arxiv: 2511.15126 · v2 · submitted 2025-11-19 · 🧮 math.AG

G-gerbes on perfectoid spaces

Xiaohuan Long , Yibin Wang , Xiangdong Wu , Ru Yi This is my paper

Pith reviewed 2026-05-17 21:16 UTC · model grok-4.3

classification 🧮 math.AG

keywords G-gerbesperfectoid spacesv-topologyétale topologyrigid groupssite morphismsdirect image functorsnon-abelian cohomology

0 comments p. Extension

The pith

The v-site and étale site of a perfectoid space give equivalent 2-categories of G-gerbes for any rigid group G.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper shows that the natural morphism of sites from the v-topology to the étale topology on a perfectoid space X over a complete non-archimedean field K induces an equivalence of 2-categories of G-gerbes, where G is any rigid analytic group over K. This directly extends Heuer's earlier equivalence of categories of G-torsors under the same site morphism. A sympathetic reader would care because gerbes encode higher-order data such as automorphisms of torsors and banded extensions, so the equivalence means these structures do not change when one passes between the finer v-covers and the coarser étale covers. If the claim holds, any classification or deformation problem involving G-gerbes can be solved indifferently in either topology, and invariants computed on one side automatically match the other.

Core claim

Let K be a complete non-archimedean field over Q_p, G a rigid group over K, and X a perfectoid space over K. The natural morphism of sites ν: X_v → X_ét is known to induce an equivalence of categories of G-torsors via the direct image functor ν_*. The paper establishes that the same functor lifts to an equivalence of 2-categories between the G-gerbes on the v-site and the G-gerbes on the étale site.

What carries the argument

The site morphism ν: X_v → X_ét together with its direct image functor ν_*, which carries the known torsor equivalence up to the 2-categorical level of gerbes.

If this is right

Every G-gerbe on the v-site corresponds canonically to a G-gerbe on the étale site and conversely.
The automorphism 2-groups of corresponding gerbes are equivalent, so banded extensions and non-abelian cohomology classes agree.
Any statement about the existence, classification, or deformation of G-gerbes proven in one topology holds automatically in the other.
Computations of gerbe cohomology or moduli problems involving G-gerbes can be carried out on the coarser étale site whenever convenient.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The same site morphism may preserve equivalences for higher gerbes or for other 2-stacks banded by G.
The result suggests that many non-abelian cohomology sets attached to rigid groups on perfectoid spaces are insensitive to the choice between v and étale topologies.
This equivalence could be used to compare gerbe-theoretic constructions in rigid geometry with constructions that appear in nearby p-adic Hodge-theoretic settings.

Load-bearing premise

The already-known equivalence of G-torsors under the direct image from the v-site to the étale site must lift naturally to the 2-categorical data of gerbes without losing or adding extra automorphisms.

What would settle it

A concrete counterexample would be any perfectoid space X and rigid group G for which the set of isomorphism classes of G-gerbes, or the automorphism 2-groups of those gerbes, differs in cardinality or structure between the v-topology and the étale topology.

read the original abstract

Let $K$ be a complete non-archimedean field over $\mathbb{Q}_p$, $G$ be a rigid group over $K$, and $X$ be a perfectoid space over $K$. We consider the natural morphism of sites $\nu: X_v \to X_{\mathrm{\acute{e}t}}$. It is known from work of Heuer that the direct image functor $\nu_*$ induces an equivalence of the categories of $G$-torsors. In this article, we show that there is an equivalence of 2-categories of $G$-gerbes on these two topologies.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper lifts Heuer's torsor equivalence to an equivalence of 2-categories for G-gerbes on perfectoid spaces, but the 2-morphism and descent checks are the part that needs the closest look.

read the letter

The paper's central result is an equivalence of 2-categories of G-gerbes on the v-topology and the étale topology of a perfectoid space, induced by the direct image from the site morphism ν. This builds on Heuer's earlier equivalence for G-torsors and moves it up one categorical level. What the paper does well is to identify a clean extension in the setting of rigid groups over p-adic fields and perfectoid spaces. The statement is direct and avoids unnecessary complications. The soft spots are in the higher categorical lifting. For the equivalence to hold, the functor needs to be fully faithful and essentially surjective not just on objects but also on the morphisms and 2-morphisms of the 2-category. Gerbes are locally trivial in a 2-sense, so one has to confirm that ν_* respects the banding and the automorphism 2-groups without introducing mismatches from the finer covers. The concern about additional descent data from v-covers is reasonable to raise. If the paper shows that any v-descent for gerbes comes from étale data, that would close the issue. Without seeing the proofs, it is hard to tell how much new work went into that part. Overall, the argument seems to rest on the torsor equivalence plus some standard 2-category facts, which is fine if those facts apply directly here. This paper is for specialists in perfectoid spaces and p-adic cohomology who might use gerbe equivalences to simplify computations in higher descent or bandings. It is the sort of result that could appear in a journal like Compositio or a similar venue after refereeing. I would recommend putting it through peer review so the details on the 2-morphisms can be vetted by someone familiar with site morphisms in algebraic geometry.

Referee Report

2 major / 2 minor

Summary. The manuscript shows that for a perfectoid space X over a complete non-archimedean field K ⊃ ℚ_p and a rigid group G over K, the direct image functor ν_* induced by the natural morphism of sites ν: X_v → X_ét yields an equivalence of 2-categories between the G-gerbes on the v-site and on the étale site. This extends the known equivalence of categories of G-torsors established by Heuer.

