AdaJEPA: An Adaptive Latent World Model
Evolving story · 1 updatesAdaJEPA: Adaptive Latent World Model for Robust PlanningTimeline →Researchers propose AdaJEPA, an adaptive latent world model that performs test-time adaptation within model predictive control to improve planning accuracy under distribution shifts.
Latent world models enable planning from high-dimensional observations by predicting future states in a compact latent space. However, these models are typically kept frozen at test time: when their predictions become inaccurate, planning can fail, especially under test-time distribution shift. To address this, we propose AdaJEPA, an adaptive latent world model that performs test-time adaptation within the closed loop of model predictive control (MPC). After training, AdaJEPA plans and executes the first action chunk, uses the observed next-state transition as a self-supervised adaptation sign
Source: AdaJEPA: An Adaptive Latent World Model. Read the full piece at the source.
Summary and analysis generated by AI (mistral). Always verify against the original sources.