Version 1.4 refines previous findings on Archimedean Stark Theorems in real quadratic fields, suggesting new insights on their implications.
Version 1.4 is a correction-and-certificate release layered over the immutable v1.3 results companion. Its reserved release DOI is 10.5281/zenodo.21712478. It preserves the proved Engine-C packet polynomials, magnitudes, and the sigma-positive Fourier convention, while adding the later provenance correction that withdraws the claim of a data-independent direct/inverse character choice in the five-control phase replay. The retained numerical statement is only that one of the two conjugate orientations has a unique quarter-turn match in each case. The release also records the proved replacement corollary: in each of the five certified cyclic-quartic Stark cases satisfying Roblot's hypotheses (A1)--(A3), his weak solution differs from the proved Stark unit by a trivial group-ring unit, so their character coefficients have ratio in mu_4. No raw quarter-turn label is promoted without exact Artin transport. Finally, version 1.4 adds the first fully worked nonzero imprimitive Engine-A row, RQ-000013. Exact conductor, Euler-factor, class-number, unit-norm-kernel, determinant-index, root-isolation, and Artin-action checks give E_chi=2, I_chi=2, and X_[0]=u^2, X_[1]=u^(-2). The included bnrL1 point-value comparison is explicitly quarantined and is not part of the proof. The top-level paper differs from v1.3 only by the release-DOI update and by correcting the Tangedal--Young bibliography pagination from 1022--1045 to 1045--1061; an exact source-diff audit is included. The v1.3 supplement changes only its archive DOI. A new top-level certified addendum and deterministic correction companion contain the correction records and replay.
No takes yet. Share an insight, caveat, or question.
Hainan Zhao (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: