🔢 mathematics

Adjoint Bloch--Kato Selmer groups of regular algebraic automorphic Galois representations

This paper proves the vanishing of the adjoint Bloch–Kato Selmer group for Galois representations associated with regular algebraic automorphic representations of general linear groups over CM fields, marking a significant advancement by requiring conditions solely on the $p$ -adic representations rather than their residual counterparts.

Original authors: Lambert A'Campo, Bence Hevesi, Jack A. Thorne, Dmitri Whitmore

Published 2026-07-14

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Lambert A'Campo, Bence Hevesi, Jack A. Thorne, Dmitri Whitmore

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine the universe of numbers as a vast, bustling city called F. In this city, there are secret agents known as Galois Representations. These agents are like master spies who carry encrypted maps of the city's hidden structures. Sometimes, these maps are so complex they look like tangled knots of string.

Mathematicians have long been trying to solve a specific mystery about these agents: Do they have any "ghostly" sidekicks? In the language of math, these sidekicks are called Bloch–Kato Selmer Groups. If a group of sidekicks exists, it means the agent's map has some hidden, unexplained wiggle room. But if the group is empty (or "vanishes"), it means the agent's map is perfectly rigid, with no hidden secrets left to find.

For a long time, mathematicians could only prove these sidekicks didn't exist if they first checked the agent's "residual" identity—a sort of blurry, low-resolution photo of the spy taken modulo a prime number $p$ . It was like saying, "We can only confirm the spy has no sidekicks if we first verify their blurry ID card is clean." This was a huge hurdle.

The Big Breakthrough
In this paper, the team of authors (A'Campo, Hevesi, Thorne, and Whitmore) has finally proven that for a very specific and important class of spies—those associated with regular algebraic automorphic representations over CM fields (a special type of number city)—the sidekicks do not exist.

Here is the magic twist: They proved this without ever needing to check the blurry ID card (the residual representation). They looked directly at the high-definition, $p$ -adic Galois representation itself. It's as if they proved the spy has no sidekicks just by looking at the spy's sharp, clear face, ignoring the blurry photo entirely.

The "Enormous" and "Pure" Rules
To make this proof work, the spies had to meet two strict criteria, which the authors describe with colorful names:

The "Enormous" Image: The spy's group of friends (the image of the representation) must be "enormous." Think of this as a spy network so vast and diverse that it contains every possible type of agent imaginable. If the network is too small or boring, the proof doesn't hold. The paper explicitly states that if the image isn't enormous, we can't guarantee the sidekicks are gone.
The "Pure" Condition: The spy's local maps (at specific places in the city) must be "pure." Imagine a map that is perfectly balanced, with no smudges or distortions. If the map is "impure," the proof breaks down.

How They Did It: The "Ultra-Patching" Machine
So, how did they prove the sidekicks were gone without the blurry ID card? They used a technique called patching, but they had to invent a new, super-charged version of it called ultra-patching.

Imagine you are trying to build a giant, perfect Lego tower. Usually, you build it one block at a time. But here, the blocks (mathematical objects) were messy and kept changing shape. The authors realized they couldn't just stack them; they had to use a "time machine" (an ultrafilter) to look at an infinite sequence of these towers simultaneously.

They built a "patched complex"—a massive, multi-layered structure that combined all these infinite versions of the towers. By using this super-structure, they could smooth out the messiness. They showed that even though the individual pieces were wobbly, the final, giant structure was so rigid that it forced the "sidekick" group to collapse into nothingness.

What They Ruled Out
The paper is very clear about what it does not do. It does not claim that every Galois representation has vanished sidekicks. If the spy's image isn't "enormous" or if the map isn't "pure," the sidekicks might still be hiding. The authors explicitly state that their result relies on these specific conditions being met. They also note that while they solved the problem for these specific "regular algebraic" spies, other types of spies (like those from modular abelian surfaces) are still a mystery in this specific context.

How Sure Are They?
The authors are not guessing or simulating; they have proven it. They didn't just suggest the sidekicks were gone; they constructed a mathematical argument that leaves no room for doubt, provided the "enormous" and "pure" conditions are met. They proved that the "Zariski tangent space" (a fancy way of measuring the wiggle room) is exactly zero.

The Takeaway
In the end, this paper is a victory for the "rigid" view of the universe. It tells us that for these specific, high-level mathematical spies, their maps are locked tight. There are no ghostly sidekicks hiding in the shadows, and we know this for sure because we finally learned how to look at the spies directly, without needing a blurry backup photo. It's a clean, sharp proof that opens the door to understanding even deeper connections in the city of numbers.

Technical Summary: Adjoint Bloch–Kato Selmer Groups of Regular Algebraic Automorphic Galois Representations

Problem Statement
The paper addresses the vanishing of the adjoint Bloch–Kato Selmer group, $H^1_f(F, \text{ad } \rho)$ , associated with Galois representations $\rho$ arising from regular algebraic automorphic representations of $\text{GL}_n$ over a CM number field $F$ . This group is conjectured to vanish by the Fontaine–Mazur, Beilinson, and Bloch–Kato conjectures. While unconditional results exist for specific cases (notably the "defect 0" or Taylor–Wiles setting), extending these results to the "positive defect" setting (where the cohomology of the associated locally symmetric spaces has torsion and is not concentrated in a single degree) has remained a significant challenge.

