RESEARCH PAPERYear 2026
DrivingWorld: Constructing World Model for Autonomous Driving via Video GPT
Classification
View four quadrants- Major category
- WAMs
- Architecture
- One Model
- Prediction paradigm
- Joint prediction
- Subcategories
- Autonomous drivingJoint video-action modeling
- Source review status
- Verified from primary sources
Category review. DrivingWorld uses one multimodal autoregressive predictor to generate future ego poses followed by future image tokens, and evaluates predicted ego trajectories for driving planning. This is a specialized driving world-action forecasting system, not a general historical foundation. Reading evidence
Contribution
A contribution summary has not been added yet.
Abstract
An abstract has not been added yet.
Affiliations
Not listed in the collection.