ForeAct3D: Policy-Grounded Future World Modeling for VLA Policies
ForeAct3D: Policy-Grounded Future World Modeling for VLA Policies
This paper presents ForeAct3, a framework for policy-grounded future world modeling within VLA policies. It uses learnable geometric queries to decode depth, semantic segmentation, and camera pose from policy representations, enabling accurate prediction of future 3D scene states. The method improves manipulation performance through semantic 3D supervision, physical consistency, and action conditioning, achieving 98.3% average success on LIBERO and 37.8% success in real-world tasks.
ForeAct3D introduces a novel framework for policy-grounded future world modeling in Vision-Language-Action (VLA) policies, enhancing manipulation performance through semantic 3D supervision, physical consistency, and action conditioning. The method achieves high success rates on LIBERO and CALVIN benchmarks without requiring robot pretraining, and shows significant improvement in real-world tasks.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- code: Available Artifact link Open source S6
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
ver the base policy. The project page and code are available at https://github.com/anthonytao80-crypto/ForeAct3D.
Open source S6
Source：arXiv Robotics — research abstracts · arxiv.org