The product-beating conjecture for non-degenerate genus-one systems

Let L=(L1,,Lt)L=(L_1,\ldots,L_t) be a non-degenerate genus-one translation-invariant system of equations over Fq\mathbb{F}_q, where each LjL_j is a translation-invariant equation in ss variables. A set AFqnA\subseteq\mathbb{F}_q^n is LL-free if the only simultaneous solutions of Lj(x1,,xs)=0L_j(x_1,\ldots,x_s)=0 for 1jt1\leq j\leq t in AA are diagonal solutions with x1==xsx_1=\cdots=x_s. The system has genus one when there is no nonempty proper subset I[s]I\subsetneq [s] such that iIci,j=0\sum_{i\in I}c_{i,j}=0 for every 1jt1\leq j\leq t, where Lj(x1,,xs)=i=1sci,jxiL_j(x_1,\ldots,x_s)=\sum_{i=1}^s c_{i,j}x_i. The product-beating conjecture. If AFqnA\subseteq\mathbb{F}_q^n is LL-free and has size cnc^n, then there exists an LL-free set BFqmB\subseteq\mathbb{F}_q^m for some m>nm>n such that

B>cm.|B|>c^m.

This would extend the corresponding product-beating result from a single translation-invariant equation to systems of equations; the non-degeneracy assumption is necessary because a degenerate system makes the whole space LL-free. The claim is presented as an expectation for systems and no resolution is supplied here.

Sources & referencesView supporting material

Primary source

Paul Hametner and Fred Tyrrell, “Beating Product Constructions for Linear Equations Over Finite Fields”, arXiv:2606.12194 (2026).

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.