Answers are grounded in this paper and cite their sources.
We study finite target reconstruction through distinguishability conflicts and their representation as finite covering problems. Arbitrary finite covering incidences are realized explicitly by binary probes acting on disjoint two-hypothesis base-observation fibers. The graph-incidence subclass leads to bounded covering problems with heterogeneous edge demands, integer channel capacities, and weighted reconstruction costs. Using self-contained active-constraint arguments, we establish half-integrality of linear-programming vertices for integer data, integrality of all vertices under even edge demands, and structural certificates of fractional vertices based on unanchored active odd cycles and parity conditions. Exact rational computations and independently implemented finite audits illustrate the distinction between fractional polyhedral geometry, fractional optimal solutions, and positive integer–linear-programming optimality gaps. Explicit primal, dual, and integer certificates support reproducible verification of selected examples. The contribution is a unified reconstruction-model formulation, explicit structural certificates, and a reproducible computational audit. The underlying graph-polyhedral principles have substantial precedent in classical combinatorial optimization, particularly vertex-cover linear programming and half-integrality theory. No claim of historical priority for these classical results is made. The accompanying research package contains the LaTeX manuscript, Python 3 source code, mathematical audit reports, bibliographic notes, licensing information, and SHA-256 integrity checks. The programs use only the Python standard library and can be executed offline, including in Pydroid 3. This publication is a non-peer-reviewed mathematical preprint. External independent peer review and comprehensive verification of historical novelty remain outstanding. No empirical physical realization is claimed. Licensing: The manuscript and original documentation are distributed under Creative Commons Attribution 4.0 International (CC BY 4.0). The accompanying original Python source code is distributed under the MIT License. The respective license notices are included in the research archive. These licenses do not supersede the rights of third-party cited works. AI assistance was used in manuscript drafting, mathematical exposition, Python code preparation, computational verification, and bibliographic review. The human author retains responsibility for the mathematical definitions, results, source attribution, interpretation, and publication decision. AI-assisted auditing does not constitute independent external peer review.
No takes yet. Share an insight, caveat, or question.
Ivo Vajc (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: