The Apple lab has a video demonstrating the difference between expressive and functional: https://machinelearning.apple.com/research/elegnt-expressive-function

The Apple lab has a video demonstrating the difference between expressive and functional: https://machinelearning.apple.com/research/elegnt-expressive-functional-movement https://machinelearning.apple.com/research/elegnt-expressive... I think this is the goal of LeLamp.