AI Dynamics

Global AI News Aggregator

Language Supervision Unnecessary for Multimodal Visual Representations

By

@ylecun

–

New paper from FAIR+NYU:
Q: Is language supervision required to learn effective visual representations for multimodal tasks? A: No.

→ View original post on X — @ylecun,

Commentaires

Leave a Reply

Your email address will not be published. Required fields are marked *

MORE ARTICLES