On p-adic L-functions for GL2n in finite slope Shalika families

Abstract

In this paper, we propose and explore a new connection in the study of p-adic L-functions and eigenvarieties. We use it to prove results on the geometry of the cuspidal eigenvariety for GL2n over a totally real number field F at classical points admitting Shalika models. We also construct p-adic L-functions over the eigenvariety around these points. Our proofs proceed in the opposite direction to established methods: rather than using the geometry of eigenvarieties to deduce results about p-adic L-functions, we instead show that non-vanishing of a (standard) p-adic L-function implies smoothness of the eigenvariety at such points. Key to our methods are a family of distribution-valued functionals on (parahoric) overconvergent cohomology groups, which we construct via p-adic interpolation of classical representation-theoretic branching laws for GLn × GLn ⊂ GL2n. More precisely, we use our functionals to attach a p-adic L-function to a non-critical refinement π of a regular algebraic cuspidal automorphic representation π of GL2n/F which is spherical at p and admits a Shalika model. Our new parahoric distribution coefficients allow us to obtain optimal non-critical slope and growth bounds for this construction. When π has regular weight and the corresponding p-adic Galois representation is irreducible, we exploit non-vanishing of our functionals to show that the parabolic eigenvariety for GL2n/F is \'etale at π over an ([F:Q]+1)-dimensional weight space and contains a dense set of classical points admitting Shalika models. Under a hypothesis on the local Shalika models at bad places which is empty for π of level 1, we construct a p-adic L-function for the family.

0

Turn this paper into a lesson

ArcXiv compiles a structured reading guide from this paper's metadata: plain-English importance, contributions, prerequisite concepts, which sections to read first, flashcards, and a quiz. Grounded in the abstract, never invented.

Discussion (0)

Sign in to join the discussion.

Loading comments…