← All problems

Furusho–Yamashita direct-sum conjecture

Let ζ(n1,…,nr)=∑k1>⋯>kr≥1k1−n1⋯kr−nr\zeta(n_1,\ldots,n_r)=\sum_{k_1>\cdots>k_r\ge 1}k_1^{-n_1}\cdots k_r^{-n_r} be a classical multiple zeta value, where r≥1r\ge 1, n1≥2n_1\ge 2, and ni≥1n_i\ge 1 for i>1i>1. For each weight w≥2w\ge 2, let Zw\mathcal{Z}_w be the Q\mathbb{Q}-vector space spanned by all such multiple zeta values with n1+⋯+nr=wn_1+\cdots+n_r=w, and let Z=Q+∑w≥2Zw\mathcal{Z}=\mathbb{Q}+\sum_{w\ge 2}\mathcal{Z}_w. The Furusho–Yamashita direct sum conjecture asserts that Z=Q⊕⨁w≥2Zw\mathcal{Z}=\mathbb{Q}\oplus\bigoplus_{w\ge 2}\mathcal{Z}_w; equivalently, the weight decomposition is a direct sum, so there are no nontrivial Q\mathbb{Q}-linear relations between multiple zeta values of distinct weights.

References

Progress summary

Refreshed
Open · progress reported

Recent work extends a related function-field result to all finite places, but the classical number-field conjecture remains open.

The Furusho–Yamashita conjecture predicts a direct-sum, weight-graded structure for relations among classical number-field multiple zeta values. The retrieved literature reports only positive-characteristic function-field analogues, not a resolution of the original conjecture.

Known results

  • For every finite place vv, the vv-adic values satisfy the same k‾\overline{k}-algebraic relations as the corresponding ∞\infty-adic values, with Z‾/(ζA(q−1))≅Z‾v\overline{\mathcal{Z}}/(\zeta_A(q-1))\cong\overline{\mathcal{Z}}_v.
  • For degree-one vv, Zv=k⊕⨁w≥1Zv,wZ_v=k\oplus\bigoplus_{w\ge1}Z_{v,w}, and the canonical grading map is bijective; this is identified as a function-field analogue of the conjecture.
  • Earlier results established the corresponding function-field relation and grading statements in narrower settings.

October 18, 2026 function-field extension

A newly reported preprint by Hung-Chun Tsui claims the function-field analogue over arbitrary finite places, extending earlier degree-one results and broadening the known weight-gradings and rationality of relations. This is progress on an analogue, not a direct result for the classical number-field conjecture.

Current status (as of October 2026): The classical number-field Furusho–Yamashita conjecture remains open; function-field analogues are established or newly claimed for increasingly general finite places.

Sources

Solutions 0

No solutions have been posted yet.