causal

Generative Video Prediction Fails on High-Dimensional Continuous Data

Building world models by training generative models to predict future video frames fails because you cannot represent probability distributions over high-dimensional continuous spaces (all possible frames) the way you can over a finite dictionary of text tokens; the world is too rich in unpredictable detail.

causalpending

Speaker

Yann LeCun

Evidence Quote

“Text is discrete, video is high-dimensional and continuous. A lot of details in this.”

Source

Research Pioneer— Lex Fridman Podcast
Created: 6/13/2026, 3:30:53 AM

Your notes

Loading notes...