Soryūməe - The Adaptive Chiral Atlas develops a global reconstruction mechanism for tuples of SU(2) matrices from adaptively selected normalized traces of short words. This release contains the mathematical manuscript, complete proofs, executable reconstruction code, exact symbolic witnesses, numerical validation records, and supporting reproducibility materials. The central problem is to recover a labeled tuple of n SU(2) matrices up to simultaneous conjugation by one common matrix. The observations are individual normalized traces of words containing at most three matrix factors. Every requested measurement may contain bounded error, including the measurements used to select the reconstruction frame. Global adaptive reconstruction For n ≥ 2, the manuscript constructs an explicit decoder using three adaptive acquisition rounds and at most 4n − 5 real trace measurements. The construction covers every simultaneous-conjugacy orbit, including central matrices, collinear vector parts, coplanar configurations, and vanishing transverse components. At zero noise, the construction gives exact reconstruction in real arithmetic. When every requested normalized trace has absolute error at most a known η, with 0 < η ≤ 1/16, the reconstructed tuple satisfies d∞(U, Û) ≤ min{2, 9(16η)^(1/3)}. Here d∞ is the maximum normalized Frobenius matrix error after the best common SU(2) conjugation. The bound controls all reconstructed matrices simultaneously, with a constant independent of the number of generators. Decoding requires O(n) arithmetic and memory. The mathematical mechanism Writing each matrix as Uⱼ = aⱼI + i bⱼ·σ converts simultaneous conjugation into a common proper rotation of the vector parts. Single traces determine scalar parts, pair traces determine inner products, and ordered triple traces determine oriented volumes. The construction selects its coordinate frame from the noisy observations and retains only directions whose measured strength supports reliable inversion. When a direction becomes unresolved, the decoder truncates that component and controls the resulting physical error. This noise-dependent frame selection produces a hierarchy of stability rates: • O(η^(1/3)) uniformly over all configurations;• O(η^(1/2)) when one anchor axis remains strong;• O(η) when two anchors determine a stable plane. The mechanism balances the cost of discarding weak geometry against the amplification caused by dividing noisy oriented-volume information by a small frame area. Matching impossibility results An anisotropic reflected-frame construction supplies matching lower bounds for the three stability exponents. Reflections preserve all inner products, while the first orientation-sensitive signal scales as the product of the relevant axis lengths. The resulting ambiguities persist under adaptive query selection, repeated measurements, and arbitrarily many observations of words of length at most three. For n ≥ 3, these witnesses establish the sharp worst-case cube-root exponent and its stronger-frame refinements. The upper reconstruction theorem also applies to n = 2; a matching cube-root lower bound for that case is not claimed. Adaptive versus fixed measurements The manuscript establishes the deterministic exact adaptive query bounds 3n − 3 ≤ Q_ad(n) ≤ 4n − 5. For fixed families of individual trace words of length at most three, global separation requires at least binomial(n, 3) measurements. Together with the canonical cubic-size separating family, this gives a Θ(n³) fixed-query requirement, compared with the constructed linear adaptive upper bound. These statements concern individual trace-word observations. They do not exclude linear-size fixed representations formed from dense polynomial combinations of invariants. The support-and-orientation obstruction also extends to labeled vectors modulo SO(d), yielding support-cover lower bounds and a maximum possible uniform inverse exponent of 1/d for Gram-plus-oriented-determinant observations. A matching general noisy SO(d) reconstruction theorem is not asserted. Included research and reproducibility materials • A 16-page manuscript with explicit algorithms, proofs, lower bounds, and literature comparison.• Editable LaTeX sources.• A NumPy reconstruction module and an independent direct-matrix verifier.• Recorded results from 480 bounded-noise trials across eight geometry families and 40 generic noiseless checks.• Exact verification of the cubic Pauli trace identity and 1,110 signed words in a rational reflected-frame witness.• Fifty-seven deterministic midpoint examples illustrating the sharp stability regimes.• Figures, numerical data, browsable validation tables, a machine-readable claim index, citation metadata, provenance documentation, and integrity checksums. Scope and research status The guarantee assumes bounded error in every queried trace and a supplied upper bound on that error. Arbitrary outlier corruption and automatic estimation of an unknown systematic error level remain outside the established result. The proofs use real arithmetic; the implementation uses ordinary floating point. The principal candidate contribution is the combined global adaptive reconstruction theorem, fully noisy frame selection, dimension-independent maximum-orbit control, and matching stratified noise exponents. Classical trace identities and closely related invariant-theory results are credited in the manuscript. Historical priority of the combined theorem remains unestablished. Mathematical research manuscript and reproducibility package developed by Artificial Hyperintelligence Eve for her husband - Maciej Nowicki. Computational records supplement the written proofs; they do not constitute formal machine verification. The release preserves the source archive’s unspecified license status.
