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

Data and code archive for: Spectral structure of noise thresholds for contextuality: Walsh modes, a heralded-CHSH invariant, and model dependence

Igor Postanovskyi

Published Oct 13, 2026• 0 citations
Publisher

Abstract

Certificates, checkers and data accompanying the preprint "Spectral structure of noise thresholds for contextuality: Walsh modes, a heralded-CHSH invariant, and model dependence" (I. Postanovsky, 2026): cone covers of RP^3 with per-leaf moment-LP certificates for the bit-flip and joint-reset noise models, the ball-arithmetic certificate of the expansion of the CH-branch constant kappa*_n, exact certificates over Q(sqrt5) for the counterexamples, and the data of all figures and tables. MANIFEST.txt gives the SHA-256 of every file.

DOI
10.5281/zenodo.23348742

Related papers

Spectral Structure of Noise Thresholds for Contextuality: Walsh Modes, A Heralded-CHSH Invariant, And Model Dependence

Igor Postanovskyi

•Jan 1, 2026•0 citations

Noise Thresholds Of Products Of Odd-Cycle Contextuality Scenarios: A CHSH Window, An Exact Face Threshold For KCBS × KCBS, And Two Regimes

Igor Postanovskyi

•Jan 1, 2026•0 citations

How Far Can a Qutrit Beat a Qubit on Antipodal Tables? An Exact Qubit Test, Bounds, and a Conjectured Constant for Three Queries

Vasily Psykovsky

Third note in a series on qubit–qutrit equivalence for centrally symmetric prepare-and-measure correlations (doi:10.5281/zenodo.24093433, doi:10.5281/zenodo.24121836). Setting: fixed binary measurements, no shared randomness, and tables of antipodal pairs of target expectation vectors. The question: by how much must a qutrit table be shrunk before a qubit can reproduce it? Proved: A qubit reproduces such a table exactly if and only if it factors through 3-dimensional real space with all rows and columns in the unit ball. With at most three queries this is the condition γ₂(T) ≤ 1 on the factorization norm, a single semidefinite program. Every qutrit table has γ₂(T) ≤ 2/√3, while the same argument gives exactly 1 for qubits. The worst-case factor κ(n) never decreases with the number of queries n. For three queries, γ₂(T)² ≥ I(T)/9, where I ≤ 9 is the qubit inequality of the previous note. Averaging a core over relabellings of the measurements can never raise the shrink factor. Certified in exact rational arithmetic (computer-assisted): κ(3) ≥ 1.002813276 and κ(4) ≥ 1.007479. For a fixed rational reference triple A*, close to the numerically best symmetric triple, the shrink factor lies in [1.002813276, 1.00281429]. The upper bound uses a new second-order covering of the direction sphere. Conjectured (not proved): κ(3) = √(I*)/3 = 1.0028142848…, and the worst triple is cyclically symmetric. No numerical search found a larger value. The note explains why the obvious proof strategies fail. Files: the preprint (PDF), and a reproducibility archive (ZIP) with five preregistered research stages, including failed and stopped runs, all certificates, and a SHA-256 manifest. After unpacking, two pure-Python exact checks reproduce the certificates in about 20 s, without any numerical solver. They check the certificates, not the surrounding proofs. 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

Exact Noise Thresholds of the Minimal State-independent Contextuality Set: Critical State, Full Noise Profile, and a Machine-checked Gap between Contextuality and Incompatibility

Igor Postanovskyi

•Jan 1, 2026•1 citation

Twenty-Three Certified Decimals: Rigorous Ground-State Energies, Correlations and a Coupling Curve for the Half-Filled 4×4 Hubbard Model: certificate bundle v1.0.0

