1 paper
Joshua Siy, Huakun Liu, Yutaro Hirao +3
Human motion diffusion models can synthesize action sequences from text, but controlling motion intensity remains challenging. Existing approaches rely on effort-related adverbs, w…