LeCun's recent talks and papers on JEPA (Joint-Embedding Predictive Architecture) and energy-based models. His core argument: autoregressive LLMs are fundamenta

LeCun's recent talks and papers on JEPA (Joint-Embedding Predictive Architecture) and energy-based models. His core argument: autoregressive LLMs are fundamentally flawed because they predict in pixel/token space rather than abstract representation space.