← All problems

Leary’s conjectures on groups of deck transformations

For every admissible Bestvina–Brady datum (L,M,S)(L,M,S) defining a group of deck transformations GLM(S)G_L^M(S), the group GLM(S)G_L^M(S) is virtually torsion-free and residually finite. Moreover, whenever π(M,L)\pi(M,L) is finite and S+n=SS+n=S, the semidirect product GLM(S)⋊nZG_L^M(S)\rtimes_n\mathbb{Z} is virtually special.

References

Progress summary

Refreshed
Open · solution reported

An October 2026 preprint claims to prove the conjectures in the specified Bestvina–Brady setting, but the claim has not been independently checked.

The conjectures concern finiteness and separability properties of groups of deck transformations, including groups GLM(S)G_L^M(S) associated with Bestvina–Brady constructions.

Known results

  • A 2022 study formulated criteria and conjectures for virtual torsion-freeness and residual finiteness of GLM(S)G_L^M(S), proving many special cases.
  • It also conjectured virtual specialness of GLM(S)⋊nZG_L^M(S)\rtimes_n\mathbb{Z} when π(M,L)\pi(M,L) is finite and S+n=SS+n=S.
  • The virtual-specialness conjecture would imply the other two through standard properties of virtually special groups.

October 2026 claimed proof

Vladimir Vankov’s unrefereed preprint Large profinite genus and extensions in Bestvina–Brady groups claims to prove the cited Leary conjectures for the relevant deck-transformation groups. It further claims that strong separability yields uncountable families with identical profinite completions but distinct quasi-isometry types, and introduces a finite congruence-extension property.

Current status (as of October 2026): The conjectures are claimed proved for the stated Bestvina–Brady setting, but the preprint is unrefereed and the claim remains unverified.

Sources

Solutions 0

No solutions have been posted yet.