Pink's conjecture on unlikely intersections and families of semi-abelian varieties (Q2184861)

This paper concerns the Conjecture 1.3 in \textit{R.~Pink}'s 2005 preprint [``A common generalization of the conjectures of André-Oort, Manin-Mumford, and Mordell-Lang''], which states that if \(Y\) is a closed irreducible Hodge generic subvariety of a mixed Shimura variety \(S\), then \[ Y\cap S^{\geqslant\dim(Y)+1} \] is not Zariski dense in \(Y\), where \(S^{\geqslant\dim(Y)+1}\) denotes the union of special subvarieties of \(S\) of codimension at least \(\dim(Y)+1\). In the same preprint [loc. cit.], Pink also states a relative version of the Manin-Mumford conjecture for families of semi-abelian varieties, namely, the Conjecture 6.1: if \(B\to X\) is a family of semi-abelian varieties over \(\mathbb{C}\) and \(Y\) is a closed irreducible subvariety of \(B\) that is not contained in any proper closed subgroup scheme of \(B\to X\), then \[ Y\cap B^{\geqslant\dim(Y)+1} \] is not Zariski dense in \(Y\), where \(B^{\geqslant\dim(Y)+1}\) denotes the union of algebraic subgroups of codimension at least \(\dim(Y)+1\) of the fibers of \(B\to X\). Pink claims that the Conjecture 1.3 implies the Conjecture 6.1 (the Theorem 6.3 of [loc. cit.]). But \textit{D. Bertrand} found counterexamples against the Conjecture 6.1 in [``Special points and Poincaré bi-extensions'', Preprint, \url{arXiv:1104.5178}]. In this paper, the authors sharpen the results in [loc. cit.] and explain that the counterexamples are not against the Conjecture 1.3; in fact, they provide evidence for Conjecture 1.3. The authors also clarify what is wrong in the proof of the Theorem 6.3 of [Pink, loc. cit.]. The counterexamples are provided by the Ribet sections, which are constructed under various denominations by the authors. The paper also provide a nice self-contained example-based introduction to mixed Hodge structures and mixed Shimura varieties (the Section 4).