X · @ylecun
· X / Twitter
RT Haiyu Wu: Learning causality from internet videos in latent space first, and then using RL to teach the foundation model how to act. This approach …
RT Haiyu WuLearning causality from internet videos in latent space first, and then using RL to teach the foundation model how to act.This approach is 30× cheaper than Gemini 3.1 Flash on pretraining and achieves a better result. JEPA is all you need!However, actions are still learned during post-training. Figuring out