Pretrained to Imagine, Fine-Tuned to Act: The Rise of World-Action Models | NVIDIA Technical Blog
Quick glossary for readers new to VLA/WAM terminology VLA Vision-Language-Action model: a robot policy that starts from a pretrained VLM backbone and adapts it to generate actions from visual…
Quick glossary for readers new to VLA/WAM terminology VLA Vision-Language-Action model: a robot policy that starts from a pretrained VLM backbone and adapts it to generate actions from visual observations and language instructions. Large-scale VLM pretraining is a core part of the recipe. See Pi-0 and GR00T N1 . WAM World-Action Model: a policy that starts from a pretrained world-model or video backbone and adapts it to represent or predict how the scene changes over time and emit corresponding actions. We use WAM as the term throughout this post. VLM Vision-Language Model: a model pretrained
saved by
related reading
- General Instinct | Any frontier model. Any edge device.general-instinct.com
- World Action Model Atlasjoeclinton.me
- [2603.16666] Fast-WAM: Do World Action Models Need Test-time Future Imagination?arxiv.org
- [2602.10098] VLA-JEPA: Enhancing Vision-Language-Action Model with Latent World Modelarxiv.org
- State of Vision-Language-Action (VLA) Research at ICLR 2026 – Moritz Reussmbreuss.github.io
- What Matters for Latent Actions in Robot Learningcarldegio.github.io
- [2410.11758] Latent Action Pretraining from Videosarxiv.org
- Vision-Language-Action (VLA) Models: A Review of Recent Progressxxxxyu.github.io
- Moritz Reuss — Robotics & VLA Researchmbreuss.github.io
- 1X World Model | From Video to Action: A New Way Robots Learn1x.tech
- Introducing S1: In-Context Learning for Roboticsskild.ai
- State of Robot Learning, December 2025vedder.io