Searcharxiv⌕ Search

arXiv · 2609.39904

On a question of Jones on tracelike vectors, cusp forms, and von Neumann algebras

Abstract

Previous work of Vaughan Jones built a bridge between cusp forms and random matrices using the language of von Neumann algebras. Specifically, for a discrete subgroup of $\operatorname{PSL}_2(\mathbb Z)$, there is a natural von Neumann algebra associated to it. Jones showed that for any "tracelike'' vector, there is a corresponding (anti)-isomorphism between this algebra and its commutant. Jones also showed that such vectors exist. As he noted, this connection is not useful without an explicit formula, which he posed as an open question. Here, we resolve Jones' question and discuss its context for a broad number theory audience. As we were finalizing this preprint for submission, we learned of an independent paper of Abreu. Both papers employ a closely-related polar normalisation of a kernel orbit to obtain the Jones tracelike vector at the critical parameter. The novelty of the present work lies in its extension to all $1<s\leq13$ and in the detailed analysis of the resulting von Neumann algebra and cusp form correspondences. This paper also arose out of a near solution of the authors from several years ago, which discussions with ChatGPT helped us finalize into a full solution. The relation to Abreu's work, and the exact nature of our use of AI, are detailed in the introduction.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Nikolaos Diamantis, Larry Rolen. 2026-09-30. On a question of Jones on tracelike vectors, cusp forms, and von Neumann algebras. https://arxiv.org/abs/2609.39904

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

KEEP EXPLORING

Related papers

An explicit type number formula for quaternion orders of level $(N_1,N_2)$

We give an explicit type number formula for the quaternion orders of level $(N_1,N_2)$ specified in this paper, where $N_1=p_1^{2u_1+1}\cdots p_w^{2u_w+1}$, the primes $p_i$ are distinct, $u_i\ge0$, $w$ is odd, and $\gcd(N_1,N_2)=1$. The formula includes Eichler orders and the orders with odd prime-power ramified level considered by Boyd, and allows nonmaximal local orders at several ramified primes, including $2$. We express the answer in terms of a generalized modified Hurwitz class number and explicit local correction factors. The proof combines the correspondence between quaternion orders and ternary quadratic forms with the Siegel--Weil formula and local representation densities. We tabulate class and type numbers for $N_1N_2\le100$. As an application, a mass bound and a finite exact calculation determine the $27$ pairs in this family with type number one.

math.NT↗

Nombres de Pisot, nombres de Salem et la conjecture de Lehmer

We investigate the relationship between the set S of Pisot numbers and the set T of Salem numbers. Salem first established that: " every Pisot number is an accumulation point of the set T ". Building on Boyd's method, we show that every accumulation point of T belongs to S. Together, these results imply that the union S U T forms a closed subset of the real half-line ]1,+infinity[. Consequently, this settles Boyd's conjecture while disproving Lehmer's conjecture.

math.NT↗

Multiplicative Subgroups of $\mathbb{Z}_p^*$ that are Generalized Arithmetic Progressions

We prove that a multiplicative subgroup $A_k$ of $\mathbb{Z}_p^*$ is a generalized arithmetic progression if and only if $|A_k| = 2,\ 4,$ or $p-1$. Much of the argument builds upon recent work studying additive decompositions of subgroups, and we generalize a result of Hanson and Petridis to show that any additive $n$-decomposition of a subgroup must be a direct sum. We also show how this classification quickly follows from Kalmynin's recent work resolving Sárközy's conjecture for quadratic residues.

math.NT↗