World2Act: Latent Action Post-Training via Skill-Compositional World Models
Paper
• 2603.10422 • Published
An Vuong
andvg3
·
AI & ML interests
CV & Robotic Learning
Recent Activity
authored
a paper
about 2 hours ago
World2Act: Latent Action Post-Training via Skill-Compositional World Models authored
a paper
5 months ago
Language-Driven 6-DoF Grasp Detection Using Negative Prompt Guidance authored
a paper
5 months ago
Improving Robotic Manipulation with Efficient Geometry-Aware Vision
Encoder Organizations
authored a paper about 2 hours ago