Loose crystalline lifts and overconvergence of \'etale $(\varphi, \tau)$-modules

TL;DR

This paper proves that all étale $(, au)$-modules over $p$-adic fields are overconvergent, extending classical results and introducing new methods involving loose crystalline lifts and Kisin models.

Contribution

It demonstrates the overconvergence of étale $(, au)$-modules using novel approaches with loose crystalline lifts and Kisin models, differing from classical techniques.

Findings

All étale $(, au)$-modules are overconvergent.

Existence of loose crystalline lifts for $p$-power-torsion representations.

Construction of Kisin models to establish overconvergence.

Abstract

Let $p$ be a prime, $K$ a finite extension of $\mathbb Q_p$, and let $G_K$ be the absolute Galois group of $K$. The category of \'etale $(\varphi, \tau)$-modules is equivalent to the category of $p$-adic Galois representations of $G_K$. In this paper, we show that all \'etale $(\varphi, \tau)$-modules are overconvergent; this answers a question of Caruso. Our result is an analogy of the classical overconvergence result of Cherbonnier and Colmez in the setting of \'etale $(\varphi, \Gamma)$-modules. However, our method is completely different from theirs. Indeed, we first show that all $p$-power-torsion representations admit loose crystalline lifts; this allows us to construct certain Kisin models in these torsion representations. We study the structure of these Kisin models, and use them to build an overconvergence basis.