Representation collapse in joint embedding architectures—where networks trivially output identical embeddings for all inputs—was the primary unsolved problem preventing joint embeddings from learning useful representations until Barlow Twins and related approaches.

factualpending

Speaker

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

Evidence Quote

“This problem is known as representation collapse.”

Source

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

Your notes

Loading notes...