On the $v$-adic values of G-functions I

Basic Information

• Paper ID: 2510.11806
• Title: On the $v$-adic values of G-functions I: Splittings in $\mathcal{A}_2$
• Author: Georgios Papas
• Classification: math.NT (Number Theory), math.AG (Algebraic Geometry)
• Publication Date: October 15, 2025
• Paper Link: https://arxiv.org/abs/2510.11806

Abstract

This is the first paper in a series studying families of G-functions associated with one-parameter families of abelian varieties. In particular, the paper constructs relationships between values of special significance in unlikely intersection problems in both Archimedean and non-Archimedean settings. In this first article of the series, the author systematically establishes the theoretical foundations of the series in a unified manner, then investigates values corresponding to "splittings" in $\mathcal{A}_2$ related to the Zilber-Pink conjecture.

Research Background and Motivation

Problem Background

1. G-function Theory: G-functions were originally introduced by C.L. Siegel in the late 1920s. Later, through pioneering work by E. Bombieri and Y. André in the 1980s, G-function theory established clearer connections with arithmetic geometry by studying the values of G-functions at "points of special interest."
2. Unlikely Intersection Problems: In recent years, this circle of ideas has been revitalized due to connections with so-called "unlikely intersection" problems. This connection was initially discovered by C. Daw and M. Orr, who used G-functions and properties of their values at points related to the Zilber-Pink conjecture to provide the first unconditional results for certain cases of this conjecture in $\mathcal{A}_2$.
3. Research Paradigm: Traditional research has focused on Paradigm 1.1 (involving degenerate families with singular values), while this paper shifts to Paradigm 1.2 (studying honest abelian varieties rather than degenerate families).

Research Motivation

1. Theoretical Refinement: Establishing the theoretical foundations of the G-function method under the new paradigm
2. Zilber-Pink Conjecture: Providing new tools and results for the Zilber-Pink conjecture in $\mathcal{A}_2$
3. Height Bounds: Establishing height bounds through relationships between G-function values, thereby proving finiteness results

Core Contributions

1. Theoretical Framework: Establishes a unified theoretical foundation for G-function theory under Paradigm 1.2
2. Relationship Construction: Constructs polynomial relationships between G-function values in both Archimedean and non-Archimedean settings
3. Height Bound Theorem: Proves effective height bounds for splitting points in $\mathcal{A}_2$ (Theorem 1.5)
4. Zilber-Pink Application: Obtains new results for the Zilber-Pink conjecture in specific cases (Corollary 1.6)
5. Technical Innovation: Minimizes the dependence of relationships on finite primes $v$ by utilizing fundamental information from $\phi$-module structures in $p$-adic Hodge theory

Methodology in Detail

Task Definition

Study one-parameter families of abelian surfaces $f: X \to S$, where $S$ is a smooth irreducible curve defined over a number field $K$. The goal is to establish relationships between G-function values for points $s \in S(\overline{\mathbb{Q}})$ where the fiber "splits" (i.e., $X_s \sim E_s \times E'_s$).

Core Technical Framework

1. G-function Construction

Given a family $f: X \to S$ and a base point $s_0 \in S(K)$, construct a family of G-functions through the following steps:

• Select a Hodge basis $\Gamma_{dR}(X) = \{\omega_i, \eta_j\}$
• Obtain a differential system via the Gauss-Manin connection: $\frac{d}{dx}Y = A \cdot Y$
• Construct a normalized fundamental solution $Y_G \in M_{2g}(\overline{\mathbb{Q}}[[x]])$ whose entries are G-functions

2. Comparison Isomorphisms

Utilize de Rham-Betti and de Rham-crystalline comparison isomorphisms: $P_v(s) = \iota_v(Y_G(x(s))) \cdot \Pi_v(X_0)$

where $\Pi_v(X_0)$ is the $v$-adic period matrix of $X_0$.

3. Relationships in the Splitting Case

For a splitting $X_s \sim E_s \times E'_s$, using the isomorphism $\theta_s: X_s \to E_s \times E'_s$, obtain the key relationship: $\iota_v(J_{2,3} \cdot [\theta_s]_{dR} \cdot Y_G(x(s)) \cdot [\theta_0^{\vee}]_{dR} \cdot J_{2,3}) = \begin{pmatrix} \Pi_v(E_s) & 0 \\ 0 & \Pi_v(E'_s) \end{pmatrix} \cdot \Theta \cdot \begin{pmatrix} \Pi_v(E_0)^{-1} & 0 \\ 0 & \Pi_v(E'_0)^{-1} \end{pmatrix}$

Technical Innovations

1. Unified Treatment of Non-Supersingular Reduction

For finite primes with non-supersingular reduction, utilize $\phi$-module structures to make the relationship $R_{s,v}$ nearly independent of the prime $v$, which represents an important improvement over the work of Beukers and others.

2. Categorical Treatment Strategy

• Ordinary Reduction: Utilize diagonalization of the Frobenius action
• Supersingular Reduction: Construct relationships dependent on the prime
• Archimedean Primes: Utilize the Legendre relation

3. Conjectural Framework for Bad Reduction

Propose Conjectures 6.1 and 6.2, utilizing Hyodo-Kato cohomology to handle bad reduction cases.

Experimental Setup

Theoretical Verification

This is primarily a theoretical work, with verification of the non-triviality of constructed polynomial relationships through Mathematica code:

1. Polynomial Computation: Compute relationship polynomials in various cases
2. Gröbner Basis Reduction: Verify that polynomials do not lie in the ideal $I(SP_4)$
3. Coefficient Analysis: Analyze coefficient structure through factorization

Specific Cases

• $E \times CM$ Points: Cases where one elliptic curve is a CM curve
• $E^2$ Points: Cases where two elliptic curves are isomorphic
• Different Reduction Types: Ordinary, supersingular, and bad reduction

Main Results

Theorem 1.3 (Main Technical Result)

Let $f: X \to S$ be a family as described above, and let $s_0, s \in S(\overline{\mathbb{Q}})$ be $E^2$ points or $E \times CM$ points. For $v \in \Sigma_{\mathbb{Q}(S,s_0,s)}$ such that $s$ and $s_0$ are "$v$-adically close," there exists a polynomial $R_{s,v} \in \overline{\mathbb{Q}}[Z_1,\ldots,Z_n]$ satisfying:

1. $\iota_v(R_{s,v}(Y(s))) = 0$
2. $R_{s,v}(Y) \neq 0$ at the functional level
3. $R_{s,v}$ is independent of $v$ (except when $v$ is an Archimedean prime or a supersingular reduction prime of $X_{s_0}$)

Theorem 1.5 (Height Bound)

Under appropriate assumptions, there exist constants $c_1, c_2 > 0$ such that for all splitting points $s$: $h(s) \leq c_1 \cdot (|\Sigma_{\mathbb{Q}(s_0),ssing}(s,0)| \cdot [K(s) : \mathbb{Q}])^{c_2}$

Corollary 1.6 (Zilber-Pink Application)

Let $Z \subset \mathcal{A}_2$ be a smooth irreducible curve not contained in any proper special subvariety. Then for any $N \in \mathbb{N}$, the set $X_{ZP-split,N}(Z) := \{s \in Z(\mathbb{C}) : s \text{ is an } E \times CM \text{ or } E^2 \text{ point and } |\Sigma_{\mathbb{Q}(s_0),ssing}(s,0)| \leq N\}$ is finite.

Related Work

Historical Development

1. Siegel (1929): Introduced the concept of G-functions
2. Bombieri (1981), André (1989): Established connections between G-functions and arithmetic geometry
3. Daw-Orr (2021-): Applied G-functions to unlikely intersection problems

Contributions of This Paper

Compared to existing work, this paper:

• Extends to Paradigm 1.2 (non-degenerate cases)
• Handles splitting cases of abelian surfaces
• Provides a more general theoretical framework
• Improves control over dependence on finite primes

Conclusions and Discussion

Main Conclusions

1. Successfully establishes the theoretical foundations of G-function theory under Paradigm 1.2
2. Constructs effective relationships for splitting cases in $\mathcal{A}_2$
3. Obtains new Zilber-Pink type finiteness results

Limitations

1. Everywhere Potential Good Reduction Assumption: Current results require the central fiber $X_{s_0}$ to have everywhere potential good reduction
2. Supersingular Prime Dependence: At supersingular reduction primes, relationships still depend on the prime
3. Computational Complexity: Verifying non-triviality of relationships requires complex computations

Future Directions

1. Higher-Dimensional Generalizations: Extend to splittings in $\mathcal{A}_g$ for $g \geq 3$
2. Bad Reduction Treatment: Verify Conjectures 6.1 and 6.2 to remove the good reduction assumption
3. Supersingular Proximity Conjecture: Study Conjecture 5.3 to fully resolve the Zilber-Pink problem

In-Depth Evaluation

Strengths

1. Theoretical Depth: Provides important generalizations and a unified framework for G-function theory
2. Technical Innovation: Cleverly utilizes $p$-adic Hodge theory to simplify prime dependence
3. Application Value: Provides new effective tools for the Zilber-Pink conjecture
4. Rigor: Theoretical constructions are rigorous and proofs are detailed

Weaknesses

1. Assumption Restrictions: The everywhere potential good reduction assumption is quite strong
2. Computational Dependence: Key non-triviality verifications rely on computer assistance
3. Completeness: Certain cases still require conjectural support

Impact

1. Academic Contribution: Provides important tools for unlikely intersection theory
2. Methodology: Demonstrates the power of G-function methods in modern arithmetic geometry
3. Subsequent Research: Establishes foundations for subsequent work in the series

Applicable Scenarios

This method is particularly suitable for:

• Unlikely intersection problems in Shimura varieties
• Special point counting in moduli spaces of abelian varieties
• Diophantine problems requiring effective height bounds

References

The paper cites 75 related references, primarily including:

• Foundational G-function theory by André, Bombieri, and others
• Unlikely intersection work by Daw-Orr
• Crystalline cohomology theory by Berthelot-Ogus
• $p$-adic Hodge theory by Hyodo-Kato

This paper represents an important advance in the application of G-function theory to modern arithmetic geometry, providing new effective tools for resolving the Zilber-Pink conjecture, with significant theoretical value and promising applications.