Predictive representation learning
Yann LeCun
Coauthor of I-JEPA and V-JEPA 2, predictive architectures for learning representations and planning.
- Tracked papers
- 8
- Past year
- 5
- Editor picks
- 1
Proving and Improving the Platonic Representation Hypothesis with JEPAs
The authors prove a “Platonic Representation Theorem” for JEPAs: if modalities share the same positive-pair distribution formed from two observations of the same scene (“good views”), their learned representations match up to one rotation. …
LeHami: Symplectic Latent Dynamics for World Models
LeHami is a JEPA-style latent world model where the predictor is replaced by an action-conditioned Hamiltonian flow on a structured latent phase space (q,p), integrated with an exactly symplectic leapfrog step. The authors prove the rollout…
Preserving Unstable Modes Through Inverse Dynamics in JEPA World Models
The authors show that standard JEPA training (next-step prediction + anti-collapse like SIGReg) can minimize loss while collapsing controllable unstable modes, preventing stabilization from the learned latent representation. They add an end…
H-JEPA: End-to-End Learning of Hierarchical World Models for Visual Planning
H-JEPA is an end-to-end hierarchical, action-conditioned JEPA world model. It trains multiple JEPA levels where each level predicts farther ahead in its own learned latent space. Top-down planning uses higher levels to predict subgoals for …
AD-E2E-JEPA: A Joint-Embedding Predictive Architecture For End-to-End Autonomous Driving
AD-E2E-JEPA is an action-conditioned JEPA world model for end-to-end autonomous driving that enables goal-conditioned zero-shot planning without training a driving policy. It adds a SIGReg-regularized learnable projector that compresses pat…
V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planning
Pretrains an action-free JEPA video model on over 1 million hours of internet video, then post-trains a latent action-conditioned world model (V-JEPA 2-AC) on under 62 hours of unlabeled robot video and uses it to plan pick-and-place with i…
Revisiting Feature Prediction for Learning Visual Representations from Video
V-JEPA trains video models solely with a feature-prediction objective (predicting masked spatio-temporal regions in representation space), with no pretrained image encoders, text, negatives or reconstruction, on 2 million public videos.
Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecture
I-JEPA learns image representations by predicting the representations of several target blocks of an image from a single context block, in representation space and with no hand-crafted data augmentations.