A motivic approach to rational p p -adic cohomologies
Federico Binda, Alberto Vezzani · 2025
Read A motivic approach to rational p p -adic cohomologies by Federico Binda, Alberto Vezzani (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-16196.txt.
Introduction
Federico Binda and Alberto Vezzani survey new ways to define p-adic cohomology through motives. Their 2025 paper reviews recent work in this corner of arithmetic geometry. Cohomology theories for p-adic settings are hard to build and compare. Motive theory, a unifying frame for such theories, now offers fresh definitions.
The survey revisits the proof of a known conjecture on weights and monodromy for smooth projective hypersurfaces. The new proof leans on the motivic picture of nearby cycles and monodromy operators. That lens makes each step of the old argument cleaner to state. The authors aim to guide readers through this fast-moving area. Specialists in motives and L-functions are the natural audience.
Catalog record
| CREATOR | Federico Binda, Alberto Vezzani |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 114 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2508-16196
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
A motivic approach to rational p p -adic cohomologies Thanks: The authors are partially supported by the PRIN 20222B24AY “The arithmetic of motives and L L -functions” Federico Binda, Alberto Vezzani Abstract . We survey over some recent applications of motivic homotopy theory in the definition and the study of p p -adic cohomology theories. In particular, we revisit the proof of the p p -adic weight-monodromy conjecture for smooth projective hypersurfaces in light of the motivic definition of nearby cycles and monodromy operators. 1. Introduction 1.1. From the special fiber to the generic fiber One of the first motivations for the introduction of analytic geometry over non-archimedean fields is the attempt of defining a p -adic Weil cohomology theory (as opposed to the \ell -adic étale cohomology with \ell\neq p ) for varieties over a finite field k of characterstic p , and the study of the associated zeta functions. Following ideas of Dwork, Monsky and Washnitzer [ 47 ] introduced a way to attach to an affine, smooth variety over k a vector space H^{i}_{MW}(X/K) over K=\Frac W(k) . This can be thought as an “overconvergent” de Rham cohomology of a (dagger rigid analytic) generic fiber of a weakly complete smooth lift of X over W(k) . We may interpret Monsky-Washnitzer’s plan was as a way to define a “functor” (1.1) \Sm/k\leadsto\FSm/\mathcal{O}_{K}\leadsto\RigSm/K\leadsto\RigSm^{\dagger}/K with K being the p -adic non-archimedean field W(k)[1/p] , and where \Sm [resp. \FSm resp. \RigSm resp. \RigSm^{\dagger} ] stands for the category of smooth varieties [resp. formal schemes, resp. rigid analytic variety, resp. dagger varieties] over the the base. Nonetheless, of all the arrows sketched above, only the middle one is an actual functor. As a matter of fact, the existence of some formal lift is only granted locally, for affine varieties X . It is not even enough to work locally (that is, considering the Zariski topos): indeed, even if affine smooth lifts are unique up to isomorphism, their “weak completions” depend on a choice of coordinates for X . The same ambiguity exists when trying to lift maps between affine smooth varieties. It was only thanks to Berthelot’s introduction of the overconvergent site [ 13 ] that Monsky–Washnitzer cohomology could be turned into a well-defined functorial cohomology theory, extendable to arbitrary varieties over k and with the usual properties of a Weil cohomology theory (a Künneth formula, finite dimensionality etc.). We propose an alternative way to “straighten up” the squiggly arrows of ( 1.1 ) based on motivic homotopy theory. Indeed, by results of Elkik [ 28 ] and Arabia [ 3 ] , we know that potential lifts of maps are unique, up to homotopy . Such results suggest that one should consider a more flexible environment, in which not only can one work Zariski-locally (or even étale locally) but also up to homotopy . More or less by definition, this environment is the so-called category of (derived, étale, rational) motives that we will denote by \DA or \RigDA or \RigDA^{\dagger} in the algebraic resp. analytic resp. dagger setting. The aim of Section 3 , in which we collect results from [ 59 , 44 , 29 ] is to sketch