Preprint
Sep 2026
Devol-ONE: One Autoregressive Mixture of Transformers to Unify Vision-Language-Action and Latent World Modeling
Devol-ONE is presented, a Mixture of Transformers architecture that unifies vision language understanding, latent world dynamics prediction, and action generation within a single autoregressive framework.
Hong-Yi Cai, Yi Herng Ong, Ting-Ting Wu et al.
· 0 citations