Smooth Splitting and Zeros from On-Shell Recursion

TL;DR

This paper introduces a new recursive approach to explain hidden zeros and smooth splitting in various tree-level amplitudes, revealing new kinematic limits and conditions for zeros, with implications for gauge theories and gravity.

Contribution

It presents a novel linear shift in kinematic space and a contour integration argument to derive splitting formulae, extending understanding of amplitude behavior in multiple theories.

Findings

Derived generalized splitting formulae for near-zero kinematics

Proved improved UV scaling and splitting in special Galileon

Identified conditions for hidden zeros in four dimensions

Abstract

We describe a new approach to understanding the origins of recently discovered "hidden zeros" and "smooth splitting" of tree-level amplitudes in $\text{Tr}\phi^3$, Non-Linear Sigma Model (NLSM), Yang-Mill-Scalar (YMS) and the special Galileon. Introducing a new type of linear shift in kinematic space we demonstrate that the mysterious splitting formulae follow from a simple contour integration argument in the style of on-shell recursion. The argument makes use of only standard notions of tree-level factorization on propagators, but assumes improved UV behavior in the form of the absence of a residue at infinity. In the case of $\text{Tr}\phi^3$ and NLSM this is proven by identifying our shift as a special case of a more general construction called a $g$-vector shift; in the case of YMS it remains an unproven conjecture. This recursive perspective leads to numerous new results: we derive generalizations of the splitting formulae on more relaxed near-zero kinematics, including interesting new kinematic limits in which the amplitude splits into a triple-product; we also demonstrate that the uncolored special Galileon model has improved UV scaling and hence also splits. We also investigate the possible realization of hidden zeros in four dimensions. The conditions under which the dimensionality constraints are compatible with zero kinematics is investigated in detail for $\text{Tr}\phi^3$ and YMS; for the latter we find they can be realized only with certain restrictions on external helicity states. The realizable 4d zeros are proven by a similar recursive argument based on BCFW and is found to generalize to a new class of intrinsically 4d "helicity zeros" present in all sectors of YM and also gravity.