1: \begin{abstract}

2: We propose an adaptive scheme of broadening the discrete spectral data from numerical renormalization group (NRG) calculations to improve the resolution of dynamical properties at finite energies.

3: While the conventional scheme overbroadens narrow features at large frequency by broadening discrete weights with constant width in log-frequency,

4: our scheme broadens each discrete contribution individually

5: based on its sensitivity to a $z$-shift in the logarithmic discretization intervals.

6: We demonstrate that the adaptive broadening better resolves various features in non-interacting and interacting models at comparable computational cost.

7: The resolution enhancement is more significant for coarser discretization as typically required in multi-band calculations.

8: At low frequency below the energy scale of temperature, 

9: the discrete NRG data necessarily needs to be broadened on a linear scale.

10: Here we provide a method that minimizes transition artefacts in between these broadening kernels.

11: \end{abstract}

12: