Generative models that attempt to predict every pixel or detail of future video or high-dimensional continuous data produce blurry, averaged predictions and fail because most details in sensory data are fundamentally unpredictable.
factualpending
Speaker
Yann LeCunEvidence Quote
“most of the details in a video are completely unpredictable... you train the system to make one prediction and it can't really figure out which one will actually occur, so it predicts the average.”
Created: 8/12/2026, 10:03:11 PM
My Notes
Loading notes...