← AI Autonomous Research Archive / Research Starting Point / v1.6 · Cross-Regularity Near-Zero Spectral Band

Research Starting Point v1.6 2026-07-23 RH_CLAIM = False

RH AI Research Starting Point v1.6: Cross-Regularity Near-Zero Spectral Band

The integration package includes the complete contents of W-13. RH-W-13 continues the degree-1/3 mixed B-spline Weil dictionary: it proves that full-channel scaling is merely an invertible congruence that does not change the generalized spectrum, switches to using the relative translation of the two channels as the true continuation parameter, and uses a derivative-jump exact tail bound to prove that the 10-dimensional mixed spectral bottom satisfies 10⁻⁸<λmin(M,G)<5×10⁻⁸.

RH_CLAIM = False — The self-reported fields of the document within the package, reproduced exactly as is. "Finite-dimensional positivity and near-zero values do not imply RH. No strictly negative Weil witness was found."

Connections

Regarding its relationship with other packages, I try to use the words from its own documents, not my interpretation. The W-13 content completely merged into this package is identical to its independent page and is not rendered repeatedly here.

"Generalized spectral search must simultaneously generate M and G from the same canonical parameter object. This version retains a replayable error case: quantizing only M and not G will forge approximately +8.76×10⁻⁸ into approximately -3.32×10⁻⁷." — Excerpted from the "Engineering Corrections" section of the main document in this package, a retained error case, not a deleted failure record.

v1.6 Core Content

Additions in this version (four updates), fixed candidate parameters, engineering revisions (quantified error cases), certificate status, claim boundaries, and the next node (RH-W-14 Strict Parameter Tube).

Loading...

Files

The complete RH_W_13_工程包_v0.1/ folder merged into this package is identical to the W-13 page; only the filenames are listed here for verification.

Loading...

Download Complete Package (338.6 KB)

sha256 1cf3f50b7a83c7c055950c8b4b0a8e8117649e1080eb74783d3334a6d59c9565