VLA-Precision: Asymmetric Co-Bootstrapping for Efficient Real-World Online RL of Vision-Language-Action Models
Paper • 2609.04355 • Published • 9
None defined yet.
VLA-Precision: Asymmetric Co-Bootstrapping for Efficient Real-World Online RL of Vision-Language-Action Models
FACET: Preserving Source Intent and Executable State in Terminal Task Synthesis