The \(h\)-vector of a Gorenstein toric ring of a compressed polytope (Q2570995): Difference between revisions

An integral (convex) polytope is called compressed if all its triangulations obtained by pulling vertices are unimodular. If \(\mathcal P\) is a polytope, then a \((q-1)\)-simplex \(\Sigma\) whose vertices are vertices of \(\mathcal P\) is special if each facet of \(\mathcal P\) contains exactly \(q-1\) vertices of \(\Sigma\). It was shown by \textit{C. A. Athanasiadis} [J.~Reine Angew.~Math. 583, 163--174 (2005; Zbl 1077.52011)] (who introduced the concept) that, if \(\mathcal P\) is compressed and has a special simplex, then its toric ring \(K[\mathcal P]\) is Gorenstein and has a unimodal \(h\)-vector. In this paper, the authors prove the converse: if \(\mathcal P\) is compressed and \(K[\mathcal P]\) is Gorenstein, then \(\mathcal P\) has a special simplex.