Fusing shallower encoder layers into a Representation Autoencoder (RAE) sharpens pixel reconstruction but degrades diffusion generation. Prior studies and the research community commonly attribute this trade-off to the latent content itself, assuming detail-rich features are harder to model. We show that this gap stems...
Hong-Yang Du, Yun-Fei Xie, Jun-Jie Ye et al.· 1 citation
Joint-Embedding Predictive Architectures (JEPA) offer an appealing approach to self-supervised learning, but are typically prone to representation collapse. LeJEPA resolves this by introducing a theoretically grounded distribution-matching objective that drives the embedding towards an isotropic Gaussian. However, this...
Kartik Tandon, Alessio Borgi, Pavel Snopov et al.· 0 citations
SpecWM is introduced, a JEPA framework that explicitly specifies desired relationships between observations through a target similarity kernel and trains the encoder to reproduce those relationships in its embedding geometry and suggests that explicitly specifying representation geometry provides a simple way to steer...
Holger Molin, William Peng, Marco Bagatella et al.· 0 citations
Joint-embedding predictive architectures (JEPAs) learn compact latent world models without pixel reconstruction. Existing JEPA pipelines typically train visual representations and action-conditioned dynamics before adding a control policy or solver. We introduce LeWAM, an end-to-end JEPA for control with: (1) a joint t...
H-JEPA is introduced, 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 in which each level's predictions become subgoals for the planner below it.
Wan-Cong Zhang, Basile Terver, Michael G. Rabbat et al.· 14 citations· ⚡4
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.