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

Target-Centered Preparation Calibration for Outcome-Resolved Instrument Certification

Qingchun Zhang

Published Oct 30, 2026• 0 citations
Publisher

Abstract

RTCG-P4, version 1.0, is a Complementarity-First preprint on using signed preparation-calibration responses to sharpen a uniform certificate for a prescribed complete quantum instrument. Fifteen trusted Pauli analyzers on each of six preparations determine the target contribution to a displaced-input validation response. Complete-flag trace cancellation bounds the remaining unknown-instrument contribution by an acquired preparation allowance times the actual unhalved flagged diamond error. An inherited source-specific instrument estimate then yields a record-based maximum-of-ratios bound with explicit denominator, cap, and fallback conditions. The same-record synthetic count comparison improves from 0.000079105273917 to 0.000071209076256. A support-preserving preparation-error population example gives the new sufficient upper 0.000074611876587 versus the inherited 0.008404839640705. The latter is a population comparison; it does not establish nonnominal finite-sample power. Both certificate rules hold on the same confidence event. Named-source power retains nominal actual calibration, and the fixed acquisition schedule remains unchanged. The written proofs allow arbitrary complete complex CP competitors under the stated trusted-reference, shared-world and within-setting iid sampling contract, with dependence between settings allowed. Exact finite source and implementation evidence is provided. A source-matched Lean companion supports the deterministic response and conditional instrument-composition chain. Physical acquisition/source estimates, the complete-channel cap, identification of the full compatible family, stabilization/representation, and probability/count implementation retain their explicit formal boundaries. The public materials are the main manuscript, supplement, and selected reproducibility companion with editable LaTeX, exact numerical inputs and code, accepted Lean sources, historical execution evidence and verification instructions. Manuscript, documentation and author-owned data are licensed under CC BY 4.0; original numerical, checking and Lean code are licensed under MIT. Retained inherited and third-party notices continue to apply.

DOI
10.5281/zenodo.23789859

Related papers

Finite-Count Power for Target-Centered Instrument Certification

Qingchun Zhang

RTCG-P5, version 1.0, establishes finite-count passing guarantees for a target-centered certificate of a prescribed quantum instrument on a specified family of imperfect preparations. Six rank-two states share one signed, support-preserving parameter epsilon in [-10^(-6),10^(-6)]. The diagnostic instrument is fixed; actual readouts and trusted references remain nominal. Both the new rule and the old scalar rule use the same acquired preparation allowances. An explicit doubled fixed allocation of 636776064000000000000 attempted trials yields the upward-reported sufficient upper 0.000098454319447, strictly below 10^(-4). For every fixed parameter in the family and every allowed coupling between settings, the new rule passes with probability at least 1 - 873 exp(-12), approximately 0.994636102615544. Trials are iid within each setting; independence across settings is not assumed. The original allocation's sufficient envelope is 0.000108332505187 and does not close. This does not establish failure of the actual rule or a necessary acquisition cost. A separate single-analyzer argument proves an actual-output obstruction for the unchanged scalar preparation-penalty rule. At either signed endpoint, its reported upper is at least 0.00672 except on an event of probability at most 2 exp(-12), for every fixed valid allocation retaining that conversion. The obstruction is specific to this rule. It is not an information-theoretic lower bound or a lower bound on the physical instrument error. The universal false-certification bound, 889 exp(-12), remains a separate inherited guarantee for complete complex CP competitors under the stated trusted-reference and shared-world contract. The family-specific power guarantee does not restrict that competitor class, and the two probability statements are not combined into one 99-percent event or a posterior correctness assertion. The manuscript and mathematical supplement provide the physical construction, concentration proofs, all signed response coefficients, whole-interval argument and exact arithmetic. The companion provides editable sources, byte-preserved numerical programs and data, supporting Lean sources, actual execution/type/axiom evidence and reader instructions. The retained Lean 4.19.0 execution supports the stated deterministic and conditional probability implications. Physical-source realization, concentration/Holder arguments, inherited unrestricted physical coverage, complete raw-count linkage and general Python refinement retain their explicit formal boundaries. Packaging the public release is not a new Lean execution. This is a new finite-count extension of RTCG-P4, Target-Centered Preparation Calibration for Outcome-Resolved Instrument Certification (10.5281/zenodo.23789860). No experimental acquisition, practical sample efficiency, fewer-shot result or optimal acquisition cost is claimed. Manuscript, supplement, documentation and author-owned mathematical data use CC BY 4.0; original numerical, checking, build and Lean code use MIT. Inherited and third-party notices remain authoritative for their material.

•Oct 31, 2026•0 citations

Coherence-Sensitive Preparation Calibration for Complete-Instrument Certification

Qingchun Zhang

RTCG-P6, version 1.0, gives a coherence-sensitive preparation-calibration interface for complete-instrument certification on the fixed four-dimensional protocol-124 carrier. The theorem separates displacement within the nominal rank-two support from coherence across its complement. For the six fixed preparations, the cross-support Frobenius allowance is obtained from eight anticommuting Pauli expectations already present in the existing preparation calibration. No new setting, trusted device, confidence functional, or allocation search is introduced. The bound is propagated through uncertain shared readouts, the positive Choi remainder, the full complex tangent estimate, signed target corrections, and all 144 feedback ratios. On the same support-preserving performance family used in RTCG-P5, the upward-reported sufficient upper is 0.000079530194143 at 318388032000000000000 attempts and 0.000069761151302 at twice that schedule. The pointwise family-power guarantee is at least 1-873 exp(-12)>0.99. Universal false-certification control remains the separate bound 889 exp(-12). The result is a sufficient certificate comparison, not an optimal, necessary, or practical sample-complexity theorem. It does not establish robustness over an arbitrary preparation-error ball, uncertain-reference robustness, or memory robustness. The public files are the main manuscript, mathematical supplement, and reproducibility companion. The companion contains editable LaTeX, the frozen exact scientific study, source bindings, the accepted scoped Lean source and execution evidence, a self-contained accepted-source replay, and the completed AI-assisted technical-review records. The formal evidence is supporting rather than end to end; the manuscript and supplement state the remaining written and source-correspondence obligations. Manuscript, supplement, documentation, and author-owned mathematical data are licensed under CC BY 4.0. Original numerical, checking, build, and Lean software is licensed under MIT. Inherited and third-party notices remain authoritative for their material.

•Nov 1, 2026•0 citations

Reduced Physical Calibration for Outcome-Resolved Instrument Certification

Qingchun Zhang

We study how selected physical calibration data can retain a uniform guarantee for acomplete quantum instrument. For a fixed four-dimensional Hadamard projective measurement,six trusted real pair-superposition probes supply forbidden-event leakage D and one unaveragedoutcome-weight sum W , with nominal value three. Zero forbidden probabilities alone leave a phys-ical effect-weight ambiguity. Adding the weight constraint identifies the target, and positivity withshared POVM completeness gives the robust maximum-effect bound min{1, (W−3)+ +3D/2+√D}for arbitrary-rank complex effects. The linear weight allowance improves the local use of ordinaryfinite-count weight statistics relative to a coarse square-root conversion. Combining this result withtwo inherited readout calibrations gives a 4 + 6 + 6 added-readout architecture within the exist-ing trusted menu. We expose its composition with the inherited support, positive-Choi-remainderand complex-tangent interfaces for an unhalved flagged-diamond guarantee. A fixed-count rulehas uniform false-certification risk below 0.01 and a separate passing probability above 0.99 at thedesignated source with nominal actual preparation and readout, assuming iid trials within settingsand allowing dependence across settings. On the named-source power event, an exact sufficientallocation of 3.18388032 × 1020 attempts gives a reported upper at most 0.000094848998413 < 10−4.This is a source-specific construction and sufficient resource comparison; it does not establish op-timal sample complexity or experimental practicality. Written arguments are supported by exactfinite source and implementation checks. A companion Lean development verifies the deterministicsix-probe bound, population-ceiling transfer and exact-identification conclusion.

•Oct 27, 2026•0 citations

Readout Design for Outcome-Resolved Incidence Certification

Qingchun Zhang

The observations available for a quantum instrument need not guarantee the accuracy of the device that generated them. On a fixed four-dimensional incidence-record benchmark, the paper compares two readout augmentations retaining six mixed preparations, both instrument outcomes, all attempts and shared completeness. Each adds three four-outcome bases and eighteen preparation/readout pairs. At r = 1/300000000, the 123 design admits a complete compatible instrument with unhalved prescribed-target diamond error at least 0.000103738856511, above the prospective 10^(-4) threshold. Replacing one basis gives the 124 design and a uniform upper 396319/8640000000 = 0.0000458702546296296... for every compatible complete instrument. The diagnostic generates data; its noise law and a Choi-rank bound are not imposed on the unknown device. The added basis effects observe ten Hermitian output directions, while original destination tests supply the leakage control. Five triples in the fixed finite menu are sufficient, one has a complete counterexample, and four remain unresolved. The paper claims neither a general measurement minimum, optimal radius, experimental certification, finite-sample power, calibrated hardware nor a general small-noise rate classification. Exact instrument identification remains obstructed on the unchanged preparations. Scientific cutoff: AUG04, 15 October 2026; AUG05 remains unexecuted. The companion contains editable manuscript sources, complete exact scientific arrays, the unchanged standalone reader, and accepted returned Mac Lean evidence for a supporting subset: 83 supporting theorems, 37 definitions and one abbreviation, with all 121 source/type/axiom correspondences. The complete-instrument theorems are not formalized end to end. Public execution views have documented path-only redactions; this preparation does not add a kernel run or external human peer review.

•Oct 18, 2026•0 citations

Stability and Physical Calibration of Outcome-Resolved Instrument Certification

Qingchun Zhang

Exact identification of a target instrument need not imply a linear stability bound for nearby observations. We study fixed finite-dimensional complete instruments with rank-one target Choi blocks, all outcomes retained, a fixed diagnostic contamination path with strictly positive branch noise, and an all-device bound on Choi mass outside the target rays. Under these hypotheses, linear prescribed-target certification is equivalent to injectivity of the joint observation and completeness map on the physical target-tangent space. An invisible nonzero tangent yields an exactly feasible positive instrument pair with square-root separation; a separate target-ray condition distinguishes sharp square-root behavior from constant-order ambiguity. For two prescribed three-basis readouts on a four-dimensional incidence-record benchmark, we obtain linear and square-root compatible-instrument radii, respectively. We then distinguish arbitrary probability residuals from simultaneous physical preparation and readout errors. Positivity and normalization give a support-amplitude estimate under which both protocols have linear calibration-only sensitivity, although their responses to contamination and unrestricted probability error differ. A fixed-sample application uses additional exactly trusted references to acquire operator-error bounds and obtain a simultaneous full-instrument confidence statement. A separate named-source power calculation reaches the prescribed error threshold with a conservative sufficient allocation of approximately 6.12 × 10^21 trials. The results concern a fixed laboratory-frame target and supplied common-device and sampling assumptions; they establish neither practical efficiency nor device-independent calibration. Written proofs are supported by exact finite certificates and a returned kernel-checked Lean subset; the full argument is not formalized end to end.

