Symmetric power functoriality for holomorphic modular forms. II. (Q2067687)

The paper under review proves the following remarkable theorem that the \(n\)-th symmetric power of a non-CM regular algebraic cuspidal automorphic representation \(\pi\) of \(\mathrm{GL}_{2}(\mathbb{A}_{\mathbb{Q}})\) is a regular algebraic cuspidal automorphic representation of \(\mathrm{GL}_{n+1}(\mathbb{A}_{\mathbb{Q}})\) for any \(n \geqslant 1\). It is a continuation of [Part I, the authors, Publ. Math., Inst. Hautes Étud. Sci. 134, 1--116 (2021; Zbl 1503.11085)], where the same authors proved this theorem under the assumption that \(\pi\) is not supercuspidal at any \(p\)-adic place. To remove this assumption, the authors' idea is to find another regular algebraic cuspidal automorphic representation congruent to \(\pi\) with less supercuspidal places and apply certain automorphy lifting theorem for the symmetric power of the associated \(l\)-adic Galois representation. One of the main difficulties is that all known automorphy lifting theorems such as in [\textit{T. Barnet-Lamb} et al., Ann. Math. (2) 179, No. 2, 501--609 (2014; Zbl 1310.11060); \textit{P. B. Allen} et al., Compos. Math. 156, No. 11, 2399--2422 (2020; Zbl 1480.11063)] do not apply directly in this case. So a large part of this paper is devoted to establishing a new kind of automorphy lifting theorem tailored for the symmetric power of certain \(2\)-dimensional \(l\)-adic Galois representation.