Lattice QCD with mixed actions

TL;DR

This paper addresses the challenge of performing lattice QCD with mixed actions, specifically a sea of improved staggered quarks combined with overlap valence quarks. It proposes a mass-matching framework, $m_{val} = \zeta(a) m_{sea}$ with $\zeta(a) \to 1$ in the continuum, and investigates unitarity-violating effects that arise when $m_{val}$ and $m_{sea}$ are mismatched, using scalar-correlator diagnostics. The authors report exploratory results for the light-hadron spectrum, pseudoscalar decay constants, and some charm-physics observables, demonstrating that overlap valence on a staggered sea can yield clean signals despite limited statistics and potential lattice artifacts. The work highlights the potential computational benefits of mixed-action approaches and motivates further refinement of quark-mass matching and smearing strategies to control short-distance effects and discretization errors, enabling more precise QCD predictions with reduced cost.

Abstract

We discuss some of the implications of simulating QCD when the action used for the sea quarks is different from that used for the valence quarks. We present exploratory results for the hadron mass spectrum and pseudoscalar meson decay constants using improved staggered sea quarks and HYP-smeared overlap valence quarks. We propose a method for matching the valence quark mass to the sea quark mass and demonstrate it on UKQCD clover data in the simpler case where the sea and valence actions are the same.