# Subspace-Decomposed JEPAs: Disentangling Progression and Content in Latent World Models

arXiv: https://arxiv.org/abs/2605.31111 · HTML: https://mlsanity.com/paper/2605.31111/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Placed in Latent predictive world models, one of 11 directions in World models.
- Builds on 20 of its 26 references that are in the map.
- First posted on arXiv 29 May 2026.

## Placed in

- World models / [Latent predictive world models](https://mlsanity.com/wm/latent-predictive-world-models/)

## Builds on (most cited first, 20 of 20)

- [Bootstrap your own latent: A new approach to self-supervised Learning](https://mlsanity.com/paper/2006.07733/) (2020)
- [Mastering Diverse Domains through World Models](https://mlsanity.com/paper/2301.04104/) (2023)
- [Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecture](https://mlsanity.com/paper/2301.08243/) (2023)
- [DeepMind Control Suite](https://mlsanity.com/paper/1801.00690/) (2018)
- [Cosmos World Foundation Model Platform for Physical AI](https://mlsanity.com/paper/2501.03575/) (2025)
- [Genie: Generative Interactive Environments](https://mlsanity.com/paper/2402.15391/) (2024)
- [Latent Action Pretraining from Videos](https://mlsanity.com/paper/2410.11758/) (2024)
- [LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels](https://mlsanity.com/paper/2603.19312/) (2026)
- [LeJEPA: Provable and Scalable Self-Supervised Learning Without the Heuristics](https://mlsanity.com/paper/2511.08544/) (2025)
- [OGBench: Benchmarking Offline Goal-Conditioned RL](https://mlsanity.com/paper/2410.20092/) (2024)
- [GAIA-2: A Controllable Multi-View Generative World Model for Autonomous Driving](https://mlsanity.com/paper/2503.20523/) (2025)
- [Temporal Cycle-Consistency Learning](https://mlsanity.com/paper/1904.07846/) (2019)
- [Learnable latent embeddings for joint behavioral and neural analysis](https://mlsanity.com/paper/2204.00673/) (2022)
- [On the duality between contrastive and non-contrastive self-supervised learning](https://mlsanity.com/paper/2206.02574/) (2022)
- [HunyuanWorld 1.0: Generating Immersive, Explorable, and Interactive 3D Worlds from Words or Pixels](https://mlsanity.com/paper/2507.21809/) (2025)
- [Brain-JEPA: Brain Dynamics Foundation Model with Gradient Positioning and Spatiotemporal Masking](https://mlsanity.com/paper/2409.19407/) (2024)
- [Learning Latent Action World Models In The Wild](https://mlsanity.com/paper/2601.05230/) (2026)
- [Temporal Straightening for Latent Planning](https://mlsanity.com/paper/2603.12231/) (2026)
- [AI-Generated Video Detection via Perceptual Straightening](https://mlsanity.com/paper/2507.00583/) (2025)
- [Joint Embedding Predictive Architectures Focus on Slow Features](https://mlsanity.com/paper/2211.10831/) (2022)

## Built on by (most cited first, 11 of 11)

- [Fast LeWorldModel](https://mlsanity.com/paper/2606.26217/) (2026)
- [Temporal-Distance JEPA: Plan-Aware Representation Learning for Latent World Model Predictive Control](https://mlsanity.com/paper/2607.25337/) (2026)
- [PhyLatent: Learning Dynamics-Relevant Representations for JEPA World Models](https://mlsanity.com/paper/2608.05720/) (2026)
- [QQWorld: Quantile-Quantile Matching for World Model Regularization](https://mlsanity.com/paper/2607.28415/) (2026)
- [DWM: Separating World Effects from Actions in Latent World Models](https://mlsanity.com/paper/2607.18715/) (2026)
- [VIScore: Diagnosing Planning-Relevant Quality in Latent World Models](https://mlsanity.com/paper/2608.11174/) (2026)
- [Support Operation Factorization: Compositional Readout of Frozen Vision Encoders under Controlled Interventions](https://mlsanity.com/paper/2608.06174/) (2026)
- [Capturing Cardiac Cyclicity through Phase-Equivariant Self-Supervised Learning](https://mlsanity.com/paper/2608.21147/) (2026)
- [Beyond One-Step Accuracy: State-Affine Latent Transition for Reliable Visual Planning](https://mlsanity.com/paper/2609.33595/) (2026)
- [LRC-JEPA: Disentangling Dynamics and Residual Context for Efficient World Models](https://mlsanity.com/paper/2609.34375/) (2026)
- [Beyond a single latent space: a dual-latent world model for long-horizon planning](https://mlsanity.com/paper/2609.37644/) (2026)

## Abstract (arXiv)

World models used for planning learn to predict how a scene evolves, yet they rarely hold a global view of how far a task has progressed. We argue that such a view should be part of the representation, and that it can be compact: describing a scene takes many dimensions, but where the scene stands on its way to the goal can be summarised by only a few. We propose SD-JEPA, a joint-embedding predictive world model whose latent is split into a small progression subspace and a large content subspace. A temporal triplet loss keeps consecutive frames of a trajectory close in the progression coordinates and separates different trajectories, while a distributional regulariser keeps the content coordinates from collapsing. On four control benchmarks, the model plans as well as the same architecture without the split. In return, its latent becomes easier to read. A few percent of its dimensions are enough to predict progress within an episode, and the angle they form follows the task, moves backward when the agent backtracks and changes fastest around events such as grasps and releases. Surprise also becomes anchored in the trajectory: when the scene changes mid-episode, the angle spikes and then relocalizes on a new part of the progression, anchoring surprise in the agent's own map of the task.