Papers
arxiv:2602.23058

GeoWorld: Geometric World Models

Published on Feb 26
· Submitted by
Zeyu Zhang
on Feb 27
Authors:
,
,
,

Abstract

GeoWorld addresses limitations in energy-based predictive world models by utilizing hyperbolic geometry to preserve latent state structures and improve long-horizon prediction performance.

AI-generated summary

Energy-based predictive world models provide a powerful approach for multi-step visual planning by reasoning over latent energy landscapes rather than generating pixels. However, existing approaches face two major challenges: (i) their latent representations are typically learned in Euclidean space, neglecting the underlying geometric and hierarchical structure among states, and (ii) they struggle with long-horizon prediction, which leads to rapid degradation across extended rollouts. To address these challenges, we introduce GeoWorld, a geometric world model that preserves geometric structure and hierarchical relations through a Hyperbolic JEPA, which maps latent representations from Euclidean space onto hyperbolic manifolds. We further introduce Geometric Reinforcement Learning for energy-based optimization, enabling stable multi-step planning in hyperbolic latent space. Extensive experiments on CrossTask and COIN demonstrate around 3% SR improvement in 3-step planning and 2% SR improvement in 4-step planning compared to the state-of-the-art V-JEPA 2. Project website: https://steve-zeyu-zhang.github.io/GeoWorld.

Community

Accepted to CVPR 2026.

the project page is not available.

·

I still need more time to work on the website, but more information has been included in the paper.

Sign up or log in to comment

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2602.23058 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2602.23058 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2602.23058 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.