RI-(S)MOM to $\overline{\rm MS}$ conversion for $B_K$ at two-loop order

TL;DR

This paper refines the calculation of the Kaon bag parameter $\hat{B}_K$ to NNLO, providing precise world averages and scheme conversion techniques crucial for flavor physics and beyond Standard Model tests.

Contribution

It introduces a two-loop matching between RI-(S)MOM and $\overline{\mathrm{MS}}$ schemes and offers updated, comprehensive lattice-based averages for $\hat{B}_K$ in 3- and 4-flavour theories.

Findings

Precise $\hat B_{K}^{(f=3)}$ value of 0.7627(60)

Updated $|\epsilon_K|$ value of 2.171(65)×10^{-3}

Inclusion of NNLO corrections and scheme conversion

Abstract

The Kaon bag parameter $ {\hat{B}}_K $ plays a critical role in constraining the parameters of the CKM matrix and in probing physics beyond the Standard Model. In this work, we improve the precision of $ \hat{B}_K $ to next-to-next-to-leading order (NNLO) and provide world averages for both $3$- and $4$-flavour theories. In the course of this, as our main technical development, we carry out the two-loop matching between the RI-(S)MOM and $\overline{\mathrm{MS}}$ schemes. Our world averages combine all available lattice data, including conversion between the 3- and 4-flavour theories as appropriate. We obtain the result $\hat B_{K}^{(f=3)} = 0.7627(60)$, which comprises the complete set of $3$- and $4$-flavour lattice results and can be used directly in phenomenological applications. The error is dominated by lattice uncertainties and missing higher-order corrections (residual scale dependence). Our averages include a PDG rescaling factor of 1.28 reflecting a mild tension among the lattice inputs after inclusion of NNLO corrections in the scheme conversion and matching across flavour thresholds. Our averages imply an updated value $|\epsilon_K|=2.171(65)_\text{pert.}(71)_\text{non-pert.}(153)_\text{param.} \times 10^{-3}$. We briefly discuss applications of our results to $D$-meson mixing.