A palindromicity criterion for the h h -polynomials of bipartite edge rings
Yuta Hatasa · 2026
Read A palindromicity criterion for the h h -polynomials of bipartite edge rings by Yuta Hatasa (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-26865.txt.
Introduction
Hatasa studies symmetry in the h-polynomials of edge rings of bipartite graphs. He proves that if the ring is pseudo-Gorenstein and the first and next-to-last coefficients match, then the ring is Gorenstein, so its h-polynomial reads the same forwards and backwards. The proof handles the 2-connected case via a tight-separation condition, then extends to all bipartite graphs by block decomposition, also building a minimal Gorenstein closure.
Catalog record
| CREATOR | Yuta Hatasa |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 33559 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.26865
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
A palindromicity criterion for the h h -polynomials of bipartite edge rings Yuta Hatasa Abstract . We study a symmetry problem for the h h -polynomials of edge rings of bipartite graphs. Let G G be a bipartite graph and write h ( 𝕜 [ G ] , t ) = h 0 + h 1 t + ⋯ + h s t s h(\Bbbk[G];t)=h_{0}+h_{1}t+\cdots+h_{s}t^{s} . We prove that if 𝕜 [ G ] \Bbbk[G] is pseudo-Gorenstein and h 1 = h s − 1 h_{1}=h_{s-1} , then 𝕜 [ G ] \Bbbk[G] is Gorenstein. Equivalently, under these assumptions the h h -polynomial of 𝕜 [ G ] \Bbbk[G] is palindromic. The proof treats the 2 2 -connected case first by translating the numerical condition h 1 = h s − 1 h_{1}=h_{s-1} into a tight-separation condition for non-edges, and then passes to arbitrary bipartite graphs using the block decomposition. We also construct a blockwise minimal Gorenstein closure, obtained by adjoining all non-edges not separated by tight acceptable sets, and show that this construction preserves the next-to-leading coefficient of the h h -polynomial. 2020 Mathematics Subject Classification Primary 13F65; Secondary 13D40, 05C25, 52B20 1. Introduction Let R=\bigoplus_{k\geq 0}R_{k} be a Cohen–Macaulay homogeneous domain of dimension d over a field R_{0}=\Bbbk . The Hilbert series of R has the form \sum_{k\geq 0}(\dim_{\Bbbk}R_{k})t^{k}=\frac{h_{0}+h_{1}t+\cdots+h_{s}t^{s}}{(1-t)^{d}},\quad h_{s}\neq 0. The numerator h(R;t)=h_{0}+h_{1}t+\cdots+h_{s}t^{s} is called the h -polynomial of R . The ring R is called pseudo-Gorenstein if h_{s}=1 . When R is a Cohen–Macaulay homogeneous domain, Stanley’s theorem states that R is Gorenstein if and only if the h -vector is symmetric, that is, h_{i}=h_{s-i} for all i [ 5 ] . We study this question for edge rings of bipartite graphs. For a finite simple graph G on the vertex set V(G)=[d] , let E(G) denote its edge set. The edge ring of G is \Bbbk[G]=\Bbbk[t_{i}t_{j}:\{i,j\}\in E(G)]\subset\Bbbk[t_{1},\ldots,t_{d}]. The edge rings of bipartite graphs are normal, hence Cohen–Macaulay, by the theorem of Ohsugi–Hibi and Simis–Vasconcelos–Villarreal [ 2 , 4 ] . For bipartite graphs, Hatasa, Kowaki, and Matsushita proved that \Bbbk[G] is pseudo-Gorenstein if and only if every block of G is matching-covered [ 1 ] . In particular, if G is 2 -connected and bipartite, the pseudo-Gorenstein property is equivalent to G being matching-covered. The natural next question is whether a small amount of additional symmetry of the h -vector already forces full symmetry. Our main result gives an affirmative answer for bipartite graphs. Theorem 1.1 . Let G be a bipartite graph and write h(\Bbbk[G];t)=h_{0}+h_{1}t+\cdots+h_{s}t^{s}. If h_{s}=1 and h_{1}=h_{s-1} , then \Bbbk[G] is Gorenstein. In particular, h(\Bbbk[G];t) is palindromic. The same method also gives a blockwise Gorenstein closure for pseudo-Gorenstein bipartite edge rings. Theorem 1.2 . Let G be a bipartite graph and write h(\Bbbk[G];t)=h_{0}+h_{1}t+\cdots+h_{s}t^{s}. Suppose that h_{s}=1 . For each block B of G , choose a bipartition X_{B}\sqcup Y_{B} . Let \Fill(B) be the set of non-edges \{x,y\} of B , with x\in X_{B} and y\in Y_{B} , that are not separated by any tight acceptable set T\subset X_{B} , in the sense that no such T satisfies x\in T,\qquad y\notin N_{B}(T). For each block B , let \widehat{B} be the graph on the same vertex set as B with E(\widehat{B})=E(B)\cup\Fill(B). Let \widehat{G} be the graph obtained from G by replacing each block B by \widehat{B} . Then the following assertions hold. (i) \Bbbk[\widehat{G}] is Gorenstein. (ii) Writing h(\Bbbk[\widehat{G}];t)=\widehat{h}_{0}+\widehat{h}_{1}t+\cdots+\widehat{h}_{\widehat{s}}t^{\widehat{s}}, one has \widehat{s}=s and \widehat{h}_{s-1}=h_{s-1} . (iii) The construction is blockwise minimal: if B is a block of G with at least one edge and K_{B} is a bipartite graph with bipartition X_{B}\sqcup Y_{B} such that E(B)\subset E(K_{B}) and \Bbbk[K_{B}] is Gorenstein, then E(\widehat{B})\subset E(K_{B}). The proof first handles the 2 -connected case. In that case, we express the difference h_{s-1}-h_{1} as the number of specific interior lattice points of the edge polytope arising from non-edges of G . Thus h_{1}=h_{s-1} is equivalent to the assertion that every non-edge is separated by a tight acceptable set on the X -side, and symmetrically on the Y -side. We then prove a purely graph-theoretic statement: in a 2 -connected matching-covered bipartite graph, this X -side separation condition is equivalent to the tightness of every acceptable set. By the Ohsugi–Hibi criterion, this is exactly the Gorenstein condition. Finally, for a bipartite graph, the edge ring decomposes as a tensor product over its blocks, so its h -polynomial is the product of the h -polynomials of the blocks. This reduces the general bipartite case to the 2 -connected one. 2. Preliminaries 2.1. Edge polytopes and facets Throughout this paper, all graphs are finite and simple. Let G be a graph on the vertex set V(G)=[d] . For an edge e=\{i,j\} , set \rho(e)=\mathbf{e}_{i}+\mathbf{e}_{j}\in\mathbb{R}^{d} . The edge polytope of G is P_{G}=\conv\{\rho(e):e\in E(G)\}\subset\mathbb{R}^{d}. If G is connected and bipartite, then \dim P_{G}=d-2 , and \dim\Bbbk[G]=d-1 . Moreover, since \Bbbk[G] is normal, the Hilbert function of \Bbbk[G] is the Ehrhart function of P_{G} : \dim_{\Bbbk}\Bbbk[G]_{k}=\lvert kP_{G}\cap\mathbb{Z}^{d}\rvert. Let G be bipartite with bipartition V(G)=X\sqcup Y . For T\subset X , set N_{G}(T)=\left\{y\in Y:\{x,y\}\in E(G)\text{ for some }x\in T\right\},\qquad\delta_{G}(T)=\lvert N_{G}(T)\rvert-\lvert T\rvert. We use the same notation for subsets of Y . Following Ohsugi–Hibi [ 2 ] , a non-empty set T\subset X is called acceptable if the bipartite graph B(T) induced by the edges between T and N_{G}(T) is connected and the induced graph on (X\setminus T)\sqcup(Y\setminus N_{G}(T)) is connected with at least one edge. Acceptable subsets of Y are defined in the same way. In this paper, we call a subset T tight if \delta_{G}(T)=1 . We also use the terminology of Ohsugi–Hibi [ 2 ] : a vertex v of a connected graph is called ordinary if G\setminus v , the induced graph obtained by deleting v , is connected. With this terminology, a connected graph is called 2 -connected if all of its vertices are ordinary. The facet description of Ohsugi–Hibi states that, for a connected bipartite graph, the coordinate facets of P_{G} are precisely the facets supported by z_{v}=0 for ordinary vertices v . The remaining facets are supported by the hyperplanes \sum_{y\in N_{G}(T)}z_{y}-\sum_{x\in T}z_{x}=0 for acceptable sets T\subset X ; equivalently, one may use the symmetric formulation with acceptable subsets of Y . In particular, if G is 2 -connected, then every vertex is ordinary, so every coordinate hyperplane z_{v}=0 supports a facet of P_{G} . 2.2. Blocks A block of G is a maximal 2 -connected subgraph. We use the following decomposition of bipartite edge rings along blocks. Proposition 2.1 ( [ 6 , Proposition 10.1.48] ) . Let G be a bipartite graph, and let B_{1},\ldots,B_{m} be the blocks of G . Then \Bbbk[G]\cong\Bbbk[B_{1}]\otimes_{\Bbbk}\cdots\otimes_{\Bbbk}\Bbbk[B_{m}] as standard graded \Bbbk -algebras. In particular, h(\Bbbk[G];t)=\prod_{i=1}^{m}h(\Bbbk[B_{i}];t). 2.3. Matching-covered bipartite graphs A connected graph G is matching-covered if every edge of G is contained in a perfect matching. The proof of the main theorem uses the following elementary strict Hall property for the 2 -connected matching-covered blocks that occur there. Lemma 2.2 . Let G be a 2 -connected matching-covered bipartite graph with bipartition X\sqcup Y . Then \lvert X\rvert=\lvert Y\rvert and, for every non-empty proper subset A\subsetneq X , \lvert N_{G}(A)\rvert\geq\lvert