Answers are grounded in this paper and cite their sources.
This working paper develops a unified study of zero localization, spectral counting inequalities, and arithmetic positivity for the de Bruijn–Newman deformation of the Riemann xi function. It connects the motion of zeros under the backward heat equation with finite-dimensional spectral inequalities, local statistics of zeta zeros, and the Weil positivity criterion. The results include unconditional localization and counting theorems, complete numerical certificates for a finite-support positivity bound, and explicit arithmetic conditions for stronger zero counts and global positivity. The paper is organized into three parts. The first treats positive-time heat flow and the localization of possible collisions. The second studies zero-time counting through Hermitian operators, geometric inequalities, higher traces, and bounded local-count transforms. The third develops finite Schur certificates for the Weil form and derives exact growth laws for Gaussian spectral quantities and compensated prime covariance. Throughout, multiplicities and zeros away from the critical line are retained where the arithmetic formulas require them. Heat-flow localization and an explicit upper bound A uniform shrinking-time theorem proves that sufficiently high zeros are real and simple above an explicit logarithmic transition threshold. The threshold is 4911678521 divided by 1933561194, approximately 2.54022398476. For any fixed larger constant, the theorem applies at positive heat times up to one fifth whenever the product of heat time and the logarithm of the ratio of the absolute real coordinate to four times pi exceeds that constant. The starting height is asymptotic. The proof combines a five-pair exponential-sum partition, complex approximation, and Rouché's theorem. An all-orders arithmetic expansion also gives phase and zero-location information in the transition region. A separate exterior-attraction argument proves that the de Bruijn–Newman constant is less than 0.17997. It combines contraction of the nonreal zero strip, a positive resolvent measuring attraction from exterior zeros, prime-power cancellation, and a certified zero-free vertical boundary. Directed-rounding estimates and a scalar comparison control the passage from a finite initial strip to extinction of nonreal zeros. These results constrain possible positive-time collisions without asserting their exclusion at every positive time. Unconditional simple and distinct zero proportions At zero time, a Hermitian operator associated with a finite conjugation-invariant zero multiset converts spectral information into lower bounds for simple critical-line zeros and distinct zeros. The paper proves finite counting inequalities, averages them over consecutive blocks, and combines the resulting geometric correction with an unconditional arithmetic trace formula. Position-dependent seven-point weights retain the contribution of each pair, two gap-pressure terms, and the boundary costs of overlapping windows. Complete interval certificates and a 387-point block conversion yield limiting lower proportions of at least 67.3320065557 percent for simple critical-line zeros and 83.6660032778 percent for distinct nontrivial zeros. Both proportions are measured relative to all nontrivial zeros in the height range, counted with multiplicity. The argument builds on the trace and spectral methods of Lamzouri and Wang and on seven-point certification methods. A universal geometric ceiling explains the need for further arithmetic information. Within the specified support-one quadratic-trace method, corrections valid for every simple real configuration cannot force a simple-zero proportion above approximately 67.6026135631 percent. This limitation motivates the higher-trace and local-count criteria that follow. Analytic height transfer and a fourth-order counting criterion An unconditional transfer theorem connects analytically smoothed cyclic traces to the ordinary unweighted zero count. A bounded rational inertia certificate makes this transfer possible through Hilbert–Schmidt control of the cutoff error. The proof includes macroscopic height covariance, estimates for the complete height tail, and preservation of the Hermitian structure contributed by nonreal conjugate pairs. The first three smoothed cyclic moments are evaluated unconditionally. The fourth is expressed as a pairing contribution plus a signed residual, with exact Fourier and finite-prime representations. A strict upper bound of 182 divided by 4005, approximately 0.0454431960, for that residual in the stated ordered limits would imply a simple critical-line proportion greater than 67.8 percent. The transfer and lower-moment evaluations are proved; the required signed fourth-order arithmetic estimate remains open. Bounded local-count transforms A second counting criterion uses all-zero counts in adjacent microscopic intervals. An interval containing exactly one zero occurrence identifies a simple critical-line zero, because multiplicity or a reflected off-line partner would contribute additional occurrences. A bounded detector extracts useful pair information, and a weighted matching argument converts it into an improvement of the simple-zero proportion. Four values of the local-count generating function express the detector's average. Exact contour-winding formulas connect these counts to the zeta function, while logarithmic generating functions organize the connected coefficients. Under the stated analytic continuation, all-order coefficient envelope, and convergence hypotheses, the shared coefficients permit a signed enclosure of the complete omitted tail. The first six reference connected coefficients at each of the two interval lengths suffice for a positive margin, with a quantitative tolerance for errors in the retained logarithmic values. The reference trace calculations use interval arithmetic and an explicit Taylor-kernel remainder. A separate criterion through second and third factorial moments is also proved. Neither route supplies the required actual-zero arithmetic estimates unconditionally. Complete finite-support Weil positivity The Weil-form analysis separates a finite-dimensional head from an infinite-dimensional complement through a Legendre representation and an exact Schur reduction. Phase-dependent prime estimates, physical Gram normalization, weighted inverse-tail bounds, and complete residual estimates prove a coercive lower bound of ten to the power minus twenty-seven times the squared norm for support parameters between zero and three fortieths. This is a bound on the full form domain. The certificate includes the infinite-dimensional contribution, with explicit analytic remainders and interval enclosures for the numerical inputs. It provides a rigorous finite-support result, while the extension to unbounded supports requires an additional arithmetic estimate. Gaussian growth and a signed prime criterion for RH At one fixed Gaussian width, the paper proves an exact asymptotic growth law for the negative part of the least eigenvalue of the normalized Weil matrices. The logarithmic growth at the square-root degree scale is determined by the maximal displacement of a zero from the critical line. Consequently, a negative allowance whose logarithm grows more slowly than the square root of the degree, even along an arbitrary unbounded sequence of degrees, would imply RH. A polynomial allowance is sufficient, and no restriction on the gaps between the tested degrees is required. A compensated thermal covariance gives a corresponding scalar formulation. Its exponential growth rate is determined by the square of the maximal zero displacement. The reflected contribution is bounded, and the leading prime diagonal has an explicit polynomial asymptotic. A dyadic combination cancels that leading diagonal term and leaves a signed expression containing the off-diagonal prime interactions and continuum corrections. A polynomial upper bound for this expression along a sequence of scales tending to zero is equivalent to RH. Scope and accompanying materials The paper does not prove the Riemann hypothesis or an unconditional 67.8 percent simple critical-line proportion. Its unconditional results comprise the stated heat-flow localization, explicit upper bound, zero-counting proportions, analytic height transfer, finite-support Weil certificate, and structural identities and growth laws. The stronger conclusions depend on precisely stated arithmetic estimates. The accompanying materials contain the 79-page manuscript, editable LaTeX sources, and the numerical certificate sources and inputs required by the retained results. Interval-arithmetic receipts, exact rational checks, analytic remainder bounds, dependency information, and file hashes support inspection and reproduction of the finite computations. The work brings together analytic number theory, entire-function heat flow, spectral inequalities, explicit formulas, and computer-assisted proof to distinguish finite certification from the uniform arithmetic information needed for global conclusions.
No takes yet. Share an insight, caveat, or question.
K. Fathi (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: