Explicit group construction with infinite twin-width
Question (explicit group with infinite twin-width) · arXiv:2204.12330
Status open high confidence
As of 2025, no explicit finitely generated group with infinite twin-width (nor an explicit bounded-degree graph with unbounded twin-width) is known; the original existence proof relies on probabilistic methods via Osajda's embedding result. A 2025 paper on near-regular graphs (arXiv:2504.02342) notes that extremal bounded-degree graphs with high twin-width are asymmetric, offering a structural explanation for why explicit examples remain elusive. The question stated in arXiv:2204.12330 is still open.
Cited literature (1)
-
Shows that extremal bounded-degree graphs with high twin-width are asymmetric, partly explaining why no explicit bounded-degree graph with unbounded twin-width (and hence no explicit group with infinite twin-width) is yet known; notes that no explicit cubic graph with twin-width greater than 4 is known.
Reviewer notes. No follow-up paper resolving the question was found after 5 web calls. The related paper arXiv:2504.02342 (2025) provides structural evidence that explicit high-twin-width bounded-degree examples are hard to construct, confirming the question remains open. The original proof in arXiv:2204.12330 (Theorem 1.1) is non-constructive, using probabilistic embeddings via Osajda's result.
Context
The group constructed in Theorem 1.1 relies on probabilistic methods and starts from a sequence of bounded degree graphs with unbounded twin-width; it does not yield an explicit bounded degree graph with unbounded twin-width. A more explicit or combinatorially direct construction would significantly advance understanding of twin-width for bounded degree graphs.
Notes. Stated as a question in prose at the end of Section 1.3.
Source paper
Twin-width VII: groups
Édouard Bonnet, Colin Geniet, Romain Tessera, Stéphan Thomassé · 2022-07-15
https://arxiv.org/abs/2204.12330
PDF source