VITRA: A Foundation for Dexterous VLA via Human Video Pretraining
A scalable VLA pretraining pipeline that converts unstructured egocentric human videos into robot training data, trains a dexterous hand VLA, and fine-tunes on robot data, achieving strong zero-shot generalization and real-robot dexterous manipulation.
The skinny
The skinny isn't ready yet — notes appear once the transcript is processed.