Guglielmo Montone

Embodied AI, vision–language–action (VLA) models, JEPA-style latent prediction architectures, and robotics simulation. Multimodal learning, imitation learning.