Rankin-Selberg local factors modulo \(\ell \) (Q508442): Difference between revisions

In this paper, the authors study Rankin-Selberg local factors for \(l\)-adic and \(l\)-modular representations of the general linear group \(\operatorname{GL}_n(F)\) of Whittaker type. Here, \(F\) is a non-Archimedean local field of residual characteristic \(p\) and \(p\neq l.\) More precisely, these local factors were already defined by other authors; the emphasis here lies on the examination of the behaviour of these factors with respect to the reduction modulo \(l,\) starting from \(l\)-adic to \(l\)-modular. Here \(l\)-adic refers to smooth representations over \(\overline{\mathbb{Q}_l}\) and \(l\)-modular refers to smooth representations over \(\overline{\mathbb{F}_l},\) where \(\mathbb{F}_l\) is a finite field with \(l\) elements. The main results are the following. The authors completely describe \(l\)-modular \(L\)-factors in the generic case; they obtain that those \(L\)-factors attached to these generic representations are the same as those attached to the banal parts of these representations. As for the behaviour under the reduction modulo \(l\): it turns out that \(l\)-modular local \(\gamma\)-factor associated to a pair of Whittaker type \(l\)-modular representations coincides with the reduction modulo \(l\) of \(l\)-adic \(\gamma\)-factor of any Whittaker (\(l\)-adic) lifts of this pair. The local \(L\)-factor of this pair behaves more complex: this factor always divides the reduction modulo \(l\) of the \(l\)-adic \(L\)-factor of any of the Whittaker lifts of this pair, but the strict division may occur. Moreover, they prove that this \(l\)-modular \(L\)-factor is the greatest common divisor of the reductions mod \(l\) over all integral generic \(l\)-adic representations which contain this initial \(l\)-modular pair as subquotients after reduction modulo \(l.\) The paper is very informative; throughout the first two sections, it gives an overview of the relevant notions in the theory of \(l\)-adic and \(l\)-modular representations which are distinct from the corresponding notions in the case of representations over complex numbers and which are be encountered later in the paper. Let us mention some of the topics: Haar measures, Bernstein-Zelevinsky theory of derivatives, parabolic induction and restriction, integral structures and reduction modulo \(l.\) In the second section an overview of the representation of Whittaker type representations is given, as well as an overview of generic representations and liftings of supercuspidal representations. Then, in the third and the fourth section, based on the work of Jacquet, Piatetski-Shapiro and Shalika [\textit{H. Jacquet} et al., Ann. Math. (2) 109, 169--212, 213--258 (1979; Zbl 0401.10037); Math. Ann. 271, 319--332 (1985; Zbl 0541.12010)] in the complex case, the authors prove their main results. Of course, many adjustments are needed in this procedure, and temporary limitations in the theory force the authors to consider just the representations of the Whittaker type.