I'm confused by this statement: "Ground take-off, the most energetically demanding flight maneuver 75,76, would have been particularly challenging for enantiornithines, possibly requiring a running start to launch from the ground as inferred for Archaeopteryx 77."
Birds typically initiate flight by leaping into the air from a stationary position - regardless of whether the bird is on the ground or on a tree branch. The launch kinematics are the same. In both cases, the hindlimbs are the primary take-off accelerator (Earls, 2000 doi: 10.1242/jeb.203.4.725.). This was almost certainly true for the first flying theropods, including the earliest birds, which appear to have been capable of launching from a standing position, with the initial impulse provided by the hindlimbs (Dececchi et al., 2016; DOI 10.7717/peerj.2159).
There seems to be a misconception that weak or nascent fliers need to run in order to take-off from the ground. But among modern birds, a running take-off is typically used for launches from compliant surfaces (especially water) - as both Earls and Dececchi &c make clear.
In asserting that enantiornithines were limited to an arboreal ecology because of their poor flight abilities ("inability to perform continuous flight") this study by Wu &c might have it the wrong way round. If most enantiornithines were adapted for a life within the canopy, they might only have needed to fly short distances.