🔢 mathematics

Uniform irreducibility of Galois action on the $\ell$ -primary part of Abelian $3$-folds of Picard type

This paper generalizes Manin's classical result on uniform bounds for cyclic $\ell$ -power isogenies between non-CM elliptic curves to certain 2-dimensional families of abelian 3-folds with multiplication by an imaginary quadratic field.

Original authors: Mladen Dimitrov, Dinakar Ramakrishnan

Published 2026-02-27

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Mladen Dimitrov, Dinakar Ramakrishnan

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 you are a detective trying to solve a mystery about a very special kind of mathematical object called an Abelian 3-fold. To make this easier to understand, let's think of these objects not as abstract equations, but as complex, multi-dimensional shapes that live in a vast, invisible landscape.

Here is the story of what Mladen Dimitrov and Dinakar Ramakrishnan discovered, broken down into simple concepts.

1. The Setting: A City of Shapes

Imagine a city called Picard Land. In this city, there are millions of these special 3-dimensional shapes (Abelian 3-folds). They are built according to strict blueprints (mathematical rules involving "imaginary quadratic fields").

Every shape in this city has a hidden "security system" called a Galois Group. Think of this security system as a set of guards that watch over the shape. These guards can rearrange the shape's internal parts, but they must follow specific rules.

2. The Problem: The "Flag" Intruder

The mathematicians are worried about a specific type of intruder: a "Full Rational Flag."

The Analogy: Imagine your shape is a 3-story building. A "flag" is a specific way of stacking a 1-story tower inside a 2-story tower, which is then placed inside the 3-story building.
The Threat: If the shape has a "rational flag," it means the guards (the Galois group) are lazy. They are only rearranging the building in a very predictable, limited way (like only moving things in a straight line). They aren't doing the full, chaotic dance they are supposed to do.
The Goal: The authors want to prove that for most of these shapes, the guards are not lazy. They want to show that the guards are so active and chaotic that they cannot be trapped in a simple, predictable pattern (a "Borel subgroup").

3. The Previous Clue: The One-Dimensional Case

Fifty years ago, a famous mathematician named Yuri Manin solved a similar mystery for Elliptic Curves (which are like 1-dimensional versions of these shapes, or perhaps "1-story buildings").

Manin proved that if you look at enough of these 1-story buildings, you will eventually find a level of security where no lazy guards exist. There is a "uniform bound"—a limit on how many times you can check before you are guaranteed to find a shape with active, chaotic guards.

4. The New Challenge: The 3-Dimensional Puzzle

The authors of this paper wanted to do the same thing for Abelian 3-folds (the 3-story buildings).

The Difficulty: This is much harder. The "city" (the mathematical space) is now two-dimensional (like a surface or a map), not just a line.
The Obstacle: In the 1-story case, the city was simple. In the 3-story case, the city is messy. It has "singularities" (cracks in the pavement) and "torsion" (twists in the road) that make it hard to apply the old rules.

5. The Solution: Building a "Super-Map"

To solve this, the authors had to build a new kind of map and use some very advanced tools.

Step A: The "Gross Subgroup" (The Special Gate)

They realized that to filter out the "lazy" shapes, they needed to add a tiny bit of extra structure at the edges of the city (specifically at a prime number $D$ ).

Analogy: Imagine putting a special, high-tech gate at the entrance of the city. This gate only lets in shapes that pass a specific "symmetry test." If a shape fails this test, it's likely one of the "lazy" ones we want to avoid. This gate is called the Gross subgroup.

Step B: The "Irregularity" (The Noise Level)

To prove the guards are active, they needed to show the city is "noisy" enough. In math, this is called irregularity.

The Metaphor: Think of the city as a concert hall. If the hall is too quiet (low irregularity), you can't hear the music (the complex behavior of the guards). The authors needed to prove the hall is so loud and chaotic (high irregularity) that the music must be complex.
The Trick: They used a technique involving automorphic forms (which are like complex musical notes played on a unitary group). By carefully choosing the "notes" (characters), they proved the concert hall is indeed loud enough to guarantee the guards are active.

Step C: The Bombieri-Lang Conjecture (The "No Crowds" Rule)

They used a famous hypothesis called the Bombieri-Lang Conjecture.

The Analogy: This conjecture says that in a very complex city (one with high irregularity), you cannot have a "crowd" of rational points (shapes with simple guards) spreading out everywhere. The rational points must be sparse, like scattered islands.
The Result: Because their "city" was proven to be complex enough, the authors knew that the "lazy" shapes couldn't be everywhere. They had to be rare.