A primary difficulty in the positive defect setting is the necessity of working with complexes in the patching argument rather than simple modules. This complicates the application of $R=T$ theorems, as one must control the properties of patched complexes (e.g., ensuring they are finitely generated over diamond operator rings) without the usual control provided by residual representations satisfying the Taylor–Wiles condition. Furthermore, previous approaches often required strong conditions on the residual representation modulo $p$ , such as specific irreducibility or non-triviality conditions on $\rho|_{F(\zeta_p)}$ .

Methodology
The authors employ a refined patching argument using ultrafilters, adapted to handle complexes with potentially unbounded rank cohomology groups. The methodology proceeds through several key technical stages:

Ultra-patching for Complexes: The authors construct patched complexes $M_{\infty, \bullet}$ using a non-principal ultrafilter on a sequence of Taylor–Wiles levels. Unlike previous works that patched modules, this construction patches complexes of singular chains with coefficients in a lattice $V^\vee_\lambda$ .
Control of Nilpotent Kernels: A central technical hurdle is that the resulting $R=T$ type isomorphisms in the patched setting are only valid up to a nilpotent ideal. To overcome this, the authors utilize a technique introduced in [AC23], which leverages the smoothness of the $\mathbb{Q}_p$ -fibre of local deformation rings at generic points. This allows them to deduce properties of the Zariski tangent space (and thus the Selmer group) despite the presence of nilpotent errors in the global isomorphism.
Two-Step Patching with Discriminant Operators: The authors introduce a novel two-step patching process involving a "discriminant" Hecke operator $\delta$ $δ$ .
- First, they replace the space of modular forms $H_{Q_N}$ (free over a ring like $\mathbb{Z}_p[x]/(x^{p^r}-1)$ ) with a patched object $\tilde{H}^r$ (free over $\mathbb{Z}_p[x]/(x^{p^r})$ ).
- This enlargement ensures that the localization $\delta \tilde{H}^r_q$ is free over $\mathbb{Q}_p[x]/(x^{p^r})$ , effectively removing the difference between spherical and Iwahori levels at the Taylor–Wiles places.
- This allows the recovery of the standard patching setup over a regular local ring (e.g., $\mathbb{Q}_p[[x]]$ ) where familiar arguments can be applied.
Local-Global Compatibility: The proof relies on unconditional local-global compatibility results for $\ell \neq p$ and $\ell = p$ , established in companion papers [Hev26] and [AHTW26]. These results are applied under a "decomposed generic" condition to ensure the decomposition of cohomology with respect to the Hodge–Tate period map is well-behaved.

Key Contributions and Results
The main result establishes the vanishing of the adjoint Bloch–Kato Selmer group under conditions imposed solely on the $p$ -adic Galois representation, rather than on its residual representation.

Main Theorem (Theorem 5.1): Let $F$ be a CM field and $\pi$ a cuspidal, regular algebraic automorphic representation of $\text{GL}_n(\mathbb{A}_F)$ . Let $\rho = r_{\pi, \iota}$ be the associated Galois representation. If:
1. The image $\rho(G_F(\zeta_{p^\infty}))$ is "enormous" (in the sense of [NT23]);
2. The extension $L/F(\zeta_{p^\infty})$ cut out by $\text{ad } \rho$ satisfies $H^1(L/F, \text{ad } \rho(1)) = 0$ ;
3. $\rho|_{G_{F_v}}$ is generic for all $v|p$ or where $\pi_v$ is ramified;
  Then $H^1_f(F, \text{ad } \rho) = 0$ .
Corollaries:
- Pure Representations (Corollary 5.2): If $\rho$ is pure (a condition satisfied by many automorphic representations), the vanishing holds.
- Polarizable Representations (Corollary 5.3): For polarizable representations (e.g., those associated to conjugate self-dual automorphic forms), the vanishing holds provided the image is enormous. This result is noted to be strictly stronger than previous work [NT23] as it covers the full adjoint representation over $F$ , not just the self-dual part.
- Symmetric Powers of Elliptic Curves (Corollary 5.4): For an elliptic curve $E$ over $F$ with $\text{End}(E_F) = \mathbb{Z}$ , $H^1_f(F, \text{ad } \text{Sym}^n V_p E) = 0$ for all $n \ge 1$ .

Significance and Claims
The paper claims that its primary novelty lies in imposing conditions only on the $p$ -adic Galois representations and not on their residual representations modulo $p$ . This removes the need for the "Taylor–Wiles condition" on the residual representation, which was a significant restriction in earlier "defect 0" results.

The authors explicitly state that their argument allows for the handling of "positive defect" settings where cohomology groups have significant torsion. By developing a method to patch complexes and control the resulting nilpotent kernels, they extend the scope of Bloch–Kato vanishing results to a broader class of automorphic representations. The work is presented as an unconditional result for specific classes of representations (e.g., those with pure Weil–Deligne representations), relying on the "enormous image" condition and genericity, which are often verifiable in practice (e.g., via Serre's open image theorem for elliptic curves).

The paper does not claim to resolve the Bloch–Kato conjecture in full generality but establishes new unconditional cases where the Selmer group vanishes, thereby providing instances of the conjecture and tools for further applications in Galois deformation theory and the Langlands program.

More like this