Search papers, labs, and topics across Lattice.
The authors resolve the classification of quasi-$F$-splitting for good reductions of projective hyperkähler and primitive symplectic varieties of dimension $2n \geq 4$ in positive characteristic. They establish that quasi-$F$-splitting is strictly equivalent to ordinary Frobenius splitting—governed by the existence of a slope-zero part in $\operatorname{H}^2_{\operatorname{crys}}(X/W)[1/p]$—which forces the quasi-$F$-split height to be either $1$ or $\infty$. This result removes longstanding crystalline torsion-freeness assumptions to prove that Hodge-goodness is open in families and ensures unobstructed mixed-characteristic formal deformations for Hilbert schemes of $K3$ surfaces and generalized Kummer varieties.
Higher-dimensional hyperkähler varieties in characteristic $p$ exhibit an all-or-nothing splitting dichotomy: intermediate quasi-$F$-split heights are impossible, collapsing the property entirely to classical Frobenius splitting.
Let $X$ be the good reduction of a projective hyperkähler variety of dimension $2n\geq4$. We prove that $X$ is quasi-$F$-split if and only if it is Frobenius split, equivalently if $\operatorname{H}^2_{\operatorname{crys}}(X/W)[1/p]$ has a slope-zero part. Thus its quasi-$F$-split height is $1$ or $\infty$. The proof combines a Verbitsky slope comparison with a Witt--Euler identity and requires no crystalline torsion-freeness. The same dichotomy holds for primitive symplectic varieties in characteristic $p$, and Hodge-goodness is open in smooth proper families. Hodge-deformations of Hilbert schemes $S^{[n]}$ of $K3$ surfaces ($p>n$) and generalised Kummer varieties $K_n(A)$ ($p>n+1$) remain primitive symplectic, with torsion-free crystalline cohomology and unobstructed mixed-characteristic formal deformations.