Bryan W. Daugherty, Shawn Ryan, Gregory Ward

We prove the ground-state energy of the periodic, half-filled 4×4 repulsive Hubbard model with t = 1 to at least 23 decimal places at U = 1, 2, 3, 4, 5, 6, 7, 8, 10, 12 and 16: every enclosure has width below 10⁻²⁴. The double occupancy, the spin correlations and the staggered structure factor have certified interval widths below 10⁻¹⁰, and the kinetic energy below 10⁻⁹. Hypercube, spin and pseudospin symmetries reduce the problem to a 19,186-dimensional block that contains the ground state by Lieb's theorem and a character argument. A binary64 Cholesky factorization, verified using a rigorous rounding-error bound evaluated in exact rational arithmetic, certifies spectral separation within this block; exact Rayleigh quotients and residuals then give Temple enclosures of the energy and bounds on the angle to the ground state, which certify the observables. Separation certificates at both ends and a perturbative trial family enclose E₀(U) for every U in [3.9, 4.1], with uniform width at most 1.6 × 10⁻⁹. Two independent implementations replay the certificates, the interlacing and Temple steps are proved in Lean 4, and all computations run on one laptop. At U = 4 and 8, exact-arithmetic brackets that use neither Lieb's theorem nor floating point are also given, and the enclosures bound their deficits. At U = 4 the certified energy rounds to −13.62185 and excludes the published value −13.62192 [Physical Review B 94, 085103]. Contents of the bundle paper/ — the manuscript (Markdown and PDF), its figures, and supplementary/ with five videos generated from the certificate files. rayleigh/symmetry/ — the enclosure pipeline (block dimensions, exact basis, certificates, observables, the interval certificate) and recheck/, the independent Rust reference checker; certificates/ holds every certificate, trial vector, run log, checker log and the clean-replay report. sos/, rayleigh/demo/, cluster/ — the exact-arithmetic bracket certificates and the expected outputs of their checkers (golden_ files); the bracket checkers belong to the authors' certification software and are not included. formal/ — the Lean 4 statements (rank-one interlacing, Temple's inequality, the sum-of-squares remainder bound). tests/ — the generators of every table, figure and video of the manuscript. replay.sh, REPLAY.md — the automated replay and the step-by-step commands. AUDIT_AXIOMS.txt — #print axioms of the formal statements. SHA256SUMS — checksums of every file. Verification PYTHON=<python with numpy and scipy> ./replay.sh rebuilds the symmetry block from the Hamiltonian, reproduces every certificate from the stored integer vectors, runs the Rust reference checker, recomputes the observables and checks the manuscript's tables, and writes REPLAY_REPORT.md. It needs about 12 GB of memory and 16 GB of free disk, and takes about half an hour on the hardware above; its work directory is removed at the end. REPLAY.md lists the commands step by step; rayleigh/symmetry/certificates/REPLAY_REPORT_2026-10-14.md is a clean replay of this material. Not included: the two 662,547,600-byte full-sector vectors of the upper bracket certificates (their SHA-256 digests are in rayleigh/demo/). Integrity: shasum -a 256 -c SHA256SUMS inside the bundle; the SHA-256 of the tarball is in the .sha256 file beside it. Environment Python: 3.13.7 with numpy 2.5.4, scipy 1.18.1, matplotlib 3.11.2 (producer-requirements.txt) Rust: 1.90.0 (reference checker; cargo build --release in rayleigh/symmetry/recheck) Lean: leanprover/lean4:v4.33.1 with Mathlib v4.33.1 (formal statements) Hardware: Apple M3 Pro, 18 GiB (all production and replay) License: CC BY-NC 4.0. Commercial use requires written permission from Origin Neural AI (labs@smartledger.solutions).

