Eight-face conjecture for facets of the Newton polytope

Let N3,n\mathbf{N}_{3,n} be the Newton polytope, and let γj1,j2,j3\gamma_{j_1,j_2,j_3} be the generalized positive root

γj1,j2,j3=t=j1j22α1,t+t=j21j33α2,t.\gamma_{j_1,j_2,j_3}=\sum_{t=j_1}^{j_2-2}\alpha_{1,t}+\sum_{t=j_2-1}^{j_3-3}\alpha_{2,t}.

Assume that {j1,j2,j3}\{j_1,j_2,j_3\} is totally nonfrozen, meaning that no two indices are cyclically adjacent. The generalized positive roots are linear functions on R2×(n3)\mathbb{R}^{2\times(n-3)}.

Eight-face conjecture. The facet of N3,n\mathbf{N}_{3,n} minimizing γj1,j2,j3\gamma_{j_1,j_2,j_3} has exactly eight codimension-33 faces combinatorially isomorphic to

N3,j2j1+2×N3,j3j2+2×N3,j1j3+n+2.\mathbf{N}_{3,j_2-j_1+2}\times\mathbf{N}_{3,j_3-j_2+2}\times\mathbf{N}_{3,j_1-j_3+n+2}.

These faces correspond to the eight weakly separated collections listed in the source. The claim generalizes the observed three-split structure and remains unproved in the supplied text.

Sources & referencesView supporting material

Primary source

Nick Early, “Factorization for Generalized Biadjoint Scalar Amplitudes via Matroid Subdivisions”, arXiv:2211.16623 (2023).

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.