Claims

../

2013_01_01_de_la_breteche_ford_vandehey: The 2013 counting bounds for disjoint progressions with distinct moduli give L(m)^(-1+o(1)) <= epsilon_m <= L(m)^(-sqrt(3)/2+o(1)), so epsilon_m tends to 0; refereed, and superseded by Ho's sharp estimate.

2026_04_23_ho: Boon Suan Ho's 2026 manuscript, written with GPT-5.4 Pro, proving that the supremum of reciprocal sums over disjoint classes with distinct moduli above m is exp(-(1+o(1)) sqrt(log m log log m)); accepted on the site's credit.

2026_04_28_zribi: Malek Zribi's five-page note of 28 April 2026, written with GPT-5.5, deriving the sharp estimate for the reciprocal-sum supremum from the sharp asymptotic for Problem 202, assumed as a hypothesis; claimed and conditional.

2026_04_30_chojecki: An eight-page unsigned draft dated 30 April 2026, posted by Przemek Chojecki on ULAM's site and credited to GPT-5.5 Pro, deriving the sharp Problem 202 upper bound by a spread-core route and transferring it to Problem 1190.

---