•Oct 14, 2026•1 citation

Classical and Quantum Antipodal Tables in Every Dimension: An Envelope Theorem, the Parity of the Quantum Advantage, and a Rigorous Noise Threshold

Vasily Psykovsky

Ninth note in the QMEM series on qubit-versus-qutrit antipodal tables. Preprint, version 1; not peer reviewed. This note compares the factorization norms of classical and quantum prepare-and-measure correlations in arbitrary finite dimension. It establishes an envelope theorem, determines the optimal classical constant for antipodal tables in every dimension, identifies an even–odd distinction in the achievable quantum advantage, and provides a computer-assisted noise certificate for the qutrit dimension witness introduced in Note 8. Main results Envelope theorem. For arbitrary pairs of quantum states with average states σₖ and arbitrary binary measurements, the half-difference table D satisfies γ₂(D)² ≤ tr(ω) for every positive semidefinite operator ω dominating all σₖ. The result does not require antipodality. Minimizing tr(ω) gives a semidefinite-programming bound, unifying the common-average theorem and several earlier no-advantage results. If all averages lie in the convex hull of r states, then γ₂(D)² ≤ r. Exact classical constant. Every classical d-level antipodal table satisfies γ₂(D)² ≤ min(v, d − v), where v is the dimension of the span of the measurement response vectors. The bound implies the exact optimal classical constant κ_cl(d) = √⌊d/2⌋, attained in every dimension. The proof combines a rank bound derived from John's ellipsoid theorem, the envelope theorem, and a trace–rank inequality for nonnegative column-stochastic matrices. Parity of the quantum advantage. The optimal quantum and classical constants coincide in every even dimension, where both equal √(d/2). In every odd dimension d = 2r + 1, the quantum squared constant exceeds the classical value r by at least 0.0151. Thus the quantum advantage in this factorization-norm criterion exists precisely in odd dimensions. Dimension witnesses in every dimension. Under the assumption of independent preparation and measurement devices, exactly antipodal data satisfying γ₂(D)² > r exclude quantum systems of dimension at most 2r and classical systems with at most 2r + 1 levels. Such correlations are achievable by quantum systems of dimension 2r + 1. Rigorous noise threshold. For the certified 4 × 8 qutrit witness of Note 8, a branch-and-bound computation with exact-rational verification of every leaf establishes that classical trits with antipodality defect at most 1/100 cannot reach the qubit witness bound. Under the specified depolarising-noise model, the threshold for excluding classical trits therefore equals the rigorous qubit threshold: visibility greater than 0.9925339. This removes the conjectural dependence of the earlier threshold for this particular witness; the general conjecture remains open. Scope and limitations The comparison concerns the factorization norm of antipodal prepare-and-measure data, not a universal advantage of quantum systems for every communication task. Independence of the devices is assumed rather than certified. The noise theorem applies to the specified witness, while the general classical-trit noise conjecture remains unresolved. Reproducibility The accompanying supplementary archive documents the analytical checks, preregistered numerical tests, branch-and-bound trees, exact-rational verification procedures, execution logs, and SHA-256 manifests. The manuscript provides commands for reproducing the computer-assisted theorem. This work builds on established dimension-witness and factorization-norm literature and on the preceding eight notes in the QMEM series.

•Nov 14, 2026•0 citations
Certificates, checkers and data accompanying the preprint "Spectral structure of noise thresholds for contextuality: Walsh modes, a heralded-CHSH invariant, and model dependence" (I. Postanovsky, 2026): cone covers of RP^3 with per-leaf moment-LP certificates for the bit-flip and joint-reset noise models, the ball-arithmetic certificate of the expansion of the CH-branch constant kappa*_n, exact certificates over Q(sqrt5) for the counterexamples, and the data of all figures and tables. MANIFEST.txt gives the SHA-256 of every file.
Third note in a series on qubit–qutrit equivalence for centrally symmetric prepare-and-measure correlations (doi:10.5281/zenodo.24093433, doi:10.5281/zenodo.24121836). Setting: fixed binary measurements, no shared randomness, and tables of antipodal pairs of target expectation vectors. The question: by how much must a qutrit table be shrunk before a qubit can reproduce it? Proved: A qubit reproduces such a table exactly if and only if it factors through 3-dimensional real space with all rows and columns in the unit ball. With at most three queries this is the condition γ₂(T) ≤ 1 on the factorization norm, a single semidefinite program. Every qutrit table has γ₂(T) ≤ 2/√3, while the same argument gives exactly 1 for qubits. The worst-case factor κ(n) never decreases with the number of queries n. For three queries, γ₂(T)² ≥ I(T)/9, where I ≤ 9 is the qubit inequality of the previous note. Averaging a core over relabellings of the measurements can never raise the shrink factor. Certified in exact rational arithmetic (computer-assisted): κ(3) ≥ 1.002813276 and κ(4) ≥ 1.007479. For a fixed rational reference triple A*, close to the numerically best symmetric triple, the shrink factor lies in [1.002813276, 1.00281429]. The upper bound uses a new second-order covering of the direction sphere. Conjectured (not proved): κ(3) = √(I*)/3 = 1.0028142848…, and the worst triple is cyclically symmetric. No numerical search found a larger value. The note explains why the obvious proof strategies fail. Files: the preprint (PDF), and a reproducibility archive (ZIP) with five preregistered research stages, including failed and stopped runs, all certificates, and a SHA-256 manifest. After unpacking, two pure-Python exact checks reproduce the certificates in about 20 s, without any numerical solver. They check the certificates, not the surrounding proofs. Not peer reviewed. A large language model assisted with derivations, computations, certificate scripts and drafting; the author takes full responsibility for the content.
We prove the ground-state energy of the periodic, half-filled 4×4 repulsive Hubbard model with t = 1 to at least 23 decimal places at U = 1, 2, 3, 4, 5, 6, 7, 8, 10, 12 and 16: every enclosure has width below 10⁻²⁴. The double occupancy, the spin correlations and the staggered structure factor have certified interval widths below 10⁻¹⁰, and the kinetic energy below 10⁻⁹. Hypercube, spin and pseudospin symmetries reduce the problem to a 19,186-dimensional block that contains the ground state by Lieb's theorem and a character argument. A binary64 Cholesky factorization, verified using a rigorous rounding-error bound evaluated in exact rational arithmetic, certifies spectral separation within this block; exact Rayleigh quotients and residuals then give Temple enclosures of the energy and bounds on the angle to the ground state, which certify the observables. Separation certificates at both ends and a perturbative trial family enclose E₀(U) for every U in [3.9, 4.1], with uniform width at most 1.6 × 10⁻⁹. Two independent implementations replay the certificates, the interlacing and Temple steps are proved in Lean 4, and all computations run on one laptop. At U = 4 and 8, exact-arithmetic brackets that use neither Lieb's theorem nor floating point are also given, and the enclosures bound their deficits. At U = 4 the certified energy rounds to −13.62185 and excludes the published value −13.62192 [Physical Review B 94, 085103]. Contents of the bundle paper/ — the manuscript (Markdown and PDF), its figures, and supplementary/ with five videos generated from the certificate files. rayleigh/symmetry/ — the enclosure pipeline (block dimensions, exact basis, certificates, observables, the interval certificate) and recheck/, the independent Rust reference checker; certificates/ holds every certificate, trial vector, run log, checker log and the clean-replay report. sos/, rayleigh/demo/, cluster/ — the exact-arithmetic bracket certificates and the expected outputs of their checkers (golden_ files); the bracket checkers belong to the authors' certification software and are not included. formal/ — the Lean 4 statements (rank-one interlacing, Temple's inequality, the sum-of-squares remainder bound). tests/ — the generators of every table, figure and video of the manuscript. replay.sh, REPLAY.md — the automated replay and the step-by-step commands. AUDIT_AXIOMS.txt — #print axioms of the formal statements. SHA256SUMS — checksums of every file. Verification PYTHON=<python with numpy and scipy> ./replay.sh rebuilds the symmetry block from the Hamiltonian, reproduces every certificate from the stored integer vectors, runs the Rust reference checker, recomputes the observables and checks the manuscript's tables, and writes REPLAY_REPORT.md. It needs about 12 GB of memory and 16 GB of free disk, and takes about half an hour on the hardware above; its work directory is removed at the end. REPLAY.md lists the commands step by step; rayleigh/symmetry/certificates/REPLAY_REPORT_2026-10-14.md is a clean replay of this material. Not included: the two 662,547,600-byte full-sector vectors of the upper bracket certificates (their SHA-256 digests are in rayleigh/demo/). Integrity: shasum -a 256 -c SHA256SUMS inside the bundle; the SHA-256 of the tarball is in the .sha256 file beside it. Environment Python: 3.13.7 with numpy 2.5.4, scipy 1.18.1, matplotlib 3.11.2 (producer-requirements.txt) Rust: 1.90.0 (reference checker; cargo build --release in rayleigh/symmetry/recheck) Lean: leanprover/lean4:v4.33.1 with Mathlib v4.33.1 (formal statements) Hardware: Apple M3 Pro, 18 GiB (all production and replay) License: CC BY-NC 4.0. Commercial use requires written permission from Origin Neural AI (labs@smartledger.solutions).
Ninth note in the QMEM series on qubit-versus-qutrit antipodal tables. Preprint, version 1; not peer reviewed. This note compares the factorization norms of classical and quantum prepare-and-measure correlations in arbitrary finite dimension. It establishes an envelope theorem, determines the optimal classical constant for antipodal tables in every dimension, identifies an even–odd distinction in the achievable quantum advantage, and provides a computer-assisted noise certificate for the qutrit dimension witness introduced in Note 8. Main results Envelope theorem. For arbitrary pairs of quantum states with average states σₖ and arbitrary binary measurements, the half-difference table D satisfies γ₂(D)² ≤ tr(ω) for every positive semidefinite operator ω dominating all σₖ. The result does not require antipodality. Minimizing tr(ω) gives a semidefinite-programming bound, unifying the common-average theorem and several earlier no-advantage results. If all averages lie in the convex hull of r states, then γ₂(D)² ≤ r. Exact classical constant. Every classical d-level antipodal table satisfies γ₂(D)² ≤ min(v, d − v), where v is the dimension of the span of the measurement response vectors. The bound implies the exact optimal classical constant κ_cl(d) = √⌊d/2⌋, attained in every dimension. The proof combines a rank bound derived from John's ellipsoid theorem, the envelope theorem, and a trace–rank inequality for nonnegative column-stochastic matrices. Parity of the quantum advantage. The optimal quantum and classical constants coincide in every even dimension, where both equal √(d/2). In every odd dimension d = 2r + 1, the quantum squared constant exceeds the classical value r by at least 0.0151. Thus the quantum advantage in this factorization-norm criterion exists precisely in odd dimensions. Dimension witnesses in every dimension. Under the assumption of independent preparation and measurement devices, exactly antipodal data satisfying γ₂(D)² > r exclude quantum systems of dimension at most 2r and classical systems with at most 2r + 1 levels. Such correlations are achievable by quantum systems of dimension 2r + 1. Rigorous noise threshold. For the certified 4 × 8 qutrit witness of Note 8, a branch-and-bound computation with exact-rational verification of every leaf establishes that classical trits with antipodality defect at most 1/100 cannot reach the qubit witness bound. Under the specified depolarising-noise model, the threshold for excluding classical trits therefore equals the rigorous qubit threshold: visibility greater than 0.9925339. This removes the conjectural dependence of the earlier threshold for this particular witness; the general conjecture remains open. Scope and limitations The comparison concerns the factorization norm of antipodal prepare-and-measure data, not a universal advantage of quantum systems for every communication task. Independence of the devices is assumed rather than certified. The noise theorem applies to the specified witness, while the general classical-trit noise conjecture remains unresolved. Reproducibility The accompanying supplementary archive documents the analytical checks, preregistered numerical tests, branch-and-bound trees, exact-rational verification procedures, execution logs, and SHA-256 manifests. The manuscript provides commands for reproducing the computer-assisted theorem. This work builds on established dimension-witness and factorization-norm literature and on the preceding eight notes in the QMEM series.