/10.5281/zenodo.24093433 https://doi.org/10.5281/zenodo.24121836 https://doi.org/10.5281/zenodo.24130557 Status: Research preprint; not peer reviewed.
Seventh note in the QMEM series on qubit-versus-qutrit antipodal tables. Preprint, version 1; not peer reviewed. An antipodal table records the expectation values of n fixed binary measurements on m pairs of states with opposite values, without shared randomness. Its factorization norm γ₂ is a factorization-based measure of possible qutrit–qubit separation. The condition γ₂ ≤ 1 is necessary for qubit reproducibility and sufficient when the table has at most three rows or at most three columns. For larger tables, γ₂ > 1 certifies a separation, while γ₂ ≤ 1 does not exclude one. The constant κ is the supremum of γ₂ over all qutrit antipodal tables. Main results: Universal bound. Every qutrit antipodal table satisfies γ₂ < 1.058972, improving the previous bound √5/2 ≈ 1.118034. Consequently, 1.007479 ≤ κ < 1.058972. A simpler closed-form bound is γ₂ ≤ 3/(2√2) ≈ 1.060660. Reference states. If all measurements annihilate a state whose largest eigenvalue is at most s (½ ≤ s < 1), then γ₂² ≤ 3/2 − 1/(4s), for arbitrary numbers of pairs and measurements. Every mixture of the pairs' averages is annihilated by those measurements. At s = ½, the bound gives γ₂ ≤ 1, excluding any separation detectable by γ₂; this applies, for example, whenever the table contains an orthogonal pure pair. The underlying pointwise inequality is sharp. Distinguishability and overlap. The pair distinguishability c introduced in Note 6 satisfies c ≤ 1 − tr(ρ⁺ρ⁻) ≤ 4‖σ‖(1 − ‖σ‖) when ‖σ‖ ≥ ½, where σ is the pair's average and ‖σ‖ denotes its largest eigenvalue. Common averages in every dimension. A table whose pairs all share the same average satisfies γ₂ ≤ 1 in any finite dimension, for arbitrary numbers of pairs and measurements. Any excess of γ₂ above one therefore requires pairs with different averages. Coupled certificates. A general factorization bound couples the measurement constraints through the span of all pair averages and encompasses the certificates of Notes 5 and 6. On the certified separating tables of Note 3, a one-parameter family of coupled certificates produces upper bounds within approximately 0.002 of the certified lower bounds. The note is self-contained: Appendix A proves the two results from Note 6 needed for the universal bound. The accompanying reproducibility archive includes proof audits, verification scripts, numerical results, preregistration records, and SHA-256 manifests. The computational checks supplement, but do not replace, the mathematical proofs.
Mutually unbiased measurements are commonly expected to expose independent facets of a quantum state: a correlation that is classical in one basis should disappear in a complementary basis. In higher dimensions, however, this intuition becomes particularly subtle because correlations recovered in different settings need not represent different information. To expose this loophole, we propose a two-branch classical null test: before the setting is chosen, a shared bit selects one of two orthogonal product preparations, producing a rank-two classical--classical state, and the candidate protocol then runs unchanged. Different settings can read the same bit through different outcome patterns. This two-branch classical architecture disproves the complementary-quantum correlation (CQC) conjecture in every dimension d≥3d\geq3d≥3. A distinct rank-two classical--classical state disproves its complete-basis extension (ECQC) at d=7d=7d=7, with an overrun that grows without bound along prime dimensions. Its qutrit CQC instance also gives classical false positives for a proposed quantum-correlation measure and a proposed one-sided semi-device-independent steering criterion, and refutes a conditional-probability conjecture. The failures identify the missing requirement: information read in different settings must be nonredundant. In experiments and applications, the same low-overhead architecture can serve as a calibration test before a multibasis score is assigned quantum meaning.