how one can prove the following results. Proposition 1.2 . There is a functor \xi\colon\Sm/k\to\DA(k)\to\RigDA^{\dagger}(K) Whenever X/k admits a weak formal smooth lift \mathfrak{X}^{\dagger} over W(k) , then \xi(X)\simeq\mathsf{M}(\mathfrak{X}^{\dagger}_{K}) . We can use this functor to define Monsky–Washnitzer/rigid cohomology and prove even new features of them. In what follows, we let \Fr/S be the category of proper frames over S in the sense of Berthelot. Theorem 1.3 . Let S be a scheme over k . There is a functor \xi\colon\Sm/S\to\DA(S)\to\lim_{(T,\bar{T},\mathfrak{P})\in\Fr/S}\RigDA^{\dagger}(]{T}[^{\dagger}_{\mathfrak{P}}) which, composed to an overconvergent relative de Rham functor gives rise to a cohomology theory: R\Gamma_{\rig}\colon\colon\Sm/S\to\DA(S)\to\lim_{(T,\bar{T},\mathfrak{P})\in\Fr/S}\QCoh(]{T}[^{\dagger}_{\mathfrak{P}})^{\op}. computing (relative) rigid cohomology R\Gamma_{\rig}(X/(T,\bar{T},\mathfrak{P})) . In particular, if S=\Spec k we get an absolute Monsky–Washnitzer realization: R\Gamma_{\rig}\colon\Sm/k\to\mathcal{D}(K)^{\op}. Corollary 1.4 . Let S be a scheme over k . (1) If S=\Spec k , then (absolute) Monsky–Washnitzer cohomology is well-defined, functorial, has étale and h-descent, and can be extended to arbitrary varieties by putting R\Gamma_{\rig}(X/K)=R\Gamma_{\rig}(\mathsf{M}_{k}(X)) . It has the structure of a \varphi -module, and is finite dimensional whenever X is a smooth quasi-compact rigid variety, or the analytification of an algebraic variety. (2) Relative rigid cohomology has étale descent and h-descent. If X/S is smooth and proper, then R^{i}\Gamma_{\rig}(X/(T,\bar{T},\mathfrak{P})) is a vector bundle over ]T[^{\dagger}_{\mathfrak{P}} and is equipped with a canonical structure of an overconvergent F -isocrystal. We note that point ( 1 ) was essentially established by Berthelot and Chiarellotto–Tsuzuki, so that motivic methods help re-visit classical resuls. On the other hand, point ( 2 ) answers positively to an open conjecture by Berthelot which was previously known only in the (log-)liftable case, and for a special class of frames [ 13 , 57 , 55 ] . 1.2. From generic fibers to special fibers One of the classical ways to define an analytic variety over a non-archimedean field arises when one has a generically smooth family X_{t} of varieties over a curve C over a field k of characteristic 0 . When restricting around an infinitesimal “punctured neighborhood” of a point 0\in C , one ends up with a formal model of a smooth rigid analytic variety over k(\!(t)\!) , having a (potentially singular) special fiber X_{0} over k . By standard resolution of singularities, one may even suppose that such special fiber is semi-stable (at least after an extension of scalars). By classic work by Clemens [ 21 ] , Schmid [ 53 ] , Steenbrink [ 56 ] , Saito [ 52 ] et al. when k=\mathbb{C} , the limit Hodge cohomology of the elements of the family \lim_{t\to 0}H(X_{t}) can be expressed in terms of a nearby cycle functor \Psi X , introduced either via passing to a universal covering and using the functor \iota^{*}j_{*} , or as an explicit bi-complex of Hodge modules depending on the geometry of X_{0} . It comes equipped with a monodromy operator. In the arithmetic situation, by replacing the punctured complex disc and the non-archimedean field \mathbb{C}(\!(t)\!) with a local field K , this last procedure has been translated to the study of \ell -adic cohomology by Rapoport–Zink [ 51 ] , interpreting the Weil-Deligne cohomology of the generic fiber in terms of \ell -adic cohomology of a bi-complex \Psi X , equipped with a monodromy operator. Similarly, this procedure was adapted to the p -adic case, with the introduction of a so called Hyodo–Kato cohomology, by work of Hyodo–Kato [ 37 ] , Mokrane [ 46 ] , Nakkajima [ 49 ] , Große-Klönne [ 34 ] , Colmez–Niziol [ 22 ] and Ertl–Yamada [ 30 ] . Once again, Hyodo–Kato cohomology of the (rigid analytic) generic fiber can be described as rigid cohomology of a bi-complex \Psi X , and comes equipped with a monodromy operator. In all three circumstances, the strategy is then to consider a “nearby cycle functor” \Psi\colon\RigSm/K\leadsto\Sm/k+\text{monodromy} and then define/study a “limit” cohomology for X/K via a cohomology theory for \Psi X . Once again, this functor is not really well-defined, as typically the nearby cycle functor is given by some bi-complex of objects in