H-JEPA: End-to-End Learning of Hierarchical World Models for Visual Planning
cs.LG, cs.RO
Submitted: 2026-10-05
Updated: 2026-10-05
License: http://creativecommons.org/licenses/by/4.0/
The gist: Long-horizon planning with latent world models requires reasoning across timescales and levels of abstraction.
Terminology
Abstract
Long-horizon planning with latent world models requires reasoning across timescales and levels of abstraction. Existing task-agnostic JEPA world models predict and plan at a single timescale or with multiple horizons in one shared latent space. We introduce H-JEPA, an end-to-end recipe for training a hierarchy of action-conditioned JEPAs in which each level predicts farther ahead in its own learned latent space. Planning proceeds top-down: the top level optimizes progress toward the goal, and each level's predictions become subgoals for the planner below it. When factors in the data evolve at separated timescales, higher levels discard fast, unpredictable detail and retain slower task-relevant state. Across four simulated navigation and manipulation environments, hierarchical planning improves over a flat JEPA; on Visual AntMaze, a three-level hierarchy raises success from 18% to 73% using less planner compute. Ablations attribute these gains to both temporal decomposition and higher-level goal representations. With inverse-dynamics supervision, the approach extends to diverse real-robot videos from DROID, where hierarchy improves offline planning fidelity at lower planner compute.
Sources
- Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecture
- V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planning
- Learning by Reconstruction Produces Uninformative Features For Perception
- LeJEPA: Provable and Scalable Self-Supervised Learning Without the Heuristics
- VICReg: Variance-Invariance-Covariance Regularization for Self-Supervised Learning
- Revisiting Feature Prediction for Learning Visual Representations from Video
- Genie: Generative Interactive Environments
- Emerging Properties in Self-Supervised Vision Transformers
- A Simple Framework for Contrastive Learning of Visual Representations
- Exploring Simple Siamese Representation Learning
- Diffusion Policy: Visuomotor Policy Learning via Action Diffusion
- Learning Universal Policies via Text-Guided Video Generation
- Visual Foresight: Model-Based Deep Reinforcement Learning for Vision-Based Robotic Control
- Provable RL with Exogenous Distractors via Multistep Inverse Dynamics
- Dynamics Learning with Cascaded Variational Inference for Multi-Step Manipulation
- Deep Visual Foresight for Planning Robot Motion
- Bootstrap your own latent: A new approach to self-supervised Learning
- World Models
- Learning Latent Dynamics for Planning from Pixels
- Dream to Control: Learning Behaviors by Latent Imagination
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks