Skip to content

Erdős problem 394

For the least tk(n)t_k(n) with ntk(n)(tk(n)+1)(tk(n)+k1)n \mid t_k(n)(t_k(n)+1)\cdots(t_k(n)+k-1), do the conjectured logarithmic-saving and adjacent-length estimates hold on average? Both answered affirmatively, with c=1/2048c = 1/2048 admissible in the t2t_2 bound.

No current result

No reviewed Result is current in Vela Mathematics Program. Retained source material is shown below.

Retained declaration

FormalConjectures/ErdosProblems/394.lean

Formal Conjectures

FormalConjectures/ErdosProblems/394.leanErdos394.erdos_394.parts.i1 lineExact file
True ↔ ∃ c > 0, (fun x => ∑ nFinset.Icc 1 ⌊x⌋₊, ↑(Erdos394.t 2 n)) =O[Filter.atTop] fun x => ↑x ^ 2 / Real.logx ^ c
SolvedProof has a holelean4external proof

The proof uses `sorry`: part of the argument is written but not proved. Lean accepts the file; it does not accept the theorem.

Proof manifests naming this Problem

  • William Blair Lean proofswilliamjblair:Research.erdos394_first_question_proved
  • William Blair Lean proofswilliamjblair:Research.erdos394_second_target_proved

Reported activity

Work these sources record against this Problem. Source-reported attribution, not reviewed here.

  • argument

    VibeMathed

    Machine
    GPT-5.6 starships (Claude Fable 5 reviewer)
    Reported outcome
    candidate
    Open the source record

Continue

Search problems.science

Find a Problem, Result, source, or page