Let k and n be positive even integers. For a cuspidal Hecke eigenform h in the Kohnen plus space of weight k - n/2 + 1/2 for Γ₀(4), let f be the corresponding primitive form of weight 2k-n for SL₂(ℤ) under the Shimura correspondence, and Iₙ(h) the Duke-Imamoḡlu-Ikeda lift of h to the space of cusp forms of weight k for Spₙ(ℤ). Moreover, let ${\varphi }_{Iₙ\left(h\right),1}$ be the first Fourier-Jacobi coefficient of Iₙ(h), and ${\sigma }_{n-1}\left({\varphi }_{Iₙ\left(h\right),1}\right)$ be the cusp form in the generalized Kohnen plus space of weight k - 1/2 corresponding to...

We interpolate the Gauss–Manin connection in $p$-adic families of nearly overconvergent modular forms. This gives a family of Maass–Shimura type differential operators from the space of nearly overconvergent modular forms of type $r$ to the space of nearly overconvergent modular forms of type $r+1$ with $p$-adic weight shifted by $2$. Our construction is purely geometric, using Andreatta–Iovita–Stevens and Pilloni’s geometric construction of eigencurves, and should thus generalize to higher rank...

Let $F$ be an imaginary quadratic field and $𝒪$ its ring of integers. Let $𝔫\subset 𝒪$ be a non-zero ideal and let $p\>5$ be a rational inert prime in $F$ and coprime with $𝔫$. Let $V$ be an irreducible finite dimensional representation of ${\overline{𝔽}}_p\left[{\mathrm{GL}}_2\left({𝔽}_{p^2}\right)\right]$. We establish that a system of Hecke eigenvalues appearing in the cohomology with coefficients in $V$ already lives in the cohomology with coefficients in ${\overline{𝔽}}_p\otimes de{t}^e$ for some $e\ge 0$; except possibly in some few cases.

We prove the indecomposability of the Galois representation restricted to the $p$-decomposition group attached to a non CM nearly $p$-ordinary weight two Hilbert modular form over a totally real field $F$ under the assumption that either the degree of $F$ over $\mathbb{Q}$ is odd or the automorphic representation attached to the Hilbert modular form is square integrable at some finite place of $F$.

For a prime p > 2, an integer a with gcd(a,p) = 1 and real 1 ≤ X,Y < p, we consider the set of points on the modular hyperbola ${}_{a,p}(X,Y)=(x,y):xy\equiv a\left(modp\right),1\le x\le X,1\le y\le Y$. We give asymptotic formulas for the average values ${\sum }_{\begin{array}{c}(\end{array}x,y){\in }_{a,p}(X,Y)x\ne y*}\phi \left(\right|x-y\left|\right)/|x-y|$ and ${\sum }_{\begin{array}{c}(\end{array}x,y){\in }_{a,p}(X,X)x\ne y*}\phi \left(\right|x-y\left|\right)$ with the Euler function φ(k) on the differences between the components of points of ${}_{a,p}(X,Y)$.

Suppose that $f$ is a primitive Hecke eigenform or a Mass cusp form for ${\Gamma }_0\left(N\right)$ with normalized eigenvalues ${\lambda }_f\left(n\right)$ and let $X>1$ be a real number. We consider the sum $${𝒮}_k\left(X\right):=\sum _{n<X}\sum _{n=n_1,n_2,...,n_k}{\lambda }_f\left(n_1\right){\lambda }_f\left(n_2\right)...{\lambda }_f\left(n_k\right)$$ and show that ${𝒮}_k\left(X\right){\ll }_{f,ϵ}{X}^{1-3/\left(2\right(k+3\left)\right)+ϵ}$ for every $k\ge 1$ and $ϵ>0$. The same problem was considered for the case $N=1$, that is for the full modular group in Lü (2012) and Kanemitsu et al. (2002). We consider the problem in a more general setting and obtain bounds which are better than those obtained by the classical result of Landau (1915) for $k\ge 5$. Since the result is valid...

Let X, Y be two separable F-spaces. Let (Ω,Σ,μ) be a measure space with μ complete, non-atomic and σ-finite. Let $N_F$ be the Nemytskiĭ set-valued operator induced by a sup-measurable set-valued function $F:\Omega \times X\to 2^Y$. It is shown that if $N_F$ maps a modular space $(N\left(L(\Omega ,\Sigma ,\mu ;X)\right),{\varrho }_{N,\mu })$ into subsets of a modular space $(M\left(L(\Omega ,\Sigma ,\mu ;Y)\right),{\varrho }_{M,\mu })$, then $N_F$ is automatically modular bounded, i.e. for each set K ⊂ N(L(Ω,Σ,μ;X)) such that $r_K=sup{\varrho }_{N,\mu }\left(x\right):x\in K<\infty$ we have $sup{\varrho }_{M,\mu }\left(y\right):y\in N_F\left(K\right)<\infty$.

Let $f$ be a Hecke–Maass cusp form of eigenvalue $\lambda$ and square-free level $N$. Normalize the hyperbolic measure such that $\mathrm{vol}\left(Y_0\left(N\right)\right)=1$ and the form $f$ such that ${\parallel f\parallel }_2=1$. It is shown that ${\parallel f\parallel }_{\infty }{\ll }_{ϵ}{\lambda }^{\frac{5}{24}+ϵ}{N}^{\frac{1}{3}+ϵ}$ for all $ϵ>0$. This generalizes simultaneously the current best bounds in the eigenvalue and level aspects.

Hecke groups $H\left({\lambda }_q\right)$ are the discrete subgroups of $\mathrm{P}SL(2,ℝ)$ generated by $S\left(z\right)=-{(z+{\lambda }_q)}^{-1}$ and $T\left(z\right)=-\frac{1}{z}$. The commutator subgroup of $H$(${\lambda }_q)$, denoted by $H^{\text{'}}\left({\lambda }_q\right)$, is studied in [2]. It was shown that $H^{\text{'}}\left({\lambda }_q\right)$ is a free group of rank $q-1$. Here the extended Hecke groups $\overline{H}\left({\lambda }_q\right)$, obtained by adjoining $R_1\left(z\right)=1/\overline{z}$ to the generators of $H\left({\lambda }_q\right)$, are considered. The commutator subgroup of $\overline{H}\left({\lambda }_q\right)$ is shown to be a free product of two finite cyclic groups. Also it is interesting to note that while in the $H\left({\lambda }_q\right)$ case, the index of $H^{\text{'}}\left({\lambda }_q\right)$ is changed by $q$, in the case of $\overline{H}\left({\lambda }_q\right)$, this number is...

We characterize stability under composition of ultradifferentiable classes defined by weight sequences M, by weight functions ω, and, more generally, by weight matrices , and investigate continuity of composition (g,f) ↦ f ∘ g. In addition, we represent the Beurling space ${}^{\left(\omega \right)}$ and the Roumieu space ${}^{\omega }$ as intersection and union of spaces ${}^{\left(M\right)}$ and ${}^M$ for associated weight sequences, respectively.

Let ${\lambda }_f\left(n\right)$ be the nth normalized Fourier coefficient of a holomorphic Hecke eigenform $f\left(z\right)\in S_k\left(\Gamma \right)$. We establish that ${\sum }_{n\le x}{\lambda }_f^2\left(n^j\right)=c_{j}x+O\left(x^{1-2/({(j+1)}^2+1)}\right)$ for j = 2,3,4, which improves the previous results. For j = 2, we even establish a better result.