← All problems

Dyer’s Conjecture A on extended weak order

For every Coxeter system (W,S)(W,S) with positive root system Φ+\Phi^+, the poset of biclosed subsets of Φ+\Phi^+, ordered by inclusion and identified with the extended weak order of WW, is a complete lattice; that is, every family of biclosed subsets has both a meet and a join.

References

Source: arXiv

Progress summary

Refreshed
Open · progress reported

A new unrefereed preprint claims to settle the conjecture in rank three, while the all-rank conjecture remains open.

Dyer posed the conjecture in 1994: the extended weak order of every Coxeter group should be a complete lattice. Earlier work established important affine and universal rank-three cases.

Known results

  • Rank-33 affine Coxeter groups (Wang, 2019).
  • All affine Coxeter groups (Barkley and Speyer, 2023).
  • The universal rank-33 Coxeter group U3U_3, including the lattice property and one-root covers (Barkley, Defant, Hersh, McCammond, McConville, and Speyer, 2025).

October 2026 rank-three claim

An October 2026 preprint claims that extended weak order is a lattice for every rank-three Coxeter group and that each cover toggles one root. If correct, this completes the rank-three case, but the claim is currently unverified; the all-rank conjecture remains open.

Current status (as of October 2026): The rank-three result is claimed in an unrefereed preprint but is unverified; the conjecture for arbitrary Coxeter groups remains open.

Sources

Solutions 0

No solutions have been posted yet.