Transformers in computer vision

What experts have said about Transformers in computer vision

1 statement · 1 negative

1. Eric JangNegativeMay 15, 2026· Dwarkesh Podcast

   Transformers require more data to learn invariant local features in vision tasks.

   “you do need more data there so that you can kind of learn through data the sort of invariant, local, local features”

   Listen at 34:40

   Open the episode · Eric Jang – Building AlphaGo from scratch

Statements are attributed to the speaker as said on the episode and reflect their view at the time, not PodLume's. They are not advice.