The work here makes substantial progress towards resolving the well known
feedback bottleneck of multi-antenna coded caching, by introducing a
fundamentally new algorithm that completely untangles caching gains from
feedback costs. In the context of the $K$-user MISO broadcast channel with
cache-aided receivers and an $L$-antenna transmitter, the algorithm achieves
full multiplexing gains and unbounded near-optimal caching gains at the mere
CSIT/CSIR costs associated to achieving multiplexing gains only. While prior
state-of-art required feedback costs that scaled with $K$, the new algorithm
requires only $L$ uplink/downlink feedback acquisition slots, irrespective of
how large the achieved caching gain is, and thus irrespective of how many users
are served at a time. In the end, the result solidifies the role of coded
caching as a method for reducing feedback requirements in multi-antenna
environments.