LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels 事件

LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels arXiv:2603.19312v3 Announce Type: replace-cross Abstract: Joint Embedding Predictive Architectures (JEPAs) offer a compelling framework for learning world models in compact latent spaces, yet existing methods remain fragile, relying on complex multi-term losses, exponential moving averages, pre-trained encoders, or auxiliary supervision to avoid representation collapse. In this work, we introduce LeWorldModel (L