bibkey: lee2026quantumcerny authors: Pui Hang Lee, Pui Yee Lee and Bjørn Kjos-Hanssen year: 2026 title: “Quantum Černý complexity of binary words” doi: null url: https://arxiv.org/abs/2609.40154 claim: “The quantum Černý complexity qc(w) is the least d for which two quantum channels on d x d density matrices and a start state make w the unique shortest synchronizing word; 2 <= qc(w) <= ceil(sqrt(|w|+1)), qc(0^m) = ceil(sqrt(m+1)), qc(01^n0) = 2; open problem 1 conjectures qc(01101001) = 2.” strata_touched:
- D5/S3/Quantum/QuantumChannels/QuantumCernyThueMorseRefutation license: citation-only triage: anchor
Quantum Černý complexity of binary words
arXiv:2609.40154v1 (2026-09-30; primary quant-ph, cross-list cs.DM). Definition 1.1 reads:
Let . An instance is a tuple where are quantum channels on the density matrices and is a density matrix. For a word write (letters applied left to right), with . The reachable set is . A word is synchronizing for the instance if is constant on , i.e. there is a density matrix with for all . The quantum Černý complexity
\qc(w)of a nonempty word is the least such that some instance of dimension has as its unique shortest synchronizing word.
The discussion section lists open problems; item 1 reads
(Thue–Morse prefix.) We conjecture
\qc(01101001)=2; the bound\qc(01101001)\le3follows from Theorem~\ref{thm:kmp}. More generally, characterize\{w:\qc(w)=2\}.
and item 3 asks for \qc(w) of a uniformly random word, noting that “it is
consistent with our results that \qc(w)=2 for almost all ”.
Verified locator
- URL: https://arxiv.org/abs/2609.40154 (v1, the only version; source file
quantum-cerny__2_.tex, md57c67e915a9bfc1f8c57ffcfcce55016c). - The authors’ Lean formalization, https://github.com/bjoernkjoshanssen/quantumcerny,
proves
qc(01101001) ≤ 3and leaves the conjecture unformalized.