Implications of structured continuous maximal regularity

We study how maximal regularity estimates with respect to the continuous functions improve automatically in cases where the spatial norm is fundamentally different from the supremum norm. More precisely, we invoke properties such as weak compactness of convolution-type operators related to the mild solutions of the underlying linear evolution equations to sharpen the a priori estimates. These results have several applications: such as a new proof of Guerre-Delabriere's result on $\mathrm{L}^1$-maximal regularity and an extension of Baillon's theorem; a simplification for well-known perturbation theorems for generation of $\mathrm{C}_0$-semigroups; and we resolve an open problem on input-to-state stability from control theory for a general abstract class of systems.