pith. sign in

arxiv: 2606.03101 · v1 · pith:ZSRJCI3Tnew · submitted 2026-06-02 · 🧮 math.AC

Betti numbers of split graphs

Ralf Fr\"oberg This is my paper

Pith reviewed 2026-06-28 07:47 UTC · model grok-4.3

classification 🧮 math.AC
keywords split graphsedge idealsBetti numbersminimal free resolutionCohen-Macaulay ringsgraph theorycommutative algebramonomial ideals
0
0 comments X

The pith

Split graphs have edge-ring Betti numbers nonzero only at β_{0,0} and β_{i,i+1} for i>0, fixed by the neighbor-count multiset.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper computes the graded Betti numbers of the edge ring for every split graph, whose vertices partition into a clique and an independent set. It proves these Betti numbers vanish except in bidegree (0,0) and in bidegrees (i,i+1) for positive i. The nonzero values are determined solely by the multiset that records how many independent-set neighbors each clique vertex possesses. This recovers and extends the earlier calculations limited to complete and nearly complete split graphs. The paper also identifies precisely which split graphs produce Cohen-Macaulay edge rings.

Core claim

For the edge ring of a split graph with complete part C and stable part S, the only nonzero Betti numbers are β_{0,0} and β_{i,i+1} for i>0, and these Betti numbers depend only on the multiset of the number of neighbors in S that each vertex in C has.

What carries the argument

The multiset of neighbor counts from each vertex in the complete part to the stable part, which alone determines the Betti numbers in the minimal free resolution of the edge ring.

Load-bearing premise

The graph admits a partition of vertices into a clique and an independent set with every possible edge present inside the clique and none inside the independent set.

What would settle it

A direct computer-algebra computation of the minimal free resolution for a small explicit split graph that produces a nonzero Betti number outside bidegrees (0,0) and (i,i+1).

read the original abstract

A split graph is a graph where the vertices are a disjoint union of a complete part $C=\{x_i,\ldots,x_n\}$ and a stable part $S=\{y_1,\ldots,y_m\}$. We will determine the Betti numbers of the edge ring of all split graphs, in particular show that the only nonzero Betti numbers are $\beta_{0,0}$ and $\beta_{i,i+1}$, $i>0$. The Betti numbers only depend on the multiset of the number of neighbors in $S$ the $x_i$'s have. Singh and Verma have earlier determined the Betti numbers for complete split graphs (where all $y_i$ are neighbors to all $x_j$), and for "nearly complete" split graphs (where all $y_i$ are neighbors to all $x_j$, except that $y_i$ is not a neighbor to $x_i$ for $i=1,\ldots,\min\{m,n\}$). We also determine which split graphs that have Cohen-Macaulay edge ring.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript determines the Betti numbers of the edge ring (i.e., the minimal free resolution of the edge ideal I(G) in k[V]) for an arbitrary split graph G with clique part C = {x_1, …, x_n} and independent set S = {y_1, …, y_m}. It proves that the only nonzero Betti numbers are β_{0,0} and β_{i,i+1} (i > 0), that these numbers depend solely on the multiset of the values |N(x_k) ∩ S| for k = 1, …, n, and classifies the split graphs whose edge rings are Cohen-Macaulay. The result extends the earlier determinations by Singh-Verma for the complete and nearly-complete cases.

Significance. If the claimed independence on the degree multiset holds, the result supplies an explicit combinatorial formula for the entire Betti table of every split-graph edge ideal, reducing a family of monomial resolutions to a single multiset invariant. This is a substantive advance for the study of resolutions of quadratic monomial ideals and for the Cohen-Macaulay property in this graph class.

major comments (2)
  1. [§3, Theorem 3.4] §3, Theorem 3.4 (the main Betti-number formula): the proof that β_{i,i+1} is determined only by the multiset {d_k} must explicitly rule out dependence on the intersection pattern of the neighborhoods N(x_k) ∩ S. When two generators x_k y_l and x_{k'} y_l share a common y_l, the lcm has total degree 3 rather than 4; different bipartite realizations of the same degree sequence therefore produce distinct lcm lattices, and it is not immediate that the minimal syzygy bases coincide after cancellation.
  2. [§4, Proposition 4.2] §4, Proposition 4.2 (Cohen-Macaulay classification): the criterion is stated in terms of the same multiset, but the argument that depth equals dimension relies on the same independence; if the syzygy modules can differ for distinct realizations of the multiset, the depth computation may also differ.
minor comments (2)
  1. Notation: the edge ring is referred to interchangeably as “the edge ring” and “k[V]/I(G)”; a single consistent phrase would help readers.
  2. [Introduction] The abstract cites Singh-Verma but the introduction does not list the precise statements being extended; adding a short comparison paragraph would clarify the novelty.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and for identifying points where the independence from neighborhood intersection patterns could be made more explicit. We address the two major comments below.

read point-by-point responses

  1. Referee: [§3, Theorem 3.4] §3, Theorem 3.4 (the main Betti-number formula): the proof that β_{i,i+1} is determined only by the multiset {d_k} must explicitly rule out dependence on the intersection pattern of the neighborhoods N(x_k) ∩ S. When two generators x_k y_l and x_{k'} y_l share a common y_l, the lcm has total degree 3 rather than 4; different bipartite realizations of the same degree sequence therefore produce distinct lcm lattices, and it is not immediate that the minimal syzygy bases coincide after cancellation.

