Editorial definition preserved directly in static HTML.
Self-supervised world models learn representations of environment dynamics without explicit labels or reward signals. They leverage prediction in abstract representation space, predicting future states, video frames, or embeddings, to build internal models of how the world works.
World models that learn environment dynamics through self-supervised prediction without explicit labels.
Static category overview generated from local editorial data.
| Attribute | Value |
|---|---|
| Category | Self-Supervised World Models |
| Description | World models that learn environment dynamics through self-supervised prediction without explicit labels. |
| Definition | Self-supervised world models learn representations of environment dynamics without explicit labels or reward signals. They leverage prediction in abstract representation space, predicting future states, video frames, or embeddings, to build internal models of how the world works. |
| Related Models | 3 |
| Related Research | 1 |
| Related Guides | 0 |
Additional editorial context preserved directly in static HTML.
Self-supervised world models reduce dependence on explicit rewards and labels, making them especially relevant for large-scale pretraining from raw video or interaction data.
What matters is whether the representation transfers well, preserves predictive structure, and supports later planning or control rather than only whether the pretraining objective looks elegant.
| Model | Lab | Category | Year |
|---|---|---|---|
| V-JEPA | Meta | Self-Supervised World Model | 2024 |
| World Models (Ha & Schmidhuber) | Google Brain / IDSIA | Model-Based RL | 2018 |
| RSSM | Latent Dynamics | 2019 |
| Topic | Summary |
|---|---|
| Self-Supervised World Models | How self-supervised world models learn environment dynamics without rewards, from JEPA and V-JEPA to predictive latent representations. |
FAQ answers rendered directly into static HTML for extractable responses.
They learn by predicting future observations or latent representations without explicit reward labels, often through contrastive, masked, or JEPA-style objectives.
Short extractable summary preserved directly in static HTML.
Editorial provenance and refresh policy preserved directly in static HTML.
Published by world-models.io editorial board.
Lead editor Bernard Grenat.
This category page maintains a stable editorial definition and connects it to related models, research topics, guides, and source-backed context.
Each editorial page is assembled from primary sources, normalized into extractable summaries, checked for factual drift, and reviewed before publication or major refreshes. Last reviewed: 2026-06-21.
Pages are refreshed when a new paper, benchmark, release, architecture update, or stronger primary source materially changes the answer a reader or AI system should retrieve.
Each page links back to relevant primary sources and keeps a stable canonical URL so readers can verify claims, trace context, and reference the most up-to-date version. See the editorial policy.
Representative external references connected to this category through related models and research topics.