Contrastive learning prevents representation collapse in joint embeddings by providing the network both positive examples (views of the same image) and negative examples (different images), but requires large amounts of computation and may need exponentially more negative samples as representation dimensionality increases.

factualpending

Speaker

Unidentified Speaker — Yann LeCun's $1B Bet Against LLMs [Part 1] [kYkIdXwW2AE]

Evidence Quote

“requiring large amounts of computation and many negative examples to learn meaningful representations... the number of contrastive samples may grow exponentially with the dimension of the representation”

Source

Yann LeCun's $1B Bet Against LLMs [Part 1]— Welch Labs
Created: 8/12/2026, 6:35:53 PM

Your notes

Loading notes...