The prime number theorem in short intervals for automorphic \(L\)-functions (Q2889254)

Let \(\psi(x)=\sum_{n\leq x}\Lambda(n)\) where \(\Lambda(n)\) is the von Mangoldt function. An important problem is about the fluctuation of \(\psi(x)\) over short intervals, which is closely connected to the distribution of the zeros of the Riemann zeta-function \(\zeta(s)\). Conditionally on the Riemann hypothesis, Selberg showed that NEWLINE\[NEWLINE \int_1^X |\psi(x+h(x))-\psi(x)-h(x)|^2\,dx =o(Xh(X)^2) NEWLINE\]NEWLINE as \(X\to \infty\), where \(h(x)\leq x\) is an increasing function satisfying \(h(x)/(\log x)^2\to \infty\) as \(x\to \infty\). This result could be refined under further assumption on the pair correlation of the zeros of \(\zeta(s)\).NEWLINENEWLINEIn the paper under review, the authors extend the above investigation to the automorphic \(L\)-function \(L(s,\pi)\) for an irreducible unitary cuspidal representation \(\pi\) of \(\mathrm{GL}_m(\mathbb{A}_\mathbb{Q})\). They consider the function NEWLINE\[NEWLINE \psi(x,\pi):=\sum_{n\leq x} \Lambda_\pi(n) NEWLINE\]NEWLINE where \(\Lambda_\pi(n)\) is the \(n\)th coefficient of the Dirichlet series NEWLINE\[NEWLINE -\frac{L'}{L}(s,\pi)=\sum_{n=1}^\infty \Lambda(n)n^{-s}. NEWLINE\]NEWLINE The sequence \(\{\Lambda_\pi(n)\}\) is supported on prime powers. Each \(\Lambda_\pi(n)\) can be expressed in terms of Satake parameters but, unlike the classical case, \(\Lambda_\pi(n)\) may be non-positive. The authors prove two theorems: (1) under the Grand Riemann hypothesis (GRH) for \(L(s,\pi)\), NEWLINE\[NEWLINE \int_1^X |\psi(x+h(x),\pi)-\psi(x,\pi)|^2\,dx \ll Xh(X)(\log(Q_\pi X))^2 +\bigg(\frac{\log Q_\pi}{\log X}\bigg)^4 NEWLINE\]NEWLINE where \(Q_\pi\) is the conductor of \(\pi\) and \(h(x)\leq x\) is an increasing function. (2) Under GRH and an assumption of the pair correlation of zeros of \(L(s,\pi)\), the first summand \(Xh(X)(\log(Q_\pi X))^2\) is replaced by NEWLINE\[NEWLINE Xh(X)\log(Q_\pi X) + X(\log(Q_\pi X))^2. NEWLINE\]