Subexponential Ramsey bound for linear 3-graphs
Conjecture 4.7 · arXiv:2404.02021
Status open high confidence
Conjecture 4.7 asks for an absolute constant C > 0 such that r(F, K_n^(3)) ≤ 2^{O_F(n^C)} for every fixed linear 3-graph F. The source paper itself (Theorem 1.4) disproves the stronger polynomial upper bound, motivating this weaker subexponential form. A 2024 follow-up by essentially the same author group (arXiv:2411.13812) characterizes polynomial growth of r(H, K_n^(3)) for tightly connected and sparse-component 3-graphs, but does not establish any subexponential upper bound for arbitrary linear 3-graphs; the conjecture remains open.
Cited literature (1)
-
Characterizes polynomial growth of r(H, K_n^(3)) for tightly connected and sparse-component 3-graphs (showing polynomial iff H is contained in an iterated blowup of an edge), confirms superpolynomial growth for some linear 3-graphs, but does not prove the subexponential upper bound asserted in Conjecture 4.7.
Reviewer notes. The internal reference arXiv:2312.15572 does not address this conjecture — it is about VC-dimension in graphs, not hypergraph Ramsey theory, and should be discarded as a false positive. The most relevant post-statement work is arXiv:2411.13812 (same authors + Yu), which tackles the related polynomial-characterization question but leaves the subexponential conjecture fully open.
Context
Following Theorem 1.4, which disproves the polynomial upper bound for linear $3$-graphs, the authors propose the weaker conjecture that $r(F,K_{n}^{(3)})$ remains subexponential for linear $F$. They note possible further strengthenings (replacing $n^C$ by $n^{o(1)}$ or extending to $123$-inducible $3$-graphs) but are less certain about those variants.
Source paper
On off-diagonal hypergraph Ramsey numbers
David Conlon, Jacob Fox, Benjamin Gunby, Xiaoyu He, Dhruv Mubayi, Andrew Suk, Jacques Verstraete · 2024-04-02
https://arxiv.org/abs/2404.02021