← All problems

Yau’s number-theoretic conjecture on positive integral points in right simplices

For every integer n≥3n\ge 3 and every a1,…,an>0a_1,\ldots,a_n>0, let N(a1,…,an)N(a_1,\ldots,a_n) denote the number of positive integral points in the real right simplex {x∈R>0n:∑i=1nxiai≤1}\left\{x\in\mathbb{R}_{>0}^n:\sum_{i=1}^n\frac{x_i}{a_i}\le 1\right\}, assuming N(a1,…,an)>0N(a_1,\ldots,a_n)>0. Then

N(a1,…,an)≤(a1+⋯+annn)=1n!∏j=0n−1(a1+⋯+ann−j).N(a_1,\ldots,a_n)\le \binom{\frac{a_1+\cdots+a_n}{n}}{n}=\frac{1}{n!}\prod_{j=0}^{n-1}\left(\frac{a_1+\cdots+a_n}{n}-j\right).

Equality holds if and only if a1=⋯=an=ma_1=\cdots=a_n=m for some integer m≥nm\ge n.

References

Progress summary

Refreshed
Open · solution reported

An unrefereed 2026 preprint claims to prove the conjecture in every dimension three or higher, but the claim has not been independently checked.

Yau’s conjecture gives a sharp upper bound for positive integral points in right simplices, with equality only when all intercepts are equal integers. The claimed result also transfers this sharpness to spectral counting for quantum harmonic oscillators.

Known results

  • The weaker rough estimate holds for every dimension n≥3n \ge 3 (S. T. Yau and L. Zhang).
  • The sharper conjecture was proved in dimension n=6n=6 by Liang, Yau, and Zuo.
  • Related GLY upper-estimate work was published by Xuejun Wang and Stephen Yau in 2007.

October 13, 2026 claimed proof

Zhiwen Liu, Stephen S.-T. Yau, and Hao Zuo claim the sharp inequality and equality characterization for every dimension n≥3n \ge 3, with corresponding oscillator counting and remainder bounds. This is an unrefereed preprint, and no independent mathematical assessment was found.

Current status (as of October 2026): The conjecture is claimed proved for every dimension n≥3n \ge 3, but the new general proof remains unverified.

Sources

Solutions 0

No solutions have been posted yet.