SOURCE-LINKED INTELLIGENCE

WLA$^3$: World Latent Action Modeling for Semantics, Dynamics, and Kinematics

arXiv · AI, language, vision and robotics · article · Sep 14, 2026 · UTC

Scaling generalist policy models with heterogeneous data is limited by the lack of unified, low-noise action supervision. Human egocentric videos are abundant, but only a small fraction comes with high-quality hand-action labels. Observed world transitions offer a common source of action-related supervision across data sources. We introduce WLA$^3$ (World Latent Action Modeling for Semantics, Dynamics, and Kinematics), a unified generalist policy model framework built around representations learned by a World Latent Action Model (WLAM). WLAM first learns how multimodal world states change over

Read original source ↗ Open in workspace

recordType

paper

region

Global