Optimal exponent in arithmetic k-cycle removal

Status partial medium confidence

Question 1.4 asks for the exact optimal exponent $C_{p,k}$ in Theorem 1.3. The source paper achieves $C_{p,k} = \frac{k-2}{c_{p,3}}+1$ and shows the lower bound is at least $\frac{k-2}{c_{p,k}}+1$, leaving a gap of $O(\log p)$. A direct follow-up by L.M. Lovász (arXiv:1804.08837, 2019) tightens the k-colored sum-free bound in $\mathbb{Z}_m^n$—showing the lower-bound construction is essentially sharp—thereby pinning down $c_{p,k}$ precisely and sharpening the lower bound on $C_{p,k}$. However, closing the remaining gap between the upper bound (which uses $c_{p,3}$) and the now-exact lower bound (using $c_{p,k}$) remains open.

Cited literature (1)

• partial A lower bound for the $k$-multicolored sum-free problem in $\mathbb{Z}^n_m$ (2019)
  László Miklós Lovász · Proceedings of the London Mathematical Society · arXiv:1804.08837 · doi:10.1112/plms.12223

  Proves a lower bound for k-colored sum-free sets in $\mathbb{Z}_m^n$ that matches (up to lower-order terms) the known upper bound when $m$ is a prime power, thereby determining $c_{p,k}$ and sharpening the lower bound on the exponent $C_{p,k}$ from Question 1.4.

Reviewer notes. arXiv:1804.08837 (Lovász 2019) makes concrete progress by tightening the k-colored sum-free bound, which determines the lower bound on $C_{p,k}$. The precise value of $C_{p,k}$ (closing the gap between the $c_{p,3}$-based upper bound and the $c_{p,k}$-based lower bound) remains open as of the search date. No further post-2019 papers resolving the full question were found within the 5-call cap.

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