6. The Final Verdict

By combining these tools, the authors proved Theorem B:

If you pick a number field (a specific set of rules) and a prime number, there is a limit $r$ . If you look at any non-CM Abelian 3-fold in this city, and you check its security at level $r$ , you will never find a "full rational flag."

In plain English:
No matter which of these special 3-dimensional shapes you pick (as long as it's not a boring, repetitive one), if you look closely enough at its internal structure, you will always find that its "guards" are doing a complex, chaotic dance. They are never stuck in a simple, predictable line.

Why Does This Matter?

Uniformity: Before this, we knew this was true for individual shapes. Now we know it's true for all shapes in a whole family at once.
The "First": This is the first time this kind of proof has been done for a two-dimensional family of shapes. It's like moving from proving a rule for a single street to proving it for an entire neighborhood.
The Legacy: The paper is dedicated to Yuri Manin, who started this line of thinking 50 years ago. The authors have successfully taken his idea and expanded it into a much larger, more complex world.

Summary: The authors built a mathematical "noise machine" to prove that in a vast city of complex 3D shapes, the security guards are always too busy dancing to ever stand in a simple, predictable line.

1. Problem Statement and Context

The Core Problem:
The paper addresses the uniform boundedness of the $\ell$ -primary torsion for families of abelian varieties. Specifically, it investigates whether there exists a uniform integer $r$ (depending on the number field $k$ , the prime $\ell$ , and the family) such that for any non-CM abelian 3-fold $A$ in a specific family, the Galois representation on the $\ell^r$ -torsion points $A[\ell^r]$ does not contain a full $k$ -rational flag (a line contained in a plane, both stable under the Galois action).

Historical Context:

Manin (1970s): Proved that for non-CM elliptic curves over a number field $k$ , there is a uniform bound $r$ such that the image of the Galois representation modulo $\ell^r$ is not contained in a Borel subgroup (i.e., no rational line exists).
Cadoret & Tamagawa: Extended this to 1-dimensional families of abelian varieties.
Gap: This paper aims to establish the first result for a genuine 2-dimensional family of abelian varieties (specifically, abelian 3-folds of Picard type).

The Object of Study:

Picard Type Abelian 3-folds: Principally polarized abelian varieties of dimension 3 with multiplication by the ring of integers $\mathcal{O}_M$ of an imaginary quadratic field $M$ (with odd fundamental discriminant $-D \neq -3$ ).
Moduli Space: These varieties are parametrized by Picard modular surfaces (Shimura surfaces) associated with the unitary group $U(2,1)$ .

2. Main Results

The authors prove two main theorems regarding the non-existence of full rational flags in the $\ell$ -primary torsion:

Theorem A (Semi-stable case): Given a number field $k$ , a prime $\ell$ inert in $M$ , and a finite set of places $S$ , there exists an integer $r$ such that for any non-CM abelian 3-fold of Picard type semi-stable outside $S$ , $A[\ell^r]$ contains no full $k$ -rational flag.
Theorem B (Main Result): Given $k$ $k$ and $\ell$ $ℓ$ (inert in $M$ $M$ ), there exists an integer $r$ $r$ such that for any non-CM abelian 3-fold $A$ $A$ of Picard type satisfying a specific level condition at the discriminant $D$ $D$ (specifically, that a canonical character $\varepsilon_{A,D}$ $ε_{A, D}$ is trivial), $A[\ell^r]$ $A [ℓ^{r}]$ contains no full $k$ $k$ -rational flag.
- Note: Theorem B implies Theorem A by extending the base field to kill the obstruction at $D$ .

3. Methodology and Technical Approach

The proof strategy relies on the interplay between Arithmetic Geometry (moduli spaces, Faltings' theorems) and Automorphic Forms (representation theory of $U(3)$ ).

A. The Geometric Strategy (Bombieri–Lang Conjecture)

To prove uniformity, the authors must show that the $k$ -rational points on the relevant Shimura surfaces are not Zariski dense.

Bombieri–Lang Conjecture: If a surface is of "general type," its rational points are not Zariski dense.
Irregularity ( $q$ ): A key invariant is the irregularity $q(X) = \dim H^1(X, \mathcal{O}_X)$ . If $q(X) \geq 3$ , the Enriques–Kodaira classification implies the surface is either ruled, elliptic, or of general type. The authors aim to show $q \geq 3$ to apply the conjecture (or known cases thereof).
Obstacle: Standard Shimura varieties for these surfaces often have trivial irregularity or are not of general type. The authors must construct specific congruence subgroups (levels) to force high irregularity.

