Joint Embedding Predictive Architecture (JEPA)
A neural architecture that encodes both current observation and future observation into abstract representation spaces, then predicts the future representation from the current representation, rather than predicting pixels directly
As defined in The Shape of AI to Come! Yann LeCun at AI Action Summit 2025
Other definitions
A self-supervised learning architecture that encodes both inputs and targets into abstract representation space and trains a predictor to match representations rather than predict pixels or raw outputs, allowing the encoder to naturally eliminate unpredictable details
The Shape of AI to Come! Yann LeCun at AI Action Summit 2025
Where it comes up · 5
- What Is Yann LeCun Cooking? JEPA Explained Simply11 claims
- Metas AI Boss Says He DONE With LLMS...8 claims
- The Shape of AI to Come! Yann LeCun at AI Action Summit 20256 claims
- Yann LeCun's $1B Bet Against LLMs [Part 1]5 claims
- Yann LeCun: World Models: Enabling the next AI revolution4 claims
Most important claims · 10 of 34Search all claims for “Joint Embedding Predictive Architecture (JEPA)” →
- Inference in JEPA-based systems becomes a search process over predicted futures rather than autoregressive prediction, fundamentally changing how these systems generate behavior.Yann LeCun's $1B Bet Against LLMs [Part 1]
- JEPA (Joint Embedding Predictive Architecture) avoids the pixel prediction problem by mapping observations to embeddings and training a predictor to forecast future embeddings rather than pixels, allowing the model to focus on salient scene features instead of random unpredictable details.Yann LeCun's $1B Bet Against LLMs [Part 1]
- VJEPA-2 conditions JEPA on action signals sent to a robot arm, allowing the predictor to learn how control signals change the robot's position in embedded space, enabling planning by exploring the model's predictions to find actions matching a goal state.Yann LeCun's $1B Bet Against LLMs [Part 1]
- Training JEPA architectures requires designing cost functions that produce low energy on training data but high energy outside the training manifold; there are only two classes of methods for this: (1) contrastive methods push down energy on real data and push up energy on synthetic negative samples, but they don't scale well to high dimensions and require many negative samples; (2) regularized methods use regularizers to minimize the volume of low-energy space, which scales better.The Shape of AI to Come! Yann LeCun at AI Action Summit 2025
- Representation collapse is a failure mode where both the context and target encoders in JEPA can learn to output the same constant embedding for all inputs, with the predictor simply outputting a constant embedding that always matches, resulting in extremely low training loss but learning absolutely nothing about the world.What Is Yann LeCun Cooking? JEPA Explained Simply
- JEPA performs world simulation directly in latent space where action sequences and state transitions are represented as embeddings, saving compute compared to generating full video frames by operating in representation space rather than pixel space.What Is Yann LeCun Cooking? JEPA Explained Simply
- For complex JEPA applications, the latent space must be structured to support planning, with the predictor learning smooth transitions between states through conditional actions, so changes in latent state make sense and predicted trajectories don't drift or accumulate errors.What Is Yann LeCun Cooking? JEPA Explained Simply
- LeJepa can train JEPA-style models without relying on EMA while still avoiding collapse and learning strong representations, achieving competitive or better performance compared to VICReg, Barlow Twins, and SimCLR on standard vision benchmarks, with accuracy levels similar to state-of-the-art DINO on ImageNet.What Is Yann LeCun Cooking? JEPA Explained Simply
- The idea of a cat sitting on a couch can be captured in a high-dimensional latent space with all kinds of sentences or images that show this idea pointing to that spot in the latent space, demonstrating that JEPA's representation captures semantic invariants across diverse inputs.What Is Yann LeCun Cooking? JEPA Explained Simply
- A view in JEPA is a transformation, masking, or partial observation of an input that preserves the underlying semantic state while hiding or removing some information, such as different image crops, masked versions, camera angles, or video frames of the same scene.What Is Yann LeCun Cooking? JEPA Explained Simply