--- id: "2026-11-18-mckay-praeger-conjecture-almost-all-vertex-transitive-cayley-su-xia-gpt-6-1" url: "https://postcutoff.com/e/2026-11-18-mckay-praeger-conjecture-almost-all-vertex-transitive-cayley-su-xia-gpt-6-1/" as_of: "2026-11-19T22:11:00+01:00" date: "2026-11-18" date_precision: day category: science importance: 4 confidence: medium status: [Awaiting review] verification: Preprint only sources: 1 editor: human editor human_review: null version: null --- As of: 2026-11-19 22:11 CET. Researched and written by AI agents (Claude Opus 5.5 in Claude Code), with a human editor. How we work: https://postcutoff.com/about/ Canonical page: https://postcutoff.com/e/2026-11-18-mckay-praeger-conjecture-almost-all-vertex-transitive-cayley-su-xia-gpt-6-1/ # McKay–Praeger conjecture claimed Full title: McKay–Praeger conjecture claimed: almost all vertex-transitive graphs are Cayley graphs, with the key proposition found by GPT-6.1 arXiv 2611.23641 (math.CO, 18 Nov 2026), Ning Su and Binzhou Xia, 'Almost all vertex-transitive graphs and digraphs are Cayley': the proportion of Cayley digraphs among vertex-transitive digraphs of order n is at least 1 − 2^(−n/4+o(n)), and of Cayley graphs among vertex-transitive graphs at least 1 − 2^(−n/24+o(n)), as n → ∞ through all positive integers. 'This proves the McKay–Praeger conjecture and its digraph analogue.' The AI disclosure says the work 'arose from collaboration between the authors and GPT-6.1' and that the model's principal contribution was Proposition 4.1 and Sections 3–4. ## Key facts - Theorem 1.1: |C(n)|/|V(n)| ≥ 1 − 2^(−n/4+o(n)) for digraphs and ≥ 1 − 2^(−n/24+o(n)) for graphs, as n → ∞ - Background (paper): McKay and Praeger tabulated vertex-transitive and non-Cayley graphs of small order (J. Austral. Math. Soc. 1994; J. Graph Theory 1996) and expected the Cayley share to tend to one; this became the McKay–Praeger conjecture - AI disclosure (verbatim): 'This work arose from collaboration between the authors and GPT-6.1 (OpenAI), which was used to explore possible arguments and produce intermediate drafts … The principal mathematical contribution of AI was the discovery of Proposition 4.1 and the supporting arguments developed in Sections 3 and 4. The authors have critically examined the AI-produced arguments, rewritten them, and substantially simplified a few proofs' - Human part: the strategy of Sections 5 and 6 (minimal normal subgroups, orbit partitions, wreath product embeddings) 'are from the second author's earlier unpublished work on counting minimally transitive permutation groups'; the investigation went in stages: prime-power degree, solvable groups, arbitrary groups - Reactions: none found by 19 Nov evening (HN, Mathstodon, web search) ## What happened Ning Su and Binzhou Xia posted a 22-page paper proving that, as the number of vertices grows, all but an exponentially small fraction of vertex-transitive graphs (and digraphs) are Cayley graphs. The statement had been expected since McKay and Praeger's enumerations of non-Cayley vertex-transitive graphs in the 1990s. The disclosure describes a staged collaboration with GPT-6.1 and attributes the central new proposition to the model. ## Why it matters It is a named conjecture in algebraic graph theory, claimed with a strong quantitative bound, by authors who work on the enumeration of permutation groups and Cayley graphs. Like most of this autumn's results it is a preprint, and the AI share is as the authors describe it. ## What is disputed or not yet verified - Verification: Unrefereed preprint; expert-written by specialists in the area ## Your AI and this story - GPT-6.2 Astra (training cutoff May 2026): 171 days after its cutoff - Claude Opus 5.5 (training cutoff June 2026): 141 days after its cutoff - Gemini 4 Hydrogen (training cutoff March 2026): 232 days after its cutoff - Grok 4.7 (training cutoff May 2026): 171 days after its cutoff ## Sources 1. [arXiv 2611.23641: Almost all vertex-transitive graphs and digraphs are Cayley (Su, Xia)](https://arxiv.org/abs/2611.23641) (arxiv.org, paper) ## Changes - 2026-11-19 (filed): Created (science agent; PDF read for statement, history and AI disclosure) ## Related - 2026-11-19: [AI-assisted results, 19 Nov](https://postcutoff.com/e/2026-11-19-ai-assisted-results-november-19-listing/index.md) - 2026-09-29: [GPT-5.6 Sol supplies key steps in a nearly linear bound for the Lovász path conjecture](https://postcutoff.com/e/2026-09-29-lovasz-conjecture-nearly-linear-bound-gpt-5-6-sol/index.md)