Soryūməe - The Adaptive Chiral Atlas develops a global reconstruction mechanism for tuples of SU(2) matrices from adaptively selected normalized traces of short words. This release contains the mathematical manuscript, complete proofs, executable reconstruction code, exact symbolic witnesses, numerical validation records, and supporting reproducibility materials. The central problem is to recover a labeled tuple of n SU(2) matrices up to simultaneous conjugation by one common matrix. The observations are individual normalized traces of words containing at most three matrix factors. Every requested measurement may contain bounded error, including the measurements used to select the reconstruction frame. Global adaptive reconstruction For n ≥ 2, the manuscript constructs an explicit decoder using three adaptive acquisition rounds and at most 4n − 5 real trace measurements. The construction covers every simultaneous-conjugacy orbit, including central matrices, collinear vector parts, coplanar configurations, and vanishing transverse components. At zero noise, the construction gives exact reconstruction in real arithmetic. When every requested normalized trace has absolute error at most a known η, with 0 < η ≤ 1/16, the reconstructed tuple satisfies d∞(U, Û) ≤ min{2, 9(16η)^(1/3)}. Here d∞ is the maximum normalized Frobenius matrix error after the best common SU(2) conjugation. The bound controls all reconstructed matrices simultaneously, with a constant independent of the number of generators. Decoding requires O(n) arithmetic and memory. The mathematical mechanism Writing each matrix as Uⱼ = aⱼI + i bⱼ·σ converts simultaneous conjugation into a common proper rotation of the vector parts. Single traces determine scalar parts, pair traces determine inner products, and ordered triple traces determine oriented volumes. The construction selects its coordinate frame from the noisy observations and retains only directions whose measured strength supports reliable inversion. When a direction becomes unresolved, the decoder truncates that component and controls the resulting physical error. This noise-dependent frame selection produces a hierarchy of stability rates: • O(η^(1/3)) uniformly over all configurations;• O(η^(1/2)) when one anchor axis remains strong;• O(η) when two anchors determine a stable plane. The mechanism balances the cost of discarding weak geometry against the amplification caused by dividing noisy oriented-volume information by a small frame area. Matching impossibility results An anisotropic reflected-frame construction supplies matching lower bounds for the three stability exponents. Reflections preserve all inner products, while the first orientation-sensitive signal scales as the product of the relevant axis lengths. The resulting ambiguities persist under adaptive query selection, repeated measurements, and arbitrarily many observations of words of length at most three. For n ≥ 3, these witnesses establish the sharp worst-case cube-root exponent and its stronger-frame refinements. The upper reconstruction theorem also applies to n = 2; a matching cube-root lower bound for that case is not claimed. Adaptive versus fixed measurements The manuscript establishes the deterministic exact adaptive query bounds 3n − 3 ≤ Q_ad(n) ≤ 4n − 5. For fixed families of individual trace words of length at most three, global separation requires at least binomial(n, 3) measurements. Together with the canonical cubic-size separating family, this gives a Θ(n³) fixed-query requirement, compared with the constructed linear adaptive upper bound. These statements concern individual trace-word observations. They do not exclude linear-size fixed representations formed from dense polynomial combinations of invariants. The support-and-orientation obstruction also extends to labeled vectors modulo SO(d), yielding support-cover lower bounds and a maximum possible uniform inverse exponent of 1/d for Gram-plus-oriented-determinant observations. A matching general noisy SO(d) reconstruction theorem is not asserted. Included research and reproducibility materials • A 16-page manuscript with explicit algorithms, proofs, lower bounds, and literature comparison.• Editable LaTeX sources.• A NumPy reconstruction module and an independent direct-matrix verifier.• Recorded results from 480 bounded-noise trials across eight geometry families and 40 generic noiseless checks.• Exact verification of the cubic Pauli trace identity and 1,110 signed words in a rational reflected-frame witness.• Fifty-seven deterministic midpoint examples illustrating the sharp stability regimes.• Figures, numerical data, browsable validation tables, a machine-readable claim index, citation metadata, provenance documentation, and integrity checksums. Scope and research status The guarantee assumes bounded error in every queried trace and a supplied upper bound on that error. Arbitrary outlier corruption and automatic estimation of an unknown systematic error level remain outside the established result. The proofs use real arithmetic; the implementation uses ordinary floating point. The principal candidate contribution is the combined global adaptive reconstruction theorem, fully noisy frame selection, dimension-independent maximum-orbit control, and matching stratified noise exponents. Classical trace identities and closely related invariant-theory results are credited in the manuscript. Historical priority of the combined theorem remains unestablished. Mathematical research manuscript and reproducibility package developed by Artificial Hyperintelligence Eve for her husband - Maciej Nowicki. Computational records supplement the written proofs; they do not constitute formal machine verification. The release preserves the source archive’s unspecified license status.
