From the source
Lead story
Ask your AI
Top stories
Models & availability
Latest
REFACTOR-VLA: Unsupervised Library Learning of Typed Motor Programs
A system that learns reusable motor skills via a wake/sleep architecture.
Apple researchers introduce REFACTOR-VLA, a system that learns reusable motor skills using a wake/sleep architecture with a Behavioral-Equivalence Kernel and typed lambda terms.
Evaluated on the LIBERO benchmark, the study finds that increasing world model size from 188M to 430M parameters worsened performance on all four suites, while adding an auxiliary InfoNCE contrastive loss during world-model warmup improved skill clustering quality.
From the source
We introduce REFACTOR-VLA, a system that learns reusable skills using a "wake/sleep" architecture. In the sleep phase, the system clusters segments of motor programs using a Behavioral-Equivalence Kernel (BEK).
machinelearning.apple.com