Significance. If the central claim holds, the result supplies a higher-categorical comparison of the v and étale topologies that is useful for descent questions and cohomology computations involving banded gerbes in p-adic geometry. The extension from the 1-categorical torsor case to the 2-categorical gerbe case is a natural increment that could support further work on rigid-analytic stacks or p-adic Hodge theory.

major comments (2)

[§4.2] §4.2 (proof of the 2-categorical equivalence): the argument that ν_* induces an equivalence on 2-morphisms (automorphisms of banded gerbes) assumes without explicit verification that local triviality and descent data for v-covers descend fully to the étale site; this step is load-bearing for the 2-category equivalence and is not reduced to the torsor case alone.
[Theorem 5.1] Theorem 5.1: the claim that the induced map on automorphism 2-groups is an equivalence relies on an implicit identification H²(X_ét, G) ≃ H²(X_v, G) for the rigid band G, but no separate argument or reference is given showing that the finer v-covers do not produce additional 2-cocycle data; this isomorphism is central to the 2-categorical statement.

minor comments (2)

[§2] The definition of the 2-category of G-gerbes (objects, 1-morphisms as G-torsors, 2-morphisms as automorphisms) is introduced without a numbered display or reference to a standard source; adding an explicit reference or diagram would improve readability.
[§3] Notation for the functor ν_* on gerbes is used before it is formally defined; a short preliminary paragraph clarifying the extension from torsors to gerbes would help.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their careful reading of the manuscript and for the constructive major comments. We address each point below and have revised the manuscript to strengthen the exposition and fill in the indicated gaps.

read point-by-point responses

Referee: [§4.2] §4.2 (proof of the 2-categorical equivalence): the argument that ν_* induces an equivalence on 2-morphisms (automorphisms of banded gerbes) assumes without explicit verification that local triviality and descent data for v-covers descend fully to the étale site; this step is load-bearing for the 2-category equivalence and is not reduced to the torsor case alone.

Authors: We thank the referee for highlighting this point. The 2-morphisms in the 2-category of G-gerbes are indeed given by G-torsors (via the banding), so the equivalence on 2-morphisms reduces directly to Heuer's equivalence of categories of G-torsors. We have revised §4.2 by inserting an explicit lemma that verifies the descent of local triviality conditions and the associated descent data from v-covers to étale covers, using the torsor equivalence to identify the relevant cocycle data on double intersections. This makes the reduction fully explicit without relying on the torsor case alone. revision: yes
Referee: [Theorem 5.1] Theorem 5.1: the claim that the induced map on automorphism 2-groups is an equivalence relies on an implicit identification H²(X_ét, G) ≃ H²(X_v, G) for the rigid band G, but no separate argument or reference is given showing that the finer v-covers do not produce additional 2-cocycle data; this isomorphism is central to the 2-categorical statement.

Authors: We agree that the identification H²(X_ét, G) ≃ H²(X_v, G) requires a separate justification to avoid any appearance of circularity. In the revised manuscript we have added a preliminary lemma (now placed before Theorem 5.1) that establishes this isomorphism for rigid groups G. The argument proceeds by showing that any 2-cocycle with respect to a v-cover can be refined, up to cohomology, to an étale cover, using the fact that perfectoid spaces admit étale refinements of v-covers together with the rigidity of G (which ensures that the band does not introduce additional data at the v-level). This lemma is independent of the main 2-categorical equivalence. revision: yes

Circularity Check

0 steps flagged

No circularity: extends external torsor equivalence independently to gerbes

full rationale

The paper cites Heuer's external result for the equivalence of G-torsors under ν: X_v → X_ét and claims a separate proof for the 2-categorical equivalence of G-gerbes. No quoted step reduces the gerbe claim to a self-definition, fitted input renamed as prediction, or load-bearing self-citation chain. The derivation is self-contained against the external benchmark and does not exhibit any of the enumerated circular patterns.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claim rests on domain assumptions from p-adic geometry and prior results rather than new free parameters or invented entities.

axioms (2)

domain assumption Standard properties of perfectoid spaces and the morphism of sites ν: X_v → X_ét
Invoked to set up the topologies whose gerbes are compared.
domain assumption Heuer's equivalence of G-torsors under ν_*
Used as the base case to lift to 2-categories of gerbes.

pith-pipeline@v0.9.0 · 5396 in / 1223 out tokens · 26063 ms · 2026-05-17T21:16:00.407295+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

3 extracted references · 3 canonical work pages

[1]

Bosch, U

[BGR84] S. Bosch, U. G¨ untzer, and R. Remmert,Non-Archimedean analysis, Grundlehren der mathematischen Wissenschaften [Fundamental Principles of Mathematical Sciences], vol. 261, Springer-Verlag, Berlin, 1984, A systematic approach to rigid analytic geometry. MR 746961 [DG67] Jean Dieudonn´ e and Alexander Grothendieck, ´El´ ements de g´ eom´ etrie alg´ ...

work page 1984
[2]

[Gru66] Laurent Gruson,Th´ eorie de Fredholmp-adique, Bull. Soc. Math. France94(1966), 67–95. MR 226381 [Heu24] Ben Heuer,G-torsors on perfectoid spaces,

work page 1966
[3]

[SW20] Peter Scholze and Jared Weinstein,Berkeley lectures onp-adic geometry, Annals of Math- ematics Studies, vol

[Sch22] Peter Scholze, ´Etale cohomology of diamonds, 2022, to appear in Ast´ erisque. [SW20] Peter Scholze and Jared Weinstein,Berkeley lectures onp-adic geometry, Annals of Math- ematics Studies, vol. 207, Princeton University Press, Princeton, NJ,

work page 2022