SOURCE-LINKED INTELLIGENCE

SCULPT-VLA: Learning Structured Control through Staged Action Grounding

arXiv · AI, language, vision and robotics · article · Sep 20, 2026 · UTC

Vision-language-action (VLA) policies increasingly incorporate structured intermediate supervision beyond action labels. Yet specifying what an intermediate representation should encode leaves open how action prediction learns to depend on it. We introduce \textbf{SCULPT-VLA}, a policy that learns structured control through staged action grounding. Its action-conditioning state comprises complementary factors for task progression, scene dynamics, and spatial grounding. Training first forms these factors with teacher scaffolds, then grounds coarse action prediction through their composition as

Read original source ↗ Open in workspace

recordType

paper

region

Global