Second Moment of Central Values of Half-Integral Weight Modular Forms and Subconvexity
with Henry Twiss, Zhining Wei, and Peter Zenz
Abstract
We let $f$ be a half-integral weight modular form of weight $\kappa > 4$ on $\Gamma_0(4)$ that is an eigenfunction
of all Hecke operators $T_n$, so that $T_nf = \Lambda_f(n)n^{\frac{\kappa-1}{2}}f$. Let $\|f\|$ denote the Petersson
norm of $f$. We study a weighted second moment of the central value of the L-function associated to $f$ over an
orthogonal basis $H_\kappa(4)$ of $S_\kappa(\Gamma_0(4))$. This corresponds to studying the following sum:
$$\sum_{f \in H_\kappa(4)} \frac{\Lambda_f(n) |L(1/2, f)|^2}{\|f\|^2}$$
Using the relative trace formula, we obtain an asymptotic formula for the second moment. We then use the
method of amplification to get the subconvexity bound:
$$L(1/2, f) \ll_\epsilon (\kappa^2)^{\frac{1}{4} - \frac{1}{40} + \epsilon}$$
This is the first subconvexity result for half-integral weight modular forms in the weight aspect.
We also apply our second moment result to get a quantitative simultaneous non-vanishing result for
central values of L-functions.
arXiv:2512.20483
Mod p Iwasawa Algebras of pro-p Iwahori Subgroups
with Rudy Ariaz, Bryan Hu, Simran Khunger, Karol Koziol, Bharatha Rankothge, and Bobby Zixuan Zhang
Abstract
Suppose $F$ is a finite unramified extension of $\mathbb{Q}_p$, and $G$ is the group of $F$-points of a split,
connected, reductive group over $F$. Under a natural restriction on $p$, we determine the structure of the
graded mod $p$ Iwasawa algebra $\operatorname{gr}_{\mathfrak{m}}(\mathbb{F}_p[[I]])$, where $I$ is a pro-$p$ Iwahori subgroup of $G$. We also determine its
maximal commutative quotient, and relate these results to Gelfand–Kirillov dimensions of smooth mod $p$ representations of $G$.
arXiv:2601.09021
Explicit Zero-Free Regions for Automorphic L-Functions
with Alia Hamieh, Simran Khunger, Kaneenika Sinha, Jakob Streipel, and Kin Ming Tsang
Abstract
Let $L(s, f)$ be the L-function associated with a newform $f$ of even weight $k$, squarefree level $N$ and
trivial nebentypus. In this paper, we establish a new explicit zero-free region for $L(s, f)$.
More precisely, we prove that $L(s, f)$ does not vanish in the region
$$\operatorname{Re}(s) \ge 1 - \frac{1}{C \log\big( kN \cdot \max(1, |\operatorname{Im}(s)|) \big)}$$
with $C = 16.7053$ if $|\operatorname{Im}(s)| \ge 1$ or $|\operatorname{Im}(s)| \le 0.30992 / \log(kN)$, and
$C = 16.9309$ if $0.30992 / \log(kN) < |\operatorname{Im}(s)| \le 1$.
This improves a result of Hoey et al. where $445.994$ was shown to be an admissible value for $C$.
arXiv:2509.20873
Cancellation in Sums of Hecke Eigenvalues Over Quadratic Polynomials and Mass Equidistribution
Abstract
We study cancellation in sums of Hecke eigenvalues over irreducible quadratic polynomials
over short intervals. In particular, we look at an average over bases of Hecke forms of weight $k$ in
the range $|k-K| < K^{\theta}$ where $1/3 < \theta < 1$. We see that when averaged over this family such sums admit square
root cancellation. The key new arithmetic input for such a result is a bound on sums of Kloosterman sums
over irreducible quadratic polynomials. Then using work of Nelson, we relate such sums to the mass
equidistribution conjecture for modular forms on compact arithmetic surfaces, and we show that almost all
forms satisfy the mass equidistribution conjecture. Furthermore, such forms will satisfy the conjecture
with an effective convergence rate of $k^{-\delta}$ for any $\delta < 1/2$.
arXiv:2508.18666
A Formula for the Number of k-almost Primes
Abstract
In 2005, E. Noel and G. Panos constructed a formula to count the number
of semiprimes less than a given value $x$. In 2006, this formula was rediscovered
independently by R. G. Wilson V. However, each citation of this result was simply
through personal communication, and a formal proof has not been written down
anywhere in the literature. In this article, we generalize their formula to the
case of $k$-almost primes and square-free $k$-almost primes.
arXiv:2310.14989
A Converse Theorem in Half-Integral Weight
with Henry Twiss
Submitted to the Journal of Number Theory
Abstract
In this paper, we prove a converse theorem for half-integral
weight modular forms assuming functional equations for L-series with additive
twists. This result is an extension of Booker, Farmer, and Lee's result in [BFL22]
to the half-integral weight setting. Similar to their work, the main result of this
paper is obtained as a consequence of the half-integral weight Petersson trace formula.
arXiv:2306.02872
Limits and Colimits in the Category of Pastures
Abstract
We show that the category of pastures has arbitrary limits and colimits
of diagrams indexed by a small category.
arXiv:2103.08655
Prym-Brill-Noether Loci of Special Curves
with Yoav Len, Caelan Ritter, and Derek Wu
International Mathematics Research Notices, Volume 2022, Issue 4, February 2022
Abstract
We use Young tableaux to compute the dimension of $V^r$, the Prym-Brill-Noether locus
of a folded chain of loops of any gonality. This tropical result yields a new upper bound
on the dimensions of algebraic Prym-Brill-Noether loci. Moreover, we prove that $V^r$ is pure-dimensional
and connected in codimension 1 when $\dim V^r \ge 1$. We then compute the first Betti number of this locus for
even gonality when the dimension is exactly 1, and compute the cardinality when the locus is finite and
the edge lengths are generic.
Journal version (IMRN)
arXiv:1912.02863
Extensions of Hyperfields
Abstract
We develop a theory of extensions of hyperfields that generalizes the notion of field extensions. Since
hyperfields have a multivalued addition, we must consider two kinds of extensions that we call weak hyperfield
extensions and strong hyperfield extensions. For quotient hyperfields, we develop a method to construct strong
hyperfield extensions that contain roots to any polynomial over the hyperfield. Furthermore, we give an example
of a hyperfield that has two non-isomorphic minimal extensions containing a root to some polynomial. This shows
that the process of adjoining a root to a hyperfield is not a well-defined operation.
arXiv:1912.05919
Video series
A series of talks I gave about this paper: