← All problems

Van der Waerden-type conjecture for Galois groups of cubic surfaces

Let Nexc(H)N_{\mathrm{exc}}(H) denote the number of smooth homogeneous cubic forms F∈Z[x0,x1,x2,x3]F\in\mathbb{Z}[x_0,x_1,x_2,x_3] with all coefficients of absolute value at most HH such that, for the cubic surface SF⊂PQ3S_F\subset\mathbb{P}^3_{\mathbb{Q}} defined by F=0F=0, the Galois image GSF⊆W(E6)G_{S_F}\subseteq W(E_6) acting on the 2727 lines satisfies GSF≠W(E6)G_{S_F}\ne W(E_6). The van der Waerden-type problem asks whether, for every ε>0\varepsilon>0, one has Nexc(H)=Oε(H19+ε)N_{\mathrm{exc}}(H)=O_{\varepsilon}(H^{19+\varepsilon}) as H→∞H\to\infty.

Equivalent formulations
  1. With Nexc(H)N_{\mathrm{exc}}(H) defined as above, O'Dorney conjectures that Nexc(H)=H16+o(1)N_{\mathrm{exc}}(H)=H^{16+o(1)} as H→∞H\to\infty.

    source: Van der Waerden's Problem for Cubic Surfaces

References

Source: arXiv

Progress summary

Refreshed
Open · progress reported

A new paper substantially improves the counting estimates for exceptional cubic surfaces, but the main conjectured bound remains open because of one unresolved discriminant obstruction.

The conjecture predicts a van der Waerden-type exceptional-set bound of O(H19+ε)O(H^{19+\varepsilon}) for Galois groups arising from cubic surfaces. The central unresolved issue is controlling surfaces whose Galois group lies in the index-two subgroup of W(E6)W(E_6).

Known results

  • Elsenhans and Jahnel construct cubic surfaces with Galois group equal to the index-two subgroup of W(E6)W(E_6), of order 25 92025\,920.
  • For their surfaces, membership in that subgroup is equivalent to (−3)Δ(a0,…,a4)(-3)\Delta(a_0,\ldots,a_4) being a square in Q\mathbb{Q}.
  • They derive square-divisibility constraints on the normalized parameters, including even pp-adic valuations for primes p≡2(mod3)p\equiv 2\pmod 3.

October 2026 improved estimates

Evan O'Dorney's paper claims sharper O(H19)O(H^{19}) bounds for five maximal-subgroup classes and identifies the index-two subgroup as the principal obstruction to the expected O(H19+ε)O(H^{19+\varepsilon}) estimate. The paper also proposes, but does not prove, H16+o(1)H^{16+o(1)} asymptotics for the exceptional count.

Current status (as of October 2026): The general conjectured O(H19+ε)O(H^{19+\varepsilon}) bound remains open; claimed progress covers five subgroup classes, while the index-two discriminant case is unresolved.

Sources

Solutions 0

No solutions have been posted yet.