Home
Research Area
Recent Posts
Members
Intern Alumni
Publications
Contact
NVIDIA Research
Light
Dark
Automatic
Vision-Language-Action
Pretrained to Imagine, Fine-Tuned to Act: The Rise of World-Action Models
An exploration of World-Action Models (WAMs) as an emerging paradigm for robot foundation models, examining how video foundation model backbones offer an alternative to VLM-based approaches for building generalist manipulation policies.
Jun 15, 2026
0 min read
Blog
Cite
×