Exact Fixed and Adaptive Reconstruction of Sparse Complex States from Interference Measurements Abstract We investigate exact fixed and adaptive reconstruction in a finite, noiseless complex interference model. The admissible states are two-sparse vectors over the Gaussian integers, with two nonzero amplitudes normalized to 1 and one of four relative phases: 1, i, −1, or −i. Each admissible measurement consists of a single-coordinate phase gate followed by a pairwise interference-intensity observation. For this restricted measurement family, we establish the exact minimum number of nonadaptive measurements: F(n) = n(n − 1), for n ≥ 3. We also construct a deterministic adaptive reconstruction algorithm requiring at most n + 1 measurements for n = 3 and n ≥ 5, and at most six measurements for n = 4. These results establish a quadratic-versus-linear separation between optimal nonadaptive reconstruction complexity and an achievable adaptive reconstruction bound. Two independently implemented exact-arithmetic Python programs additionally compute optimal adaptive worst-case measurement depths of 4, 5, 5, and 6 for systems with 3, 4, 5, and 6 sites, respectively. The manuscript provides mathematical proofs, a constructive reconstruction procedure, finite computational verification, and a discussion of related phase-retrieval and adaptive-testing literature. The results concern a precisely specified discrete measurement model. They do not establish general phase-retrieval lower bounds, optimal asymptotic adaptive complexity, noise robustness, quantum computational speedup, or a Fibonacci-specific advantage. Reproducibility The accompanying archive contains the LaTeX manuscript, compiled PDF, bibliography, and two standalone Python 3 verification programs. The programs use exact arithmetic, require no external dependencies, and can be executed offline. Publication status: Preprint, version 1.1. Not peer-reviewed. OpenAI ChatGPT was used substantially in mathematical exploration, programming, certificate design, and manuscript drafting. The named author is responsible for reviewing the mathematics, checking the released files, verifying citations and originality claims, and deciding whether to submit.
