Fetching the paper…

M2D2M: Multi-Motion Generation from Text with Discrete Diffusion Models · Around