SOURCE-LINKED INTELLIGENCE
GaussianWAM: Distilling Geometry and Semantics from 3D Gaussian Fields into World-Action Models
World-Action Models (WAMs) jointly learn future visual prediction and action generation, using video dynamics as a representation-learning signal for robotic manipulation. However, their video latents are primarily optimized for visual prediction and are not explicitly encouraged to preserve cross-view geometric structure or spatially localized, object-relevant semantics. We propose \textbf{GaussianWAM}, a training-time representation-enhancement framework that organizes geometric and semantic supervision through a 3D Gaussian field. Given synchronized multi-view observations, frozen geometry
Read original source ↗ Open in workspace
- recordType
- paper
- region
- Global
Evidence & attribution
- arXiv · AI, language, vision and robotics · 2026-08-25T15:34:13.000Z
First collected: 2026-09-21T10:02:02.728Z. This is not the publication date.