B. The Automorphic Strategy (Generating Irregularity)

The authors use the theory of Arthur packets for the unitary group $U(3)$ to construct automorphic forms that contribute to the cohomology (and thus irregularity) of the modular surfaces.

Endoscopic Representations: They consider non-tempered, endoscopic representations $\pi$ of $U(3)$ attached to anti-cyclotomic Hecke characters $\lambda$ .
The Packet: The local Arthur packet at a prime $p$ contains two representations: a supercuspidal $\pi_c$ and a non-tempered $\pi_n$ .
Level Structure: The goal is to find a level subgroup (specifically Iwahori subgroups $\Gamma_0(\ell^r)$ $Γ_{0} (ℓ^{r})$ ) that admits non-zero invariants in the non-tempered representation $\pi_n$ $π_{n}$ .
- Challenge: Non-generic representations usually lack invariants at standard levels.
- Solution: The authors perform delicate computations involving matrix coefficients, Jacquet modules, and intertwining operators to prove the existence of invariants for specific tamely ramified characters.
The Gross Subgroup Condition: A crucial innovation is the introduction of an index-2 subgroup (the "Gross subgroup" $K''$ ) at the ramified primes dividing $D$ . This condition ensures that the global root number $W(\lambda^3)$ is $+1$ , allowing the non-tempered representation $\pi_n$ to be automorphic. Without this, the relevant surfaces would have trivial Albanese varieties, and the method would fail.

C. Handling Curves and Singularities

Once the Bombieri–Lang conjecture is established for the surface (showing rational points lie in a finite union of curves and points):

Curves of Genus $\leq 1$ : The authors must rule out the possibility that these curves contain infinitely many non-CM points.
Lemma 2.1 (No Complex Reflections): They prove that the specific congruence subgroups used do not contain complex reflections. This implies the modular surface has only isolated singularities.
Reduction to 1D Families: Removing singularities allows the restriction to a smooth curve $C$ , which parametrizes a 1-dimensional family of abelian 3-folds.
Cadoret–Tamagawa Theorem: For 1-dimensional families, the uniform boundedness of Galois images is already known. Thus, the Galois image on the curve is "large" (open in the Mumford–Tate group) for all but finitely many points.
Mumford–Tate Conjecture: For the remaining finite set of points (and the generic point), the Mumford–Tate conjecture (known for dimension 3) ensures the Galois image is large enough to prevent a full rational flag for sufficiently large $r$ .

4. Key Contributions

First 2D Uniformity Result: This is the first proof of uniform irreducibility for a genuine 2-dimensional family of abelian varieties (Picard type).
Representation Theoretic Breakthrough: The paper provides a complete answer to the question of levels for non-tempered endoscopic representations of $U(3)$ . Specifically, it determines exactly when non-tempered representations $\pi_n$ admit invariants under Iwahori subgroups $\Gamma_0(\ell^r)$ , a problem previously open due to the lack of a "new vector" theory for non-generic representations.
Explicit Construction of Irregularity: The authors explicitly construct automorphic forms with non-zero invariants at specific levels, proving that the irregularity of the associated Picard modular surfaces is at least 3. This validates the application of the Bombieri–Lang conjecture in this context.
The "Gross Subgroup" Trick: The introduction of the index-2 subgroup condition at the discriminant $D$ is a novel technical device that resolves sign issues in root numbers, enabling the use of non-tempered representations.

5. Significance

Generalization of Manin's Theorem: It extends a fundamental result from elliptic curves (dim 1) to abelian 3-folds (dim 3) over a 2-dimensional base, bridging a significant gap in the arithmetic of abelian varieties.
Interplay of Fields: The paper demonstrates a powerful synthesis of Shimura varieties, automorphic forms (Arthur packets), Galois representations, and Diophantine geometry (Bombieri–Lang).
Methodological Advance: The techniques developed for computing invariants of non-tempered representations via exponential sums and intertwining operators are likely to be applicable to other problems involving non-generic automorphic forms and level structures.
Foundation for Future Work: By establishing the uniformity for Picard type, it opens the door to investigating similar questions for other Shimura varieties and higher-dimensional abelian varieties, provided the necessary automorphic machinery can be developed.

In summary, the paper resolves a long-standing problem in the uniformity of Galois actions by constructing a specific geometric setting (Picard modular surfaces with high irregularity) and proving that the associated automorphic representations possess the necessary level structures to force the Galois images to be "large" (irreducible) uniformly across the family.