RESEARCH PAPER

LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels

Lucas Maes; Quentin Le Lidec; Damien Scieur; Yann LeCun; Randall Balestriero

Classification

View four quadrants
Major category
WAMs
Architecture
Not applicable
Prediction paradigm
Other mechanisms
Source review status
Verified from primary sources

Category review. LeWM is a 2026 learned latent-dynamics system used by CEM to choose goal-reaching actions and replan from observations. Under the manuscript system-level predictive-control view, learned futures and external optimization jointly determine action; it is a contemporary method, not a pre-2026 foundation or generic DINO-style encoder. Reading evidence

AT A GLANCE

Contribution

A contribution summary has not been added yet.

Abstract

An abstract has not been added yet.

Affiliations

Not listed in the collection.

BibTeX