The discrete restriction conjecture for toral eigenfunctions

Let p2p\geq 2 and d2d\geq 2. Let Td=Rd/Zd\mathbb{T}^d=\mathbb{R}^d/\mathbb{Z}^d be the square torus, and let eNe_N be an eigenfunction of Δ\sqrt{-\Delta} with eigenvalue 2πN2\pi N. The discrete restriction conjecture. One should have

eNLp(Td)εNε(Nd22dp+1)eNL2(Td).\|e_N\|_{L^p(\mathbb{T}^d)}\lesssim_{\varepsilon}N^{\varepsilon}\left(N^{\frac{d-2}{2}-\frac{d}{p}}+1\right)\|e_N\|_{L^2(\mathbb{T}^d)}.

Moreover, the extra factor NεN^{\varepsilon} can be omitted if d5d\geq 5. This conjecture gives the expected sharp numerology for restriction estimates on lattice points of spheres. The paper proves the bound without the loss when d5d\geq 5 and p>2dd4p>\frac{2d}{d-4}, while in a wider range it obtains a logarithmic loss; the full stated range remains open.

Sources & referencesView supporting material

Primary source

Daniel Pezzi, “Sharp Eigenfunction Bounds on the Torus for large p”, arXiv:2603.10927 (2026).

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.