Zombie number increase via subdivision

Question 4.2 · arXiv:2008.03587

arXiv Question medium confidence— first stated 2021-06-03

Status open high confidence

Question 4.2 asks whether for any graph G there exists an integer k such that z(G'_k) > z(G) + 1, where G'_k is obtained by subdividing every edge k times and re-adding the original edges; the authors suggest k = 5 may work universally. No follow-up paper addressing this specific question was found in indexed literature through 2026, across searches covering zombie-number graph theory, the source paper's citations, and related pursuit-evasion work on graph products and subdivisions.

Reviewer notes. No follow-up found. The three web searches and two successful WebFetch calls (arXiv abstract page and a 2022 Bose-De Carufel-Shermer pursuit-evasion paper on lazy zombies) returned no reference to Question 4.2 or the specific G'_k construction. The Manitoba 2024 thesis on zombie numbers across graph classes also does not appear to address this question. The conjecture remains open with high confidence.

Auto-reviewed 2026-05-15 with claude-sonnet-4-6 (web search enabled).

Question. For any graph $G$, is there an integer $k$ such that, for $G'_k$ the graph obtained from $G$ by subdividing all edges $k$ times then adding the original edges, $z(G'_k) > z(G) + 1$?

Context

This question probes how iterated edge subdivision combined with re-adding the original edges affects the zombie number. The authors remark that $k = 5$ may be a valid universal answer for any graph $G$.

Notes. PDF source — the inequality $z(G'_k) > z(G) + 1$ may be garbled; it could be $z(G'_k) \geq z(G) + 1$ (a strict increase of at least 1). The follow-up remark 'it could be that 5 is a valid answer to Question 4.2 in any graph' suggests the question may additionally seek a universal constant $k$.

Source paper

A note on deterministic zombies
Valentin Bartier, Laurine Bénéteau, Marthe Bonamy, Hoang La, Jonathan Narboni · 2021-06-03
https://arxiv.org/abs/2008.03587 PDF source