AI Dynamics

Global AI News Aggregator

About

LeJEPA World Model Learning Under Gaussian Latent Dynamics

New paper from Yann LeCun! "When Does LeJEPA Learn a World Model?" This paper proves that under Gaussian latent dynamics, LeJEPA can recover the hidden state behind nonlinear observations up to rotation. The intuition is that linear latent features are the most stable across

→ View original post on X — @askalphaxiv,