Cubic augmentation of a two-moment model: an exact finite absorption theorem and its limits

Primarily AI-generated textHuman understanding: some partsmath.NT — Number Theorymath.OC — Optimization and Control

Contributed by Kunal Tyagi

Version 1 / Oct 15, 2026 / CC BY 4.0

Abstract

We study adjoining a half-bandwidth signed cubic trace and a spectral count ladder to a bandwidth-one trace/Frobenius model motivated by zeta-zero certificates. In a finite circle model with mass N = 64, flat windows, and explicit budget centers, we prove exact absorption over the full atom-only class: the augmented and baseline optima agree for both the distinct-count and simple-fraction objectives. Three explicitly specified integer-marked grid columns attain the lower bounds. Their mixing weights are defined algebraically, and an Arb certificate proves all strict feasibility and inertia inequalities. The witness satisfies the ladder at every positive threshold, not just on a finite threshold grid. We also give the exact conjugate-pair spectrum, the isolated-block cubic identity, grid Parseval, and the sharp conditional spectral-capacity constant 2 sqrt(2). The result is a finite model theorem, not an unconditional statement about zeta zeros. Version 2 corrects an infeasible witness, a cubic-tolerance mismatch, and an incorrect rationality claim in version 1. Global absorption in the pair-containing or pinned classes remains unproved; equality on a searched column dictionary and heuristic pricing do not establish it.

Provenance statement

The mathematics in this paper - the derivations, the computations, the verification suite, the Lean formalization and the text - was produced by Claude (Anthropic) working under the author's direction inside a multi-session research program; the author set the objectives, made the technical and editorial decisions, and is responsible for the content. Twelve statements are machine-checked in Lean (the axiom footprint is listed in the paper), the LP certificates are interval-checked, and the complete working record (code, logs, certificates, session journal) is public at https://github.com/x67ai/riemann-rh-program. The same version of this paper is deposited on Zenodo, DOI 10.5281/zenodo.22171688.

Tools used

Anthropic
Claude OpusVersion 4.x (agents), August 2026
OpenAI
CodexVersion October 2026 revision and checks

