Sharing one discrete codebook between visual-dynamics and robot-motion branches and using the codes as auxiliary supervision improves cross-embodiment VLA success rates in the authors' real-robot tests.
Octo: An open-source generalist robot policy
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
XR-1: Towards Versatile Vision-Language-Action Models via Learning Unified Vision-Motion Representations
Sharing one discrete codebook between visual-dynamics and robot-motion branches and using the codes as auxiliary supervision improves cross-embodiment VLA success rates in the authors' real-robot tests.