← All problems

Guo–Krattenthaler positivity conjecture for q-binomial quotients

For every positive integer nn, define Pn(q)=(1−q)2(1−q10n−1)(1−q15n−1)[30n5n]qP_n(q)=\frac{(1-q)^2}{(1-q^{10n-1})(1-q^{15n-1})}\begin{bmatrix}30n\\5n\end{bmatrix}_q, where [N\K]q=∏i=1K1−qN−K+i1−qi\begin{bmatrix}N\K\end{bmatrix}_q=\prod_{i=1}^{K}\frac{1-q^{N-K+i}}{1-q^i}. Then Pn(q)∈Z[q]P_n(q)\in\mathbb{Z}[q], its coefficient of qq is −1-1, its coefficient of q125n2−25n+3q^{125n^2-25n+3} is −1-1, and every other coefficient of Pn(q)P_n(q) is non-negative.

References

Source: arXiv

Progress summary

Refreshed
Open · progress reported

A new paper claims substantial progress on the conjecture, including a proof for all sufficiently large parameter values, but the remaining finite cases are unresolved.

Guo and Krattenthaler conjectured that whenever a quotient of two qq-binomial coefficients is a polynomial, all its coefficients are non-negative. The general conjecture remains open, despite several proved parameter families.

Known results

  • The cases l=kl=k, l=k−1l=k-1, l=0l=0, l=1l=1, and implicitly l=k−2l=k-2 were known previously.
  • The cases l=k−3l=k-3 and l=2l=2 were proved by Mirzavaziri and Yaqubi (2026), with computation reported through n≤400n\le 400.
  • A later sufficient condition is 2k⌊k−12⌋+k−l<n2k\left\lfloor\frac{k-1}{2}\right\rfloor+k-l<n; in particular, the conjecture holds for n≥k2n\ge k^2.

October 2026 claimed all-large-nn advance

Mirzavaziri and Yaqubi reportedly establish the first and last 25n+125n+1 coefficients uniformly and the full positivity pattern for all sufficiently large nn. This is a claimed advance, not an independently verified resolution; the finite range below the threshold remains open.

Current status (as of October 2026): Several families and a sufficient large-nn regime are established or claimed, but the finite exceptional range and the full conjecture remain unresolved.

Sources

Solutions 0

No solutions have been posted yet.