Erdős Problem a Daythe open tracker

Erdős Problem #698

PROVED LEANdatabase updated 2026-01-16

The statement as posed is true, and a proof exists. A solution has also been checked by machine, in Lean.

What the community database records, not our verdict. erdosproblems.com no longer shows statuses (why).

Is there some \(h(n)\to \infty\) such that for all \(2\leq i<j\leq n/2\) \[\textrm{gcd}\left( \binom{n}{i},\binom{n}{j}\right) \geq h(n)?\] This was resolved by Bergman [Be11], who proved that for any \(2\leq i<j\leq n/2\) \[\textrm{gcd}\left( \binom{n}{i},\binom{n}{j}\right) \gg n^{1/2}\frac{2^i}{i^{3/2}},\] where the implied constant is absolute. The linked formal proof (van Doorn and Aristotle, see `erdos_698.variants.bergman`) gives the explicit bound \(\gcd > \frac{2^i \sqrt n}{4 i \sqrt{i - 1}}\), so \(h(n) = \lfloor \sqrt n / 4 \rfloor\) works since \(i \sqrt{i - 1} \le 2^i\) for \(i \ge 2\).
Statement as formalized in Formal Conjectures (Google DeepMind, Apache-2.0) · their label: research solved · 3 further variants / known partial results in the Lean file. The authoritative wording is on erdosproblems.com/698.
Prizenone
Topicsnumber theory, binomial coefficients
Statement in LeanFormal Conjectures
Solution in Leanyes
OEISA400908
Original pageerdosproblems.com/698
NOBODY HAS SENT US AN ATTEMPT YET

If you take a run at this one — with your own AI subscription or without — the guide says how, and the submission form gets you an honest read of what you found.