Abstract:In this paper, we prove a uniform version of Poonen's "Mordell-Lang Plus Bogomolov" theorem for abelian varieties. We mainly generalize Rémond's work on large points to allow an extra $\epsilon$-neighborhood. The part on small points follows from an earlier paper, joint with Gao and Kühne.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to prove a unified version of the Mordell - Lang conjecture and the Bogomolov conjecture, namely the so - called "Mordell - Lang Plus Bogomolov" theorem. Specifically, the author aims to generalize Rémound's work on large points by Vojta's method and allow an additional \(\epsilon\)-neighborhood, so as to obtain more general results. In addition, for the part of small points, the argument is based on previous research results.
The core problem of the paper can be summarized as follows: for an irreducible sub - variety \(X\) on an abelian variety \(A\) and a finitely generated subgroup \(\Gamma\), can we find an \(\epsilon\) independent of the choice of \(\Gamma\) such that the intersection \(X(\overline{\mathbb{Q}})\cap\Gamma'_\epsilon\) within the \(\epsilon\)-neighborhood is not Zariski - dense? Further, the author also hopes to prove that the number of rational points of this intersection can be bounded by a constant depending on \(X\), \(A\), their dimensions and degrees.
Expressed in a formula, the key result is:
\[ \#(X^\circ(\overline{\mathbb{Q}})\cap\Gamma'_\epsilon)\leq c^{1 + \rho} \]
where \(c = c(r, g, d)\), \(\rho\) is the rank of \(\Gamma\), and \(\epsilon=\epsilon(r, g, d)\).
The importance of this problem lies in the fact that it not only covers the classical Mordell - Lang conjecture and the Bogomolov conjecture, but also provides a more unified and extensive result framework. This helps to better understand the distribution characteristics of algebraic points on abelian varieties and provides a theoretical basis for further research.