Answers are grounded in this paper and cite their sources.
Mathematical analysis proves the sharp upper bound in Erdős Problem 848 across all positive integers, confirming that arithmetic progressions modulo 25 attain maximum density.
For N ≥ 1, let A be a subset of [1,N] such that ab+1 is nonsquarefree for every a,b in A, with a=b allowed. We prove the sharp bound |A| ≤ floor((N+18)/25), with equality attained by the elements of the progression 7 modulo 25. Earlier work established the same formula only above a large threshold. Our proof converts the extremal problem, by an exact Hall equivalence, into a completion inequality relative to the progression 7 modulo 25. A prefix-compatible colouring controls all N ≤ 5·106 simultaneously. Beyond this point, square-divisor estimates first eliminate defects supported on the competing progression 18 modulo 25 and then force every remaining defect to contain a large residual set. A valuation and residue decomposition supplies bounded pivot configurations; finite-prime counts and spacing between solutions of a transformed quadratic equation rule out each configuration. The resulting interval estimates are uniform, and a final monotone argument controls the unbounded tail. All finite inequalities and witnesses are exact, and the complete theorem is formally verified in Lean 4.
No takes yet. Share an insight, caveat, or question.
Alex Chengyu Li (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: