Beyond the Weyl barrier for \(\mathrm{GL}(2)\) exponential sums (Q6103490)

For \(g\), a holomorphic cusp newform of level \(D\), weight \(k\), nebentypus character \(\xi\), let \(\lambda_g(n)\) be the coefficient of Fourier expansion of \(g(z)\) given by \(g(z)=\sum_{n=1}^\infty\lambda_g(n)n^{(k-1)/2}e(nz)\), with \(z\) belonging to the upper half plane, and \(e(z)=e^{2\pi iz}\). In the paper under review, the authors estimate several variants of the following smoothed exponential sum \[ S_f(N)=\sum_{n=1}^\infty\lambda_g(n)e(f(n))V\left(\frac{n}{N}\right), \] where the weight function \(V\) belongs to the set \(C_c^\infty(0,\infty)\) and the phase function \(f\) is of the form \(f(x)=N^\beta\phi(x/N)\) with \(\beta>1+\varepsilon\) for an arbitrarily small \(\varepsilon>0\), and \(\phi(x)=\frac{1}{2\pi}\log x\) or \(\phi(x)=ax^\beta\) for a fixed real number \(a\neq 0\). The idea of estimating is to use the two-dimensional stationary phase method to transform the off-diagonal sum in the Bessel \(\delta\)-method to certain double exponential sums, and then develop two new van der Corput methods of exponent pairs to treat this type of sums with `almost separable' phase.