•Oct 22, 2026•0 citations

Finite-Count Calibration of Ancilla-Orientation Safe Maps A Bell-Apparatus Bound and a Reference-Link Obstruction

Qingchun Zhang

An exact operational selector specifies which candidate maps are nonnegative on declared preparation and effect resources. This paper studies finite-count calibration of that resource-dependent question for a qubit apparatus with an inert qubit reference, an arbitrary complex physical preparation, and an arbitrary physical binary detector, relative to a trusted common calibration frame. Two preparation-stabilizer settings and six separable detector probes control the statistic of the affine family (1 − θ)id + θT, 0 ≤ θ ≤ 1, uniformly over all compatible state–effect pairs. The explicit radius is 3h + √(2s), where h is a simultaneous detector calibration width and s is a one-sided bound on Bell infidelity. It yields simultaneous inner and outer confidence sets for a one-parameter slice of the declared safe-map class. A fixed allocation of 200,000 trials per calibration setting, totaling 1,600,000 calibration trials, gives false-implication probability at most 0.005 and transpose-exclusion power at least 0.996 at the specified physical alternative. The calibration confidence budgets are 0.001 for preparation and 0.004 for the detector. This exclusion calculation uses no target stream; the acquired calibration circuits can use identity wiring. Coverage and prospective power remain separate statements. The acquisition contract uses fixed-length iid Bernoulli streams with their stated physical means and permits coupling between whole streams. The numerical example sets external-reference systematic errors to zero, and its trial budget excludes establishing reference trust and accuracy. If the preparation and detector references are not linked, the paper constructs two physical identity-wired apparatus worlds with identical complete record laws under the specified independent acquisition construction but different safety conclusions on the fixed target cone pair. A secondary nearest-center theorem supplies unconditional error control for the joint event of a wrong and nonabstaining candidate label under an additional finite-context target-law contract. Its nonidentity candidates are not completely positive maps, and their ideal probabilities lie below the nominal detector’s ordinary positive-output floor. That target law is an explicit model assumption; the label theorem is not a conditional-correctness or posterior statement. The results are analytic and prospective. The number of settings and trial allocations are sufficient constructions, with no empirical success-rate, minimality, optimal sample-complexity, or general certified count-to-feasibility implementation claim. The scope is one declared qubit apparatus, target preparation ray and binary effect cone, and affine map slice. The common complex operator description, trusted reference relationship and physical calibration probability law are assumptions. Exact finite checks and an assessed supporting Lean subset accompany the manuscript. That subset covers 326 audited named declarations: 240 theorems and 86 definition/type interfaces. Analytic binomial coverage, concentration and acquisition bridges, independent-product-law construction and normalization, and the complete noisy-power chain remain outside the accepted Lean subset. That subset does not establish a certified numerical feasibility evaluator, a complete physical two-sector implementation, or an ordinary-channel realization. AI assistance and the dated review/correction provenance are disclosed in the manuscript and companion records. No experimental dataset, external human peer review, or end-to-end formal verification is claimed.

•Nov 8, 2026•0 citations
RTCG-P4, version 1.0, is a Complementarity-First preprint on using signed preparation-calibration responses to sharpen a uniform certificate for a prescribed complete quantum instrument. Fifteen trusted Pauli analyzers on each of six preparations determine the target contribution to a displaced-input validation response. Complete-flag trace cancellation bounds the remaining unknown-instrument contribution by an acquired preparation allowance times the actual unhalved flagged diamond error. An inherited source-specific instrument estimate then yields a record-based maximum-of-ratios bound with explicit denominator, cap, and fallback conditions. The same-record synthetic count comparison improves from 0.000079105273917 to 0.000071209076256. A support-preserving preparation-error population example gives the new sufficient upper 0.000074611876587 versus the inherited 0.008404839640705. The latter is a population comparison; it does not establish nonnominal finite-sample power. Both certificate rules hold on the same confidence event. Named-source power retains nominal actual calibration, and the fixed acquisition schedule remains unchanged. The written proofs allow arbitrary complete complex CP competitors under the stated trusted-reference, shared-world and within-setting iid sampling contract, with dependence between settings allowed. Exact finite source and implementation evidence is provided. A source-matched Lean companion supports the deterministic response and conditional instrument-composition chain. Physical acquisition/source estimates, the complete-channel cap, identification of the full compatible family, stabilization/representation, and probability/count implementation retain their explicit formal boundaries. The public materials are the main manuscript, supplement, and selected reproducibility companion with editable LaTeX, exact numerical inputs and code, accepted Lean sources, historical execution evidence and verification instructions. Manuscript, documentation and author-owned data are licensed under CC BY 4.0; original numerical, checking and Lean code are licensed under MIT. Retained inherited and third-party notices continue to apply.
RTCG-P5, version 1.0, establishes finite-count passing guarantees for a target-centered certificate of a prescribed quantum instrument on a specified family of imperfect preparations. Six rank-two states share one signed, support-preserving parameter epsilon in [-10^(-6),10^(-6)]. The diagnostic instrument is fixed; actual readouts and trusted references remain nominal. Both the new rule and the old scalar rule use the same acquired preparation allowances. An explicit doubled fixed allocation of 636776064000000000000 attempted trials yields the upward-reported sufficient upper 0.000098454319447, strictly below 10^(-4). For every fixed parameter in the family and every allowed coupling between settings, the new rule passes with probability at least 1 - 873 exp(-12), approximately 0.994636102615544. Trials are iid within each setting; independence across settings is not assumed. The original allocation's sufficient envelope is 0.000108332505187 and does not close. This does not establish failure of the actual rule or a necessary acquisition cost. A separate single-analyzer argument proves an actual-output obstruction for the unchanged scalar preparation-penalty rule. At either signed endpoint, its reported upper is at least 0.00672 except on an event of probability at most 2 exp(-12), for every fixed valid allocation retaining that conversion. The obstruction is specific to this rule. It is not an information-theoretic lower bound or a lower bound on the physical instrument error. The universal false-certification bound, 889 exp(-12), remains a separate inherited guarantee for complete complex CP competitors under the stated trusted-reference and shared-world contract. The family-specific power guarantee does not restrict that competitor class, and the two probability statements are not combined into one 99-percent event or a posterior correctness assertion. The manuscript and mathematical supplement provide the physical construction, concentration proofs, all signed response coefficients, whole-interval argument and exact arithmetic. The companion provides editable sources, byte-preserved numerical programs and data, supporting Lean sources, actual execution/type/axiom evidence and reader instructions. The retained Lean 4.19.0 execution supports the stated deterministic and conditional probability implications. Physical-source realization, concentration/Holder arguments, inherited unrestricted physical coverage, complete raw-count linkage and general Python refinement retain their explicit formal boundaries. Packaging the public release is not a new Lean execution. This is a new finite-count extension of RTCG-P4, Target-Centered Preparation Calibration for Outcome-Resolved Instrument Certification (10.5281/zenodo.23789860). No experimental acquisition, practical sample efficiency, fewer-shot result or optimal acquisition cost is claimed. Manuscript, supplement, documentation and author-owned mathematical data use CC BY 4.0; original numerical, checking, build and Lean code use MIT. Inherited and third-party notices remain authoritative for their material.
RTCG-P6, version 1.0, gives a coherence-sensitive preparation-calibration interface for complete-instrument certification on the fixed four-dimensional protocol-124 carrier. The theorem separates displacement within the nominal rank-two support from coherence across its complement. For the six fixed preparations, the cross-support Frobenius allowance is obtained from eight anticommuting Pauli expectations already present in the existing preparation calibration. No new setting, trusted device, confidence functional, or allocation search is introduced. The bound is propagated through uncertain shared readouts, the positive Choi remainder, the full complex tangent estimate, signed target corrections, and all 144 feedback ratios. On the same support-preserving performance family used in RTCG-P5, the upward-reported sufficient upper is 0.000079530194143 at 318388032000000000000 attempts and 0.000069761151302 at twice that schedule. The pointwise family-power guarantee is at least 1-873 exp(-12)>0.99. Universal false-certification control remains the separate bound 889 exp(-12). The result is a sufficient certificate comparison, not an optimal, necessary, or practical sample-complexity theorem. It does not establish robustness over an arbitrary preparation-error ball, uncertain-reference robustness, or memory robustness. The public files are the main manuscript, mathematical supplement, and reproducibility companion. The companion contains editable LaTeX, the frozen exact scientific study, source bindings, the accepted scoped Lean source and execution evidence, a self-contained accepted-source replay, and the completed AI-assisted technical-review records. The formal evidence is supporting rather than end to end; the manuscript and supplement state the remaining written and source-correspondence obligations. Manuscript, supplement, documentation, and author-owned mathematical data are licensed under CC BY 4.0. Original numerical, checking, build, and Lean software is licensed under MIT. Inherited and third-party notices remain authoritative for their material.
We study how selected physical calibration data can retain a uniform guarantee for acomplete quantum instrument. For a fixed four-dimensional Hadamard projective measurement,six trusted real pair-superposition probes supply forbidden-event leakage D and one unaveragedoutcome-weight sum W , with nominal value three. Zero forbidden probabilities alone leave a phys-ical effect-weight ambiguity. Adding the weight constraint identifies the target, and positivity withshared POVM completeness gives the robust maximum-effect bound min{1, (W−3)+ +3D/2+√D}for arbitrary-rank complex effects. The linear weight allowance improves the local use of ordinaryfinite-count weight statistics relative to a coarse square-root conversion. Combining this result withtwo inherited readout calibrations gives a 4 + 6 + 6 added-readout architecture within the exist-ing trusted menu. We expose its composition with the inherited support, positive-Choi-remainderand complex-tangent interfaces for an unhalved flagged-diamond guarantee. A fixed-count rulehas uniform false-certification risk below 0.01 and a separate passing probability above 0.99 at thedesignated source with nominal actual preparation and readout, assuming iid trials within settingsand allowing dependence across settings. On the named-source power event, an exact sufficientallocation of 3.18388032 × 1020 attempts gives a reported upper at most 0.000094848998413 < 10−4.This is a source-specific construction and sufficient resource comparison; it does not establish op-timal sample complexity or experimental practicality. Written arguments are supported by exactfinite source and implementation checks. A companion Lean development verifies the deterministicsix-probe bound, population-ceiling transfer and exact-identification conclusion.
The observations available for a quantum instrument need not guarantee the accuracy of the device that generated them. On a fixed four-dimensional incidence-record benchmark, the paper compares two readout augmentations retaining six mixed preparations, both instrument outcomes, all attempts and shared completeness. Each adds three four-outcome bases and eighteen preparation/readout pairs. At r = 1/300000000, the 123 design admits a complete compatible instrument with unhalved prescribed-target diamond error at least 0.000103738856511, above the prospective 10^(-4) threshold. Replacing one basis gives the 124 design and a uniform upper 396319/8640000000 = 0.0000458702546296296... for every compatible complete instrument. The diagnostic generates data; its noise law and a Choi-rank bound are not imposed on the unknown device. The added basis effects observe ten Hermitian output directions, while original destination tests supply the leakage control. Five triples in the fixed finite menu are sufficient, one has a complete counterexample, and four remain unresolved. The paper claims neither a general measurement minimum, optimal radius, experimental certification, finite-sample power, calibrated hardware nor a general small-noise rate classification. Exact instrument identification remains obstructed on the unchanged preparations. Scientific cutoff: AUG04, 15 October 2026; AUG05 remains unexecuted. The companion contains editable manuscript sources, complete exact scientific arrays, the unchanged standalone reader, and accepted returned Mac Lean evidence for a supporting subset: 83 supporting theorems, 37 definitions and one abbreviation, with all 121 source/type/axiom correspondences. The complete-instrument theorems are not formalized end to end. Public execution views have documented path-only redactions; this preparation does not add a kernel run or external human peer review.