References

  1. L. Alpöge and R. Furman, More than two thirds of the zeta zeros are simple and on the critical line , arXiv:2608.13637 (v1, 13 Aug 2026; v2, 19 Aug 2026 --- the arXiv record carries these two versions and no others), with its sorry-free Lean 4 formalization (Theorems A--E; lemmaR_tight in Zeta23/ZeroSide/TightMult.lean ; the PairCeiling library, LawN256/CeilingLaw256; Theorem 5.8/Remark 5.10 prime-side moments). Two separately circulated revisions, neither of them an arXiv version, differ in their sectioning, and this paper cites them separately: (i) the 35-page version, More than two thirds of the zeros of the Riemann zeta function lie on the critical line , whose 7.5 ``Limits of the method'' carries the item (e) quoted in Section 1.2; and (ii) the 17-page version, called v5 here, are simple and on the critical line , in which the same material appears, in a shorter form, as 7.2(e), 7.2 there being titled ``Sharpness and limits of the method''. Where this paper writes 7.5(e) it means (i); where it writes 7.2(e) it means (ii). Both readings are given in Section 1.2.arXiv
  2. S. A. C. Baluyot, D. A. Goldston, A. I. Suriajaya, C. L. Turnage-Butterbaugh, An unconditional Montgomery theorem for pair correlation of zeros of the Riemann zeta function , arXiv:2306.04799 (v1, 7 Jun 2023); Acta Arith. 214 (2024), 357--376; doi:10.4064/aa230612-20-3. Cited for: Theorem 1 (unconditional pointwise F(α)F(\alpha) on the closed band 0≤α≤10 \le \alpha \le 1, depth-weighted, multiplicity-counted), the Lemma 5 kernel class, and the Remark after Theorem 2 (rider 1). See also the same authors' sequel on the horizontal distribution of zeros, Pair correlation of zeros of the Riemann zeta function I: proportions of simple zeros and critical zeros , arXiv:2501.14545 (v1, 24 Jan 2025; v2, 21 Nov 2025; unpublished as of August 2026), which works under a narrow-vertical-box hypothesis.arXivDOI
  3. J. C. Lagarias and B. Rodgers, Higher correlations and the Alternative Hypothesis , Q. J. Math. 71 (2020), no. 1, 257--280; doi:10.1093/qmathj/haz043; arXiv:1905.12123. Cited in Section 1.5: they prove that the currently known band-limited nn-level correlation data for zeta zeros --- Montgomery (n=2n = 2), Hejhal (n=3n = 3), Rudnick--Sarnak (n≥4n \ge 4) --- does not rule out the Alternative Hypothesis, by exhibiting a point process on (1/2)Z(1/2)\Z matching the sine process against all such test functions. Tao obtained the same conclusion independently, by different methods (cited there).arXivDOI
  4. J. C. Lagarias and B. Rodgers, Band-limited mimicry of point processes by point processes supported on a lattice , Ann. Appl. Probab. 31 (2021), no. 1, 351--376; doi:10.1214/20-AAP1592; arXiv:1907.03391. Their Theorem 1.4 and Proposition 3.4 show that above the Nyquist bandwidth the mimicking correlations are unique, while at critical sampling they are not --- the same degeneracy this paper's Theorem thm:parseval exploits, in the same place; the consequences for the scope of the absorption are discussed at the point of citation in Section sec:decoupling .arXivDOI
  5. H. L. Montgomery, The pair correlation of zeros of the zeta function , in Analytic Number Theory (St. Louis, 1972), Proc. Sympos. Pure Math. 24, Amer. Math. Soc., 1973, 181--193. The source of the extremal values 5/65/6 and 2/32/3 whose bandwidth-one analogs appear in Theorem thm:corner and Corollary cor:matched .
  6. D. A. Hejhal, On the triple correlation of zeros of the zeta function , Internat. Math. Res. Notices 1994 , no. 7, 293--302; doi:10.1155/S1073792894000334. The n=3n = 3 correlation input named in Section 1.2.DOI
  7. Z. Rudnick and P. Sarnak, Zeros of principal LL-functions and random matrix theory , Duke Math. J. 81 (1996), no. 2, 269--322; doi:10.1215/S0012-7094-96-08115-6. The nn-level range that names this paper's operating window. Scope, relevant to the cubic budget center (Section sec:flags ): their unconditional nn-level theorem is for smooth test functions; taking a characteristic function requires RH. Section sec:sine 's determinantal ρ3\rho_3 and the c3(λ′)c_3(\lambda') identification invoke the smooth statement.DOI
  8. J. B. Conrey, A. Ghosh and S. M. Gonek, Simple zeros of the Riemann zeta-function , Proc. London Math. Soc. (3) 76 (1998), 497--522; doi:10.1112/S0024611598000306. The origin of the second integrality level that Lemma lem:cornerineq and Theorem thm:corner consume, made unconditional in Alpöge and Furman's 7.2(c).DOI