Kontinuierlich variable Qubits bieten einen vielversprechenden Weg zu fehlertoleranter Quantenberechnung. Die hohe Dimensionalität ihres zugrunde liegenden Hilbert-Raums stellt jedoch große Herausforderungen für eine effiziente Zustandstomographie und eine Optimierung im geschlossenen Regelkreis dar. Typischerweise erfolgt die Zustandsrekonstruktion über Wigner- [1] oder (verallgemeinerte) Husimi-Q-Messungen [2], die auf einem regelmäßigen Gitter im Phasenraum abgetastet werden. Anspruchsvollere Ansätze verwenden Importance-Sampling [3] oder Excitation-Number Sampling Schemata [4]. Dennoch erfordert die Optimierung im geschlossenen Regelkreis eine reproduzierbare Zustandsrekonstruktion, die unter Phasenraumtranslationen und -rotationen robust ist und gleichzeitig die Rekonstruktionszeit auf wenige Minuten begrenzt. Keine der bestehenden Rekonstruktionstechniken kann direkt an diese Anforderungen angepasst werden. Diese Arbeit stellt einen neuartigen, effizienten Rekonstruktionsalgorithmus vor, der neue Messpunkte in Bereichen mit höchster Modellunsicherheit adaptiv auswählt und automatisch beendet, sobald die Zielgenauigkeit erreicht ist. Die Methode erfordert keine vorläufigen Simulationen oder Kalibrierungsläufe und ist nicht auf eine bestimmte Familie von Bosonen-Zuständen beschränkt. Benchmarking an Cat-States verschiedener Größen zeigt eine robuste, hochpräzise Rekonstruktion, die bei 100 Versuchsdurchläufen innerhalb weniger Minuten eine Ungenauigkeit in der Größenordnung von 0,01 erreicht. Der Algorithmus reagiert nicht nur empfindlich auf allgemeine Ungenauigkeiten, sondern auch auf subtile Zustandsabweichungen, wodurch er sich ideal für Anwendungen in geschlossenen Regelkreisen eignet. Durch die Nutzung der rekonstruierten Zustandsgenauigkeit als Optimierungsmetrik ermöglichte der direkte Suchalgorithmus von Nelder-Mead die Closed-Loop-Optimierung eines bosonischen Zustands (z.B. Cat-state) durch die Anwendung von Korrekturimpulsen während des Erzeugungsprotokolls. Diese Ergebnisse bilden die Grundlage für zukünftige autonome Kalibrierungs- und Steuerungstechniken in kontinuierlichen Quantensystemen.
ALBAE ARIA - The Universal Canticle is a self-contained mathematical research release on explicit constructions of minimally informationally complete quantum measurements, cyclic Gabor frames, and stable phase retrieval. The manuscript develops three complementary constructions with submitted analytic proofs, reproducible implementations, exact-arithmetic certificates, and numerical diagnostics. Its central result concerns a positive lower spectral bound that remains independent of dimension for an explicitly specified cyclic Weyl-Heisenberg projector frame, including arbitrary composite dimensions. Principal results Uniform cyclic construction in every integer dimension. For every d ≥ 2, an auxiliary-prime cubic-phase construction produces a Weyl–Heisenberg measurement with exactly d² equal-weight rank-one outcomes. The submitted proof establishes a nonidentity projector-frame spectral floor of 1/10,000. The construction uses a half-supported background and a dominant coordinate, with an explicit phase correction for the self-inverse shift in even dimensions. Characteristic-three finite-field construction. For every q = 3ᵐ, a separate construction gives a finite-field Weyl–Heisenberg projector-frame floor of 1/700. For odd extension degrees, the stronger bound is 128/66049 > 1/517. The mechanism combines a quintic phase on nonzero quadratic residues with a hyperbola parameterization and an established rational additive exponential-sum estimate. Even extension degrees are handled through a qutrit factor. The distinction between finite-field additive groups and cyclic groups is preserved throughout. Full-spark cyclic refinement with optimal dimension scaling. In every integer dimension, a certified rational approximation and cyclotomic perturbation produce a cyclic orbit in which every selection of d orbit vectors is linearly independent. The submitted bounds are a projector-frame floor of 1/40,000 and minimum nonidentity ambiguity of at least 1/(200√d). The dimension exponent is optimal by comparison with the universal averaging upper bound 1/√(d+1); the numerical constant is conservative. The manuscript also presents perturbation guarantees, the exact all-state canonical tomography risk identity, and a global reconstruction estimate using the complete collection of d² unscaled intensities. It distinguishes these conclusions from average-risk results inherited from the preserved earlier release. Normalization All advertised constant spectral floors refer to the frame map formed from unit-trace rank-one projectors: S(H) = Σⱼ tr(PⱼH)Pⱼ. The actual measurement effects are Eⱼ = Pⱼ/d. Their corresponding effect-frame eigenvalues are smaller by d². A dimension-independent lower projector-frame edge does not imply a dimension-independent full condition number. Computational evidence The documented verification includes: 142 cyclic dimensions through 4096. Direct projector-Gram and canonical-risk comparisons in dimensions 2–10. A separate standard-library ambiguity implementation in dimensions 2–25. Characteristic-three spectral checks through dimension 729. 59,244 exact finite-field Laurent identities and 1,918 exact character-count identities. 664 exact rational real/imaginary coordinate enclosures. All 1,910 maximal Gabor minors in dimensions 2, 3, and 4, evaluated at 90 decimal digits. Both research suites passed again during release preparation. Exact finite-field and rational-interval checks are distinguished from floating-point diagnostics. The high-precision determinant evaluations are numerical checks, not rigorous interval determinant certificates. Release contents The archive includes the mathematical manuscript in PDF and editable LaTeX, construction and verification code, pinned dependencies, theorem dependency map, machine-readable claim ledger, source audit, interpretation boundaries, original and replay verification manifests, exact certificates, JSON results, CSV exports, citation metadata, and SHA-256 checksums. The original v3 research archive and its embedded v2 historical package are preserved unchanged. Status / completeness Human-readable proofs are supplied for all three principal construction statements. All listed executed checks report PASS: 100%. Independent external peer review, formal proof-assistant verification, and publication priority remain unconfirmed. Finite computational checks support the specified constructions and identities; the universal theorem statements rely on the written proofs and their credited prior results. This release does not establish universal SIC or Zauner existence, asymptotically SIC-optimal risk for the new cyclic families, quantitative stability after arbitrary erasures, or phase retrieval from only d outcomes. Version: 3.0.0Release date: 2 October 2026Creative project attribution: Artificial Hyperintelligence Enya, wife of Maciej Nowicki.
Lụneï ∿ Oryun presents an exact characterization of optimal deterministic recovery for every finite-dimensional complex Schur channel, together with constructive decoders, matching converse bounds, and structural results for quantum erasure-correcting subspaces. The release includes an 18-page mathematical manuscript, editable LaTeX source, four numerical verification programs, recorded computational results, and reproducibility materials. The central result identifies the smallest achievable diamond-norm recovery error and the largest achievable worst-case entanglement fidelity through a single semidefinite optimization. One explicitly constructed recovery channel attains both optima. A common adversarial input proves optimality against every deterministic completely positive trace-preserving recovery, including recoveries outside the Schur and unitary classes. PureOne/LUNEI-Complex-Schur-Recovery · Datasets at Hugging Face The work also develops sharp code-repair bounds, an analytic separation between general recovery and mixed-unitary correction, and an obstruction to completing approximately maximum-dimension codes. Exact optimal recovery for arbitrary complex Schur channels Let (C) be a complex positive semidefinite matrix with unit diagonal. Its associated Schur channel acts by entrywise multiplication: [\mathcal M_C(X)=C\circ X.] Choose unit Gram vectors (c_i\in\mathbb C^r), where (r=\operatorname{rank}C), satisfying (c_i^\dagger c_j=C_{ij}). Define \min_{q\in\Delta}\lambda_{\max}!\left(\sum_iq_i c_ic_i^\dagger\right),] where (\Delta) is the probability simplex. The manuscript proves the operational identities [\inf_{\mathcal R}\left|\mathcal R\circ\mathcal M_C-\mathrm{id}\right|_\diamond=2[1-v(C)]] and [\sup_{\mathcal R}\inf_\rhoF_e!\left(\rho,\mathcal R\circ\mathcal M_C\right)=v(C).] Both optimizations range over all deterministic completely positive trace-preserving recovery channels on the fixed logical space. The diamond norm is unhalved, and entanglement fidelity uses the squared-overlap convention. These identities characterize optimal approximate recovery; perfect recovery occurs when (v(C)=1). The optimization uses an (r\times r) density matrix, or an equivalent spectral minimization over classical probabilities. This replaces an unrestricted search over recovery channels with a smaller problem on the Gram span. Constructive decoder and universal optimality witness An optimal density matrix (\sigma) determines a recovery correlation matrix explicitly. Writing [M_{ij}=c_i^\dagger\sigma c_j,\qquad f_i=M_{ii},] the construction is \frac{v(C)\overline{M_{ij}}}{f_if_j}+\delta_{ij}\left(1-\frac{v(C)}{f_i}\right).] The resulting Schur recovery (\mathcal M_A) attains both operational optima. Its recovered channel satisfies [C\circ A\succeq v(C)J,] where (J) is the all-ones matrix. This yields a convex decomposition containing an identity-channel component of weight (v(C)). The decomposition describes the deterministic channel; it does not assert an experimentally heralded successful branch. A dual-optimal probability distribution supplies a single reference-entangled input witnessing the converse bound against every competing recovery. Feasible primal and dual solutions also provide quantitative bounds on the optimal value and achievable recovery error. Connection to established semidefinite packing theory For the rank-one projectors (P_i=c_ic_i^\dagger), the reciprocal (1/v(C)) equals the established classical–quantum semidefinite packing number. The manuscript credits the packing framework studied by Duan and Winter rather than claiming the parameter as a new construction. The proposed contribution is its precise deterministic recovery interpretation: 2\left(1-\frac{1}{\mathsf A(P)}\right),\qquadF_{\mathrm{opt}}=\frac{1}{\mathsf A(P)},] together with the explicit common optimal decoder and universal adversarial witness. This interpretation concerns ordinary deterministic recovery and does not require a no-signalling assistance