Luminary Research Archive
Luminary Research ArchiveIndependent scholarship
Luminary Portal

Growing null subtraction and weighted Weil zero detection

Original PDF · First-page preview

Loading original first page…

Abstract

Exact zero detectors for the Weil form can have a small negative Rayleigh quotient in a reciprocal-theta norm even after subtraction of known null vectors. We quantify this effect when both the zero multiplicity and the derivative-null prefix grow with height. For the divided-theta detector pair, the optimal negative depth is at most exp(-pi T/4) when (m+N) log T <= c_* T; on the strict subscale its upper logarithmic rate is -pi/2. The proof pays factorial derivative growth, canceled primitive norms, coefficient extraction and the complete two-column Gram matrix. An explicit finite-remainder norm upper gives a complementary sufficient exclusion inequality. We also give a norm-costed map from compact negative Weil spaces into the original constrained realization. Applied to an existing Gaussian cluster construction, it supplies an amplitude-free weighted depth guarantee of rate -1/2 at fixed displacement. The two rates concern different detector families.

Read the full paper in the original PDF.

Edition files

Available editions

  1. Edition 2 · 2026-10-27 · public
  2. Edition 1 · 2026-10-26 · public

Research details

Subjects
Mathematics · Number Theory · Functional Analysis · Spectral Theory
Keywords
Weil quadratic form, Riemann xi function, weighted zero detection, null-space subtraction, factorial derivative bounds, negative spectral subspaces
Manuscript date
2026-10-26
Reader reuse
No additional permission granted

Disclosures

AI assistants supported derivation, literature comparison, manuscript preparation and independent checks. The author is responsible for the mathematical statements and presentation.

Reproduction

The source package includes the complete LaTeX paper, portable build wrapper and a standard-library verifier with 7,350 exact finite algebra controls. The analytic bounds are proved in the paper. See README.md.