RESEARCH PAPERYear 2025

DINO-WM: World Models on Pre-trained Visual Features enable Zero-shot Planning

Gaoyue Zhou; Hengkai Pan; Yann LeCun; Lerrel Pinto

Classification

Major category: Foundational work
Quadrant: Not applicable
Architecture: Not applicable
Prediction paradigm: Not applicable
Subcategories: Classical world models & model-based RL Visual encoders & representations
Source review status: Verified from primary sources

Category review. DINO-WM is a pre-2026 foundation for decoder-free predictive control: frozen spatial features, action-conditioned latent dynamics, and external CEM/MPC. It is explicitly part of the manuscript world-model thread; using DINO does not make the paper itself a generic vision 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

@inproceedings{dino_wm,
  author       = {Gaoyue Zhou and
                  Hengkai Pan and
                  Yann LeCun and
                  Lerrel Pinto},
  title        = {{DINO-WM:} World Models on Pre-trained Visual Features enable Zero-shot
                  Planning},
  booktitle    = {ICML},
  year         = {2025},
}

Classification

Contribution

Abstract

Affiliations

BibTeX

More in this category

Critique of World Model

Precise and Dexterous Robotic Manipulation via Human-in-the-Loop Reinforcement Learning

Mean Flows for One-step Generative Modeling