Answers are grounded in this paper and cite their sources.
A pure O-sequence is the h-vector of an Artinian monomial level algebra. Equivalently, it counts, degree by degree, the monomials that divide at least one member of a fixed set of monomials of degree e, the socle degree. Migliore, Nagel and Zanello (2012) asked, for a fixed number r of variables, for the largest socle degree e such that every pure O-sequence is unimodal (Problem 3 of their survey). For r = 4 the known bounds were 4 ≤ e ≤ 6: Boyle (2012) proved unimodality through socle degree 4 and constructed non-unimodal examples in every socle degree e ≥ 7. This paper settles the two remaining cases. Every pure O-sequence in four variables of socle degree at most 6 is unimodal, so the answer to Problem 3 for r = 4 is exactly 6. In addition, every pure O-sequence in five variables of socle degree 4 is unimodal. Hence the smallest number of variables admitting a non-unimodal pure O-sequence of socle degree 4 (Problem 4 of the survey) is at least 6. The proofs are computer-assisted. Socle degree 5 in four variables, and socle degree 4 in five variables, are settled by checking all 2^35 candidate layers in degree e − 1. Socle degree 6 is reduced to two extremal quantities of the degree-5 layer, which are computed exactly by dynamic programmes over a splitting by one variable. The bounds used in both proofs are attained, by the pure O-sequences (1, 4, 8, 9, 10, 11) and (1, 4, 10, 17, 19, 19, 20).
No takes yet. Share an insight, caveat, or question.
William Huanshan Chuang (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: