arXiv Science⌕ Search

arXiv · 2610.15559

The Goldbach exceptional set: an exponent 7/12 and conditional refinement

Abstract

Let \(E(X)\) count the even integers between \(4\) and \(X\) that are not sums of two primes. We combine Pintz's explicit formula for the major arcs with the exponential-sum estimate of Maynard, Pandey, and Radziwiłł to prove \(E(X)\ll_\varepsilon X^{7/12+\varepsilon}\) whenever all Dirichlet \(L\)-functions have a common fixed zero-free half-plane \(\Rea s>θ_0\), with \(θ_0<1\). The uniform \(7/8\) theorem reported in the OpenAI mathematics release supplies this input. Under the stronger hypothesis that every nontrivial Dirichlet zero has real part at most \(θ\), where \(1/2\leqθ\leq2/3\), we prove \[ E(X)\ll_{θ,\varepsilon}X^{g(θ)+\varepsilon}, \qquad g(θ)=\frac{θ(5-4θ)}{2(2-θ)}. \] Thus every fixed \(θ<2/3\) gives an exponent strictly below \(7/12\). The latter argument uses a classical fixed-modulus zero-density estimate and is independent of the new exponential-sum theorem.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

János Pintz, Ingo Althöfer. 2026-10-11. The Goldbach exceptional set: an exponent 7/12 and conditional refinement. https://arxiv.org/abs/2610.15559

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

Sporadic points on $X_0(N)$

We determine all integers $N$ for which the modular curve $X_0(N)$ admits a sporadic CM point (of any degree), as well as all $N$ for which $X_0(N)$ admits a sporadic point, whether CM or non-CM. In a sense, our results generalize the classification of isogenies of elliptic curves over $\Q$ due to Mazur and Kenku: their work determines the $X_0(N)$ with degree 1 sporadic points, whereas we classify all $X_0(N)$ that have a sporadic point of arbitrary degree.

math.NT↗

On the Number of Prime Factors of Consecutive Integers

We prove that there are infinitely many $n$ such that $ω(n+k) \ll \log k$ for all integers $k \ge 2$. This improves on a result of Tao and Teräväinen (2025), who have $O(k)$ in place of $O(\log k)$. As corollaries, we make progress on a number of questions posed by Erdős. The proof is based on a quantitative refinement of the Tao-Teräväinen probabilistic argument, combining a refined sieve construction with upper bounds for high moments and the resulting tail estimates. Moreover, we formulate a conjecture on integers with many prime factors based on Cramér-type random models. Assuming this conjecture, the main bound is essentially sharp.

math.NT↗