Cohomological rank functions and syzygies of abelian varieties (Q2118184)

Let \(X\) be a smooth projective variety, and take \(L\) a very ample line bundle on \(X\). A very natural question one can ask is how ``nice'' is the projective embedding induced by \(L\). What is the maximum degree of the equations defining \(X\)? What is the degree of the relations between these equations? And the degree of the relations between the relations? We say that the polarized variety \((X,L)\) satisfies property \((N_p)\) if these degree are as small as possible for the first \(p\) steps of the graded free minimal resolutions of the algebra \(R_L:=\bigoplus_mH^0(X,L^{\otimes m})\) over the polynomial ring \(S_L:=\operatorname{Sym}H^0(X,L)\). In particular \(N_0\) means that the embedding is projectively normal, \(N_1\) means that \(X\) is defined by quadric equations, \(N_2\) is equivalent that the relations between the quadric equation of \(X\) are linear and so on. This paper is concerned with the study of property \((N_p)\) when \(X\) is an abelian variety and \(L\) is a primitive polarization, that is when \(L\) is not a power of an ample line bundle. The main tool is the cohomological rank function -- introduced in [\textit{Z. Jiang} and \textit{G. Pareschi}, Ann. Sci. Éc. Norm. Supér. (4) 53, No. 4, 815--846 (2020; Zbl 1459.14006)] by elaborating on ideas developed in [\textit{M. Á. Barja} et al., J. Inst. Math. Jussieu 19, No. 6, 2087--2125 (2020; Zbl 1452.14005)] -- which was first applied to the study of syzygies in [\textit{F. Caucci}, Algebra Number Theory 14, No. 4, 947--960 (2020; Zbl 1442.14140)]. The main results of these paper are progresses toward a conjecture of \textit{A. Ito} [Commun. Algebra 46, No. 12, 5342--5347 (2018; Zbl 1411.14016)] and \textit{V. Lozovanu} [``Singular divisors and syzygies of polarized abelian threefolds''. Preprint, \url{arXiv:1803.08780}], and the proof of a conjectur of Caucci [loc. cit.] in the case of a very general polarized abelian variety.