The mod-4 valuation law for Gaussian power sums

From papers

Let pp be an odd prime and define

Gn(p)=a=1p1b=1p1(a+bi)nZ[i],\mathbf G_n(p)=\sum_{a=1}^{p-1}\sum_{b=1}^{p-1}(a+bi)^n\in\mathbb Z[i],

with vp(x+yi):=min{vp(x),vp(y)}v_p(x+yi):=\min\{v_p(x),v_p(y)\}. The mod-4 valuation conjecture. For 1np21\le n\le p-2,

vp(G1(p))=1,vp(G2(p))=2,vp(G3(p))=3,v_p(\mathbf G_1(p))=1,\qquad v_p(\mathbf G_2(p))=2,\qquad v_p(\mathbf G_3(p))=3,

and for 4np24\le n\le p-2,

vp(Gn(p))={1,n0(mod4),2,n1(mod4),3,n2(mod4),4,n3(mod4).v_p(\mathbf G_n(p))=\begin{cases}1,&n\equiv0\pmod 4,\\2,&n\equiv1\pmod 4,\\3,&n\equiv2\pmod 4,\\4,&n\equiv3\pmod 4.\end{cases}

These formulas are suggested by the computed data and describe the observed valuation pattern for small exponents; their general validity is unresolved.

Progress summary

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

Sources & referencesView supporting material

Primary source

Nikita Kalinin and Faith Shadow Zottor, “A p-adic (p34) depth-5 supercongruence for Gaussian p-th power sums over a square”, arXiv:2602.00206 (2026).

Solutions 0

No solutions have been posted yet.