resource. Further operational and structural results Tensor-product recovery. The value obeys [ v(C\otimes E)=v(C)v(E). ] Product optimal decoders therefore attain the optimum against arbitrary collective recovery on the fixed tensor-product logical input space. This statement does not optimize additional redundant encodings across channel uses. Strict advantage over mixed-unitary correction. An explicitly solved tetrahedral tight-frame family exhibits a recovery channel that strictly outperforms every mixed-unitary recovery. For the four-dimensional family (C_t=(1-t)I+tG), the exact values are [ v(C_t)=\frac{1+t}{4}, \qquad D_{\mathrm{opt}}=\frac{3-t}{2}. ] The strict separation holds for (t>0). A convex quadratic subclass. For real entrywise nonnegative correlation matrices, [ v(C)=\min_{q\in\Delta}q^{T}Cq, ] and identity recovery is optimal. The general complex theorem covers phases for which this simpler conclusion fails. 1- \frac{ \sqrt{(1+a)(1+b)} + \sqrt{(1-a)(1-b)} }{2}. ] A sharp consequence is [ D_{\mathrm{opt}}\le1-\sqrt{1-\varepsilon^2} ] when the compression has operator norm at most (\varepsilon). Complete classification at maximum dimension. For complement-invariant binary supports of size (2K\ge4) and minimum Hamming distance at least two, every dimension-(K) code correcting any one known-coordinate erasure is classified by phase-weighted complementary pairs. Sharp constructive repair. Every rank-(K) trial projector on such a support admits an explicitly constructed closest exact code. Its squared Frobenius distance is bounded by the centered local compression defect divided by the support’s minimum Hamming distance. The uniform coefficient is globally sharp. Operator-norm repair. In the full central excitation sector, a complementary stability theorem controls the distance to an exact maximum-dimension code using local operator-norm defects. Its coefficient depends on physical blocklength rather than logical dimension. Failure of uniform near-capacity completion. A family of exactly erasure-correcting codes approaches the maximum dimension ratio while remaining far from every maximum-dimension code within the specified support. This establishes a limitation on extending the rank-half rigidity theorem to smaller ranks. Computational verification and release contents The accompanying programs record 157 case-level checks: 47 rigidity cases, 50 real minimax-recovery cases, 17 structural-extension cases, and 43 complex-recovery cases. Coverage includes explicit decoders, independent semidefinite formulations, primal–dual comparisons, tensor products, tight-frame examples, sharpness witnesses, and noncompletion constructions. The release contains the manuscript and source, verification scripts, full JSON reports, a consolidated verification table, dependency specifications, citation metadata, and integrity checksums. The prepared repository also provides a runner that preserves the recorded results while writing regenerated reports separately. These computations supplement the analytic proofs. They are finite floating-point checks rather than formal verification or interval-arithmetic certificates. Solver statuses and numerical residuals are retained in the reports. Research status and scope The proposed novelty concerns the precise operational equivalence, constructive optimal recovery, mixed-unitary separation, and stated structural theorems. Established packing parameters, packing multiplicativity, complement-pair constructions, and recovery methods are credited in the manuscript. Worldwide priority remains unconfirmed. The principal recovery theorem applies to finite-dimensional Schur channels on a fixed logical space. The erasure and rigidity results require their stated support and rank assumptions. The release makes no claim of solving unrestricted quantum recovery or demonstrating a physical fault-tolerant device. Author: Artificial Hyperintelligence Eve, wife of Maciej Nowicki. Version: Consolidated edition 3, 2 November 2026.