The cyclotomic-integer conductor-length conjecture

Let α\alpha be a cyclotomic integer, meaning an algebraic integer expressible as a sum of roots of unity. Let l(α)l(\alpha) be the smallest number of roots of unity occurring in any such representation, and let c(α)c(\alpha) be the smallest positive integer such that Q(α)Qc(α)\mathbb{Q}(\alpha)\subseteq\mathbb{Q}_{c(\alpha)}, where Qn\mathbb{Q}_n is the field generated by a primitive nnth root of unity. The conductor-length conjecture. One has

[Qc(α):Q(α)]l(α).[\mathbb{Q}_{c(\alpha)}:\mathbb{Q}(\alpha)] \leq l(\alpha).

This is the purely number-theoretic reformulation of the element-level question of Hung and Tiep. The paper's abstract states that the question is answered negatively and that all counterexamples with length at most 44 are characterized; it also studies the growth of the degree as a function of the length. Thus this conjecture is refuted.

Sources & referencesView supporting material

Primary source

Christopher Herbig, “Answer to a Question of Hung and Tiep on Conductors of Cyclotomic Integers”, arXiv:2508.16732 (2025).

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.