#machine learning
Mar 2026
From Diffusion To Flow: Efficient Motion Generation In MotionGPT3
Comparing diffusion and rectified flow objectives within the MotionGPT3 framework suggests that several known benefits of rectified flow objectives do extend to continuous-latent text-to-motion generation, highlighting the importance of the training objective choice in motion priors.
Jaymin Ban, J. Jeon, Sang-jun Jeong
· arXiv.org · 1 citation