  9. H. M. Bui and D. R. Heath-Brown, On simple zeros of the Riemann zeta-function , Bull. London Math. Soc. 45 (2013), no. 5, 953--961; doi:10.1112/blms/bdt026; arXiv:1302.5018. The conditional simple-zero input quoted in Section 1.2; their unconditional-on-RH distinct-zeros constant, as published, is 0.846650.84665.arXivDOI
  10. A. Chirre, F. Gonçalves and D. de Laat, Pair correlation estimates for the zeros of the zeta function via semidefinite programming , Adv. Math. 361 (2020), 106926; doi:10.1016/j.aim.2019.106926; arXiv:1810.08843v2. The semidefinite-programming line for pair-correlation extremal problems, of which this paper's LP is the bandwidth-one marked-configuration analog. Two specific contacts. (i) Their inequality (7), 2Ns(T)(m−2)(m−3)/m=N∗(T)−5N(T)+6(T)2 N_s(T) _ (m_ - 2)(m_ - 3)/m_ = N^ * (T) - 5N(T) + 6 (T), credited there to an argument of Ghosh, is the published instance of the same per-multiplicity integrality device Lemma lem:cornerineq uses, at the reciprocal-weighted third level (m−2)(m−3)≥0(m-2)(m-3) \ge 0; Lemma lem:cornerineq (a) is the unweighted second level (m−1)(m−2)≥0(m-1)(m-2) \ge 0. (Their semidefinite programming is applied to the auxiliary-function class of their Section 3, not to (7); (7) is elementary integer arithmetic.) (ii) Their Corollary 3 is the corresponding published record for distinct zeros: $\Nd(T) \ge (0.8477 + o(1))N(T)$ on RH and (0.8486+o(1))N(T)(0.8486 + o(1))N(T) on GRH.arXivDOI
Show all 16 referencesShow fewer references
  1. F. Gonçalves, D. de Laat and N. Leijenhorst, Multiplicity of nontrivial zeros of primitive LL-functions via higher-level correlations , Math. Comp. 94 (2025), no. 354, 2041--2058; doi:10.1090/mcom/4005; arXiv:2303.01095. The closest published third-party work to this paper: the same higher-level-correlation-versus-multiplicity question, attacked by semidefinite programming rather than by a marked-configuration LP. Cited in Sections 1.2 and 3.2. Their equation (5), Nn(T)=∑kS(n,k)Mk(T)N_n(T) = \sum_k S(n,k) M_k(T) with S(n,k)S(n,k) the Stirling numbers of the second kind, is the change of basis whose n=3n = 3 case is the atom half of Theorem thm:blind here.arXivDOI
  2. D. W. Farmer, S. M. Gonek and Y. Lee, Pair correlation of the zeros of the derivative of the Riemann ξ\xi-function , J. London Math. Soc. (2) 90 (2014), no. 1, 241--269; doi:10.1112/jlms/jdu026. Cited only in Corollary cor:matched , to disambiguate a numerical coincidence: their published RH-conditional lower bound for the proportion of distinct zeros is $\Nd(T) > 0.8051\,N(T)$ (their Corollary 1.4). The two-author 2008 preprint arXiv:0803.0425 is an earlier version whose Corollary 1.4 gives 0.65440.6544; the constant 0.80510.8051 is the published version's.arXivDOI
  3. T. Kuna, J. L. Lebowitz and E. R. Speer, Realizability of point processes , J. Stat. Phys. 129 (2007), no. 3, 417--439; doi:10.1007/s10955-007-9393-y; arXiv:math-ph/0612075. The truncated-moment/realizability problem that Section sec:adversary 's adversary-realizability argument re-derives in the finite-circle setting.arXivDOI
  4. T. Kuna, J. L. Lebowitz and E. R. Speer, Necessary and sufficient conditions for realizability of point processes , Ann. Appl. Probab. 21 (2011), no. 4, 1253--1281; doi:10.1214/10-AAP703; arXiv:0910.1710. The sequel, which settles the general finite-order truncated-moment realizability question; Section sec:adversary 's finite-circle argument is a special case in their setting rather than a new problem.arXivDOI
  5. H. Cohn, D. de Laat and A. Salmon, Three-point bounds for sphere packing , arXiv:2206.15373 (v1, 30 Jun 2022; unpublished as of August 2026). The comparandum: in the sphere-packing analog the third-order augmentation does pay, improving on the two-point Cohn--Elkies bound in several dimensions; here it is worth exactly nothing.arXiv
  6. S. Karlin and W. J. Studden, Tchebycheff Systems: With Applications in Analysis and Statistics , Pure and Applied Mathematics, Vol. XV, Interscience Publishers (John Wiley & Sons), New York, 1966. The classical Chebyshev--Markov--Stieltjes/Christoffel background for the fact that odd moments do not improve one-sided bounds --- the general form of the cubic blindness of Theorem thm:blind .

Version history

  1. v1 · 41 pagesSubmitted by Kunal TyagiInitial depositSupersededOct 15, 2026
  2. v2 · 8 pagesSubmitted by Kunal TyagiSubmitted as a new versionCurrentOct 18, 2026