Monomial congruence counting conjecture

Let a,b,ca,b,c be non-zero integers with a,b,ca,b,c setwise coprime, and let d1d\geqslant 1 be another integer. For a prime q3q\geqslant 3, define Na,b,c,d(L,M,N;q)N_{a,b,c,d}(L,M,N;q) to count triples (l,m,n)L×M×N(l,m,n)\sim L\times M\times N satisfying

(lambnc)d1(modq),(l^a m^b n^c)^d\equiv 1\pmod q,

but with lambnc1l^a m^b n^c\ne 1. Monomial congruence counting conjecture. There exists η0>0\eta_0>0, depending on a,b,c,da,b,c,d but not on qq, such that for every L,M,N1L,M,N\geqslant 1 and every ε>0\varepsilon>0,

Na,b,c,d(L,M,N;q)LMNε,a,b,c,dqε(LMNq+(LMN)η0).\frac{N_{a,b,c,d}(L,M,N;q)}{\sqrt{LMN}}\ll_{\varepsilon,a,b,c,d}q^\varepsilon\left(\frac{\sqrt{LMN}}{q}+(LMN)^{-\eta_0}\right).

This estimate is the conjectural input needed to control error terms in the cubic moment analysis and would establish the expected asymptotic lower-bound equality in the general case. The supplied text does not state whether the conjecture has been proved or refuted.

Sources & referencesView supporting material

Primary source

Étienne Fouvry, Emmanuel Kowalski, Philippe Michel and Will Sawin, “Toroidal families and averages of L-functions, II: cubic moments”, arXiv:2603.10746 (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.