ZimaBlue: Evolving Generalizable World Action Models through Scalable Video Pre-training
Classification
View four quadrants- Major category
- WAMs
- Quadrant
- Q4 · Dual-system × IDM
- Architecture
- Dual-system
- Prediction paradigm
- IDM
- Subcategories
- Visual planning & IDMEfficient inference & real-time controlGeneralization & action alignment
- Source review status
- Verified from primary sources
Contribution
We introduce ZimaBlue, a scalable framework for learning generalizable World Action Models (WAMs) from large-scale video.
Abstract
Robotic manipulation faces a fundamental scaling challenge: robust generalization demands broad physical experience, yet action-labeled robot trajectories are expensive to collect and inherently limited in diversity. Egocentric videos offer a far more scalable source of embodied experience, capturing object interactions, contact dynamics, tool use, and long-horizon behaviors across diverse environments. The central challenge is how to convert this abundant but action-free experience into effective robot control. We introduce ZimaBlue, a scalable framework for learning generalizable World Action Models (WAMs) from large-scale video. ZimaBlue follows a three-stage training curriculum: it first performs causal embodied video pre-training on large-scale human and robot egocentric videos, then grounds the learned visual dynamics in heterogeneous robot trajectories through video-action mid-training with a unified action representation, and finally specializes the model to a target robot for deployment. To make generative WAMs practical for real-time control, ZimaBluefurther adopts an asynchronous Slow-Fast dual-system architecture, where a high-capacity Slow world model provides generalizable spatiotemporal representations and a lightweight Fast branch enables 30 Hz action prediction on NVIDIA RTX 4090. On real-robot zero-shot evaluations, scaling from target-robot data alone to over 120,000 hours of embodied video improves success from 36.1% to 77.8%. ZimaBlue further delivers strong performance across multiple benchmarks, with particularly pronounced gains on unseen tasks.
Affiliations
Joy Future Academy 1