A vanishing Ext-branching theorem for \((\mathrm{GL}_{n+1}(F),\mathrm{GL}_n(F)\) (Q821478)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | A vanishing Ext-branching theorem for \((\mathrm{GL}_{n+1}(F),\mathrm{GL}_n(F)\) |
scientific article |
Statements
A vanishing Ext-branching theorem for \((\mathrm{GL}_{n+1}(F),\mathrm{GL}_n(F)\) (English)
0 references
20 September 2021
0 references
Let \(F\) be a \(p\)-adic field. It is an important problem to understand the restriction of a smooth representation of \(\mathrm{GL}_{n+1}(F)\) to the subgroup \(\mathrm{GL}_n(F)\), which is known as part of the local Gan-Gross-Prasad conjecture [\textit{W. T. Gan} et al., Astérisque 346, 1--109 (2012; Zbl 1280.22019)]. Let \(\pi_1\) be an irreducible representation of \(\mathrm{GL}_{n+1}(F)\) and \(\pi_2\) be an irreducible representation of \(\mathrm{GL}_n(F)\). Dipendra Prasad established a profound formula relating the Euler-Poincare pairing between \(\pi_1|_{\mathrm{GL}_n(F)}\) and \(\pi_2\) to the dimension of Whittaker functionals on \(\pi_1\) and \(\pi_2\) in [\textit{D. Prasad}, in: Proceedings of the international congress of mathematicians, ICM 2018, Rio de Janeiro, Brazil, August 1--9, 2018. Volume II. Invited lectures, 1367--1392 (2018; Zbl 1443.11086)]. More precisely, one has \[ \mathrm{EP}_{\mathrm{GL}_n(F)}(\pi_1,\pi_2)=\dim \mathrm{Wh}(\pi_1)\cdot\dim \mathrm{Wh}(\pi_2). \] On the other hand, a direct consequence of the so-called multiplicity one theorem [\textit{A. Aizenbud} et al., Ann. Math. (2) 172, No. 2, 1407--1434 (2010; Zbl 1202.22012); Sel. Math., New Ser. 15, No. 2, 271--294 (2009; Zbl 1185.22006)] is that, if both \(\pi_1\) and \(\pi_2\) are generic, one has \[ \dim \mathrm{Hom}_{\mathrm{GL}_n(F)}(\pi_1,\pi_2)=1. \] Combining the above two facts, Prasad conjectured that the higher extension groups \(\mathrm{Ext}^i_{\mathrm{GL}_n(F)}(\pi_1,\pi_2)\) vanish for \(i>0\) if both representations are generic. In the paper under review, the authors prove the above conjecture via a careful study of Bernstein-Zelevinsky derivatives [\textit{I. N. Bernshteĭn} and \textit{A. V. Zelevinskiĭ}, Russ. Math. Surv. 31, No. 3, 1--68 (1976); translation from Usp. Mat. Nauk 31, No. 3(189), 5--70 (1976; Zbl 0342.43017); Ann. Sci. Éc. Norm. Supér. (4) 10, 441--472 (1977; Zbl 0412.22015)] and their variants, namely, the left derivatives and the right derivatives. They also applied the Hecke algebra techniques to prove that \(\pi_1|_{\mathrm{GL}_n(F)}\) is projective if \(\pi_1\) is an essentially discrete series representation of \(\mathrm{GL}_{n+1}(F)\). Moreover, they also discussed how to apply the projectivity result and the cohomological duality to understand the sub-restriction problem, i.e., \(\mathrm{Hom}_{\mathrm{GL}_n(F)}(\pi_2,\pi_1)\).
0 references
branching problems
0 references
representations of general linear groups
0 references
0 references
0 references
0 references
0 references