Colour-balanced error bounds for k≥3 spanning forests

Problem 6.3 · arXiv:2604.09449

arXiv Problem high confidence— first stated 2026-04-10

Authors: Emma Hogan, Alex Scott, Dmitry Tsarev

Status open high confidence

Problem 6.3 asks to improve the leading coefficient in the linear-in-$\Delta$ error bound for $f_h(F)$ when $k \geq 3$ colours and $F$ is a bounded-degree spanning forest of $K_{kn}$; the paper leaves this optimisation entirely open. The paper was posted on 10 April 2026 and no follow-up addressing this specific problem was found in any indexed preprint or journal.

Reviewer notes. Paper verified via WebFetch of https://arxiv.org/abs/2604.09449. The paper is only ~5 weeks old; absence of follow-up is expected. No internal references were supplied for this problem.

Auto-reviewed 2026-05-14 with claude-sonnet-4-6 (web search enabled).

Problem. Improve the bounds on $f_{h}(F)$ for bounded-degree spanning forests $F$ in complete graphs when $k\geq 3$.

Context

For $k=2$, error bounds sharp up to an additive constant are known for spanning forests of maximum degree $\Delta$. For $k>2$ the paper shows the error remains linear in $\Delta$, but no attempt is made to optimise the leading coefficient.

Source paper

Colour-balanced subgraphs
Emma Hogan, Alex Scott, Dmitry Tsarev · 2026-04-10
https://arxiv.org/abs/2604.09449