Zagier's dimension conjecture for multiple zeta values

About 28 years old · traced to

Let ζ(s1,…,sr)\zeta(s_1,\ldots,s_r) be a multiple zeta value with s1>1s_1>1 and s2,…,sr≥1s_2,\ldots,s_r\ge1, and let its weight be s1+⋯+srs_1+\cdots+s_r. For k≥1k\ge1, define

Zk:=span⁡Q{ζ(s1,…,sr):r≥1, s1>1, s2,…,sr≥1, s1+⋯+sr=k}\mathcal Z_k:=\operatorname{span}_{\mathbb Q}\{\zeta(s_1,\ldots,s_r):r\ge1,\ s_1>1,\ s_2,\ldots,s_r\ge1,\ s_1+\cdots+s_r=k\}

and dk:=dim⁡Zkd_k:=\dim\mathcal Z_k. Zagier's dimension conjecture. One has d1=0d_1=0, d2=d3=1d_2=d_3=1, and

dk=dk−3+dk−2,k≥4.d_k=d_{k-3}+d_{k-2},\qquad k\ge4.

This predicts the dimensions of the weight-graded pieces of the space of multiple zeta values. In particular, the conjectural numbers dkd_k count the independent generators in each weight, and there are no linear relations between elements of different weights; the source presents this as a conjectural description supported by numerical checks.

References

Primary source

Vincel Hoang Ngoc Minh, “On the Algebraic Bases of Polyzetas”, arXiv:2510.13295 (2026).

Additional references

24 papers in this index state this conjecture (1998–2025). The statement above is taken from the most recent of them; the others are arXiv:2505.07221, arXiv:2406.13630, arXiv:2402.11539, arXiv:2301.05906, arXiv:2205.07165, arXiv:2205.09929, arXiv:2111.00051, arXiv:1708.07464, arXiv:1611.01921, arXiv:1611.08693, arXiv:1510.06519, arXiv:1412.5099, and 11 more.

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.