arXiv++ Combinatorics

We develop a geometric lifting method for inverse sumset problems in compact connected abelian groups. The first result is an analogue of Freiman's $3k-4$ theorem, that is every compact set $A\subseteq G$ of sufficiently small Haar measure satisfying $μ_G(A+A)<3μ_G(A)$ is contained in a one dimensional Bohr set of measure at most $μ_G(A+A)-μ_G(A)$. This resolves a question of Christ and Iliopoulou. The proof combines Bilu's theorem with a geometric refinement of the spillover argument. We also establish a sharp projection theorem. Under a continuous surjective homomorphism with connected kernel, a compact set of sufficiently small positive measure and doubling at most $K$, where $2\le K<3$, has image of doubling at most $2K-2$, and this factor is best possible. Further consequences include variants of the $3k-4$ theorem for popular sumsets and an inverse theorem for Tao's convolution inequality.

Eberhard and Pohoata conjectured that every $3$-cube-free subset of $[N]$ has size less than $2N/3+o(N)$. In this paper we show that if we replace $[N]$ with $\mathbb{Z}_N$ the upper bound of $2N/3$ holds, and the bound is tight when $N$ is divisible by $3$ since we have $A=\{a\in \mathbb{Z}_N:a\equiv 1,2\pmod{3}\}.$ Inspired by this observation we conjecture that every $d$-cube-free subset of $\mathbb{Z}_N$ has size less than $(d-1)N/d$ where $N$ is divisible by $d$, and we show the tightness of this bound by providing an example $B=\{b\in\mathbb{Z}_N:b\equiv 1,2,\ldots,d-1\pmod{d}\}$. We prove the conjecture for several interesting cases, including when $d$ is the smallest prime factor of $N$, or when $N$ is a prime power. We also discuss some related issues regarding $\{x,dx\}$-free sets and $\{x,2x,\ldots,dx\}$-free sets. A main ingredient we apply is to arrange all the integers into some square matrix, with $m=d^s\times l$ having the coordinate $(s+1,l-\lfloor l/d\rfloor)$. Here $d$ is a given integer and $l$ is not divisible by $d$.