Answers are grounded in this paper and cite their sources.
Salehi Nowbandegani, Esfandiari, Shirdareh Haghighi and Bibak conjectured that every cubic graph contains a cycle of some length ℓ for which the interval [2ℓ, 3ℓ] contains a power of two; by contracting triangles, this would imply the Erdős–Gyárfás conjecture for claw-free cubic graphs. Call such lengths compatible, and let S32 be the set of compatible lengths at most 32. We prove, with computer assistance, that every nonempty finite simple cubic graph of order less than 2182 contains a cycle whose length lies in S32. A second proof, which requires much less computation, gives the same conclusion for graphs of order less than 1504. The argument bounds the size of a breadth-first-search ball. Short cycles are grouped into clusters, of which there are 46 isomorphism types, and the effect of longer cycles is charged to local configurations whose total charge is bounded by exhaustive enumeration and integer optimization. As a consequence, every nonempty finite simple cubic claw-free graph of order less than 6546 contains a cycle whose length is a power of two not exceeding 64.
No takes yet. Share an insight, caveat, or question.
Yuta110972 (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: