SOURCE-LINKED INTELLIGENCE

REFACTOR-VLA: Unsupervised Library Learning of Typed Motor Programs

arXiv · AI, language, vision and robotics · article · Sep 1, 2026 · UTC

Most vision-language-action (VLA) models -- OpenVLA, $π_0$, RT-2, RDT-1B -- are monolithic: they emit raw motor commands or short action chunks without organizing behavior into reusable abstractions, so they degrade on long-horizon tasks and resist interpretation. Existing skill-discovery methods sidestep the core question of when two action sequences are behaviorally equivalent, either clustering contrastive embeddings or delegating the judgment to a language model uncalibrated to the robot's dynamics. We introduce REFACTOR-VLA, a wake/sleep system for learning reusable skills. Its sleep phas

Read original source ↗ Open in workspace

recordType

paper

region

Global