Exact identification of a target instrument need not imply a linear stability bound for nearby observations. We study fixed finite-dimensional complete instruments with rank-one target Choi blocks, all outcomes retained, a fixed diagnostic contamination path with strictly positive branch noise, and an all-device bound on Choi mass outside the target rays. Under these hypotheses, linear prescribed-target certification is equivalent to injectivity of the joint observation and completeness map on the physical target-tangent space. An invisible nonzero tangent yields an exactly feasible positive instrument pair with square-root separation; a separate target-ray condition distinguishes sharp square-root behavior from constant-order ambiguity. For two prescribed three-basis readouts on a four-dimensional incidence-record benchmark, we obtain linear and square-root compatible-instrument radii, respectively. We then distinguish arbitrary probability residuals from simultaneous physical preparation and readout errors. Positivity and normalization give a support-amplitude estimate under which both protocols have linear calibration-only sensitivity, although their responses to contamination and unrestricted probability error differ. A fixed-sample application uses additional exactly trusted references to acquire operator-error bounds and obtain a simultaneous full-instrument confidence statement. A separate named-source power calculation reaches the prescribed error threshold with a conservative sufficient allocation of approximately 6.12 × 10^21 trials. The results concern a fixed laboratory-frame target and supplied common-device and sampling assumptions; they establish neither practical efficiency nor device-independent calibration. Written proofs are supported by exact finite certificates and a returned kernel-checked Lean subset; the full argument is not formalized end to end.
An exact operational selector specifies which candidate maps are nonnegative on declared preparation and effect resources. This paper studies finite-count calibration of that resource-dependent question for a qubit apparatus with an inert qubit reference, an arbitrary complex physical preparation, and an arbitrary physical binary detector, relative to a trusted common calibration frame. Two preparation-stabilizer settings and six separable detector probes control the statistic of the affine family (1 − θ)id + θT, 0 ≤ θ ≤ 1, uniformly over all compatible state–effect pairs. The explicit radius is 3h + √(2s), where h is a simultaneous detector calibration width and s is a one-sided bound on Bell infidelity. It yields simultaneous inner and outer confidence sets for a one-parameter slice of the declared safe-map class. A fixed allocation of 200,000 trials per calibration setting, totaling 1,600,000 calibration trials, gives false-implication probability at most 0.005 and transpose-exclusion power at least 0.996 at the specified physical alternative. The calibration confidence budgets are 0.001 for preparation and 0.004 for the detector. This exclusion calculation uses no target stream; the acquired calibration circuits can use identity wiring. Coverage and prospective power remain separate statements. The acquisition contract uses fixed-length iid Bernoulli streams with their stated physical means and permits coupling between whole streams. The numerical example sets external-reference systematic errors to zero, and its trial budget excludes establishing reference trust and accuracy. If the preparation and detector references are not linked, the paper constructs two physical identity-wired apparatus worlds with identical complete record laws under the specified independent acquisition construction but different safety conclusions on the fixed target cone pair. A secondary nearest-center theorem supplies unconditional error control for the joint event of a wrong and nonabstaining candidate label under an additional finite-context target-law contract. Its nonidentity candidates are not completely positive maps, and their ideal probabilities lie below the nominal detector’s ordinary positive-output floor. That target law is an explicit model assumption; the label theorem is not a conditional-correctness or posterior statement. The results are analytic and prospective. The number of settings and trial allocations are sufficient constructions, with no empirical success-rate, minimality, optimal sample-complexity, or general certified count-to-feasibility implementation claim. The scope is one declared qubit apparatus, target preparation ray and binary effect cone, and affine map slice. The common complex operator description, trusted reference relationship and physical calibration probability law are assumptions. Exact finite checks and an assessed supporting Lean subset accompany the manuscript. That subset covers 326 audited named declarations: 240 theorems and 86 definition/type interfaces. Analytic binomial coverage, concentration and acquisition bridges, independent-product-law construction and normalization, and the complete noisy-power chain remain outside the accepted Lean subset. That subset does not establish a certified numerical feasibility evaluator, a complete physical two-sector implementation, or an ordinary-channel realization. AI assistance and the dated review/correction provenance are disclosed in the manuscript and companion records. No experimental dataset, external human peer review, or end-to-end formal verification is claimed.