We fix an error on a $3$ -cocycle in the original version of the paper ‘Endoscopy for Hecke categories, character sheaves and representations’. We give the corrected statements of the main results.

We present a Langlands dual realization of the putative category of affine character sheaves. Namely, we calculate the categorical center and trace (also known as the Drinfeld center and trace, or categorical Hochschild cohomology and homology) of the affine Hecke category starting from its spectral presentation. The resulting categories comprise coherent sheaves on the commuting stack of local systems on the two-torus satisfying prescribed support conditions, in particular singular support conditions, which appear in recent advances in the geometric Langlands program. The key technical tools in our arguments are a new descent theory for coherent sheaves or ${\mathcal{D}}$-modules with prescribed singular support and the theory of integral transforms for coherent sheaves developed in the companion paper by Ben-Zvi et al. [Integral transforms for coherent sheaves, J. Eur. Math. Soc. (JEMS), to appear].

In this paper, we extend the notion of Shintani descent to general (possibly disconnected) algebraic groups defined over a finite field $\mathbb{F}_{q}$. For this, it is essential to treat all the pure inner $\mathbb{F}_{q}$-rational forms of the algebraic group at the same time. We prove that the notion of almost characters (introduced by Shoji using Shintani descent) is well defined for any neutrally unipotent algebraic group, i.e. an algebraic group whose neutral connected component is a unipotent group. We also prove that these almost characters coincide with the ‘trace of Frobenius’ functions associated with Frobenius-stable character sheaves on neutrally unipotent groups. In the course of the proof, we also prove that the modular categories that arise from Boyarchenko and Drinfeld’s theory of character sheaves on neutrally unipotent groups are in fact positive integral, confirming a conjecture due to Drinfeld.

We consider the rigid monoidal category of character sheaves on a smooth commutative group scheme $G$ over a finite field $k$, and expand the scope of the function-sheaf dictionary from connected commutative algebraic groups to this setting. We find the group of isomorphism classes of character sheaves on $G$, and show that it is an extension of the group of characters of $G(k)$ by a cohomology group determined by the component group scheme of $G$. We also classify all morphisms in the category character sheaves on $G$. As an application, we study character sheaves on Greenberg transforms of locally finite type Néron models of algebraic tori over local fields. This provides a geometrization of quasicharacters of $p$-adic tori.

Soit $G$ un groupe algébrique réductif sur la clôture algébrique d’un corps fini $\mathbb{F}_q$ et défini sur ce dernier. L’existence du support unipotent d’un caractère irréductible du groupe fini $G(\mathbb{F}_q)$, ou d’un faisceau caractère de $G$, a été établie dans différents cas par Lusztig, Geck et Malle, et le second auteur. Dans le présent article, nous démontrons que toute classe unipotente sur laquelle la restriction du caractère ou du faisceau caractère donné est non nulle est contenue l’adhérence de Zariski du support unipotent de ce dernier. Pour établir ce résultat, nous étudions certaines représentations des groupes de Weyl, dites “bien supportées”.

Let $G$ be a reductive algebraic group over the algebraic closure of a finite field $\mathbb{F}_q$, and defined over the latter. The existence of a unipotent support for an irreducible character of the finite group $G(\mathbb{F}_q)$, or for a character sheaf on $G$, has been established in several cases by Lusztig, Geck and Malle, and the second author. In this paper, we prove that any unipotent class on which the restriction of the given character or character sheaf does not vanish is contained in the Zariski closure of its unipotent support. In order to establish this result, we study some representations of Weyl groups, which are called ‘well supported’.

The statement (and the proof) of Theorem 4.1.1 of the paper ‘Mackey formula in type A’ [Proc. London. Math. Soc. (3) 80 (2000) 545–574] is false. In this corrigendum we provide a correct statement (and a correct proof). As a consequence, we see that Corollary 4.1.2 of the paper holds. So all the other statements in the paper are correct (up to minor misprints…).

2000 Mathematical Subject Classification: 20G05, 20G40.