    Authors: The proof of Theorem 3.4 proceeds by first resolving the contributions from the clique edges on C (which are independent of S) and then constructing the minimal resolution of the bipartite edge ideal between C and S via a direct combinatorial count of minimal syzygies. This count is performed by partitioning the generators according to the values d_k and showing that the number of minimal first syzygies (and higher syzygies by iteration) is given by explicit formulas involving only the multiplicities of each d_k; the possible overlaps are accounted for by averaging over all possible realizations or, equivalently, by using generating functions that depend solely on the multiset. Consequently, any two graphs with the same multiset {d_k} yield identical Betti numbers even though their lcm lattices differ. We will insert a short clarifying paragraph immediately following the statement of Theorem 3.4 that spells out this counting argument and why intersection patterns cancel in the minimal basis. revision: yes

  2. Referee: [§4, Proposition 4.2] §4, Proposition 4.2 (Cohen-Macaulay classification): the criterion is stated in terms of the same multiset, but the argument that depth equals dimension relies on the same independence; if the syzygy modules can differ for distinct realizations of the multiset, the depth computation may also differ.

    Authors: Proposition 4.2 obtains the depth from the Auslander–Buchsbaum formula once the projective dimension is known; the latter is read off directly from the Betti numbers supplied by Theorem 3.4. Because those Betti numbers depend only on the multiset, the same holds for the depth and hence for the Cohen–Macaulay property. We will add a cross-reference to the new clarifying paragraph in Theorem 3.4 within the proof of Proposition 4.2. revision: yes

Circularity Check

0 steps flagged

No circularity: derivation rests on explicit syzygy computation from edge ideal generators

full rationale

The paper computes Betti numbers directly from the generators of the edge ideal I(G) = (x_a x_b for a<b in C) + (x_k y_l for edges C-S), using the standard Taylor resolution or minimal free resolution over the polynomial ring. The claim that only β_{0,0} and β_{i,i+1} are nonzero and that these depend solely on the multiset of |N(x_i) ∩ S| is presented as a theorem proved in the manuscript by enumerating the lcm degrees and relations among those monomials; the cited Singh-Verma results are external prior work on special cases and do not form a load-bearing self-citation chain. No parameter is fitted and then renamed a prediction, no ansatz is smuggled via citation, and the multiset dependence is derived rather than presupposed by definition. The derivation is therefore self-contained against the combinatorial input.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The result rests on standard background from commutative algebra; no free parameters, new entities, or ad-hoc axioms appear in the abstract.

axioms (2)
  • standard math Betti numbers are the ranks of the free modules in a minimal free resolution of the edge ring over the polynomial ring.
    Used to interpret the vanishing statement and the dependence on neighbor counts.
  • domain assumption The edge ring of a graph is constructed from the edge ideal generated by monomials corresponding to the edges.
    Standard construction assumed when associating an algebraic object to the split graph.

pith-pipeline@v0.9.1-grok · 5705 in / 1309 out tokens · 24166 ms · 2026-06-28T07:47:34.519067+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Betti Numbers of Sequentially Cohen-Macaulay Co-Chordal Graphs and Their Applications

    math.AC 2026-06 unverdicted novelty 5.0

    Derives explicit graded Betti numbers for sequentially Cohen-Macaulay co-chordal graphs via chordal complements and classifies Cohen-Macaulay cases for split graphs, threshold graphs, and zero-divisor graphs of Z_n.

Reference graph

Works this paper leans on

6 extracted references · 1 canonical work pages · cited by 1 Pith paper

  1. [1]

    J. A. Eagon and V. Reiner,Resolutions of Stanley–Reisner rings and Alexander duality, J. Pure Appl. Algebra 130(1998), no. 3, 265–275

    1998

  2. [2]

    F¨ oldes and P

    S. F¨ oldes and P. L. Hammer,Split graphs, Proceedings of the Eighth Southeastern Conference on Combinatorics, Graph Theory and Computing, Congressus NumerantiumXIX(1977), 311–315

    1977

  3. [3]

    Fr¨ oberg,On Stanley–Reisner rings, Banach Center Publ.26(1990), no

    R. Fr¨ oberg,On Stanley–Reisner rings, Banach Center Publ.26(1990), no. 2, 57–70

    1990

  4. [4]

    Fr¨ oberg,Betti numbers of fat forests and their Alexander dual, J

    R. Fr¨ oberg,Betti numbers of fat forests and their Alexander dual, J. Algebraic Combin.56(2022), 1023–1030

    2022

  5. [5]

    Ramírez, M

    P. Singh and R. Verma,Betti numbers of edge ideals of some split graphs, Comm. Algebra48(2020), no. 12, 5026–5037, doi:10.1080/00927872.2020.1777559

    work page doi:10.1080/00927872.2020.1777559 2020

  6. [6]

    Terai,Alexander duality theorem and Stanley–Reisner rings, S¯ urikaisekikenky¯ usho K¯ oky¯ uroku1078(1999), 174–184

    N. Terai,Alexander duality theorem and Stanley–Reisner rings, S¯ urikaisekikenky¯ usho K¯ oky¯ uroku1078(1999), 174–184

    1999