On Fujita's conjecture for a general hyperkähler manifold in the standard series of examples
Alessandro Pilastro
TL;DR
The paper advances Fujita-type questions for polarized hyperkähler manifolds of $K3^{[n]}$-type and $\mathrm{Kum}^n$-type by leveraging tautological bundles on Hilbert schemes, Grassmannian embeddings, and lattice-polarization theory. It develops explicit criteria for base point freeness and very ampleness of polarizations on general elements of the moduli spaces $\Sigma_{d,t}^n$ and $\Upsilon_{d,t}^n$, including sharp numerical bounds via the parameter $\tau=\frac{t^2}{2(t-1)}$. The work also establishes openness of these properties in families, enabling transfer of local results to generic elements within connected components, and provides necessary and sufficient non-emptiness conditions for the moduli spaces. Overall, it extends Fujita-type results to higher-dimensional hyperkähler manifolds, delivering concrete, computable thresholds and clarifying the polarization geometry in these moduli spaces.
Abstract
For the moduli spaces $Σ_{d,t}^n$ and $Υ_{d,t}^n$ of polarized hyperkähler manifolds of Hilb$^n$(K3)-type and Kum$^n$-type respectively, with polarization with square $2d$ and divisibility $t$, we study general base point freeness and very ampleness of the polarization. We provide cases where these moduli space are connected and a formula characterizing when these spaces are non-empty.
