3 ms·
Language Quantized AutoEncoders: Towards Unsupervised Text-Image Alignment
- sebg 4y agoTwitter Thread that led me to the link: https://twitter.com/0xmaddie_/status/1627777203013398528 https://twitter.com/0xmaddie_/status/1627777203013398528 Specifically this language: " if i'm understanding this correctly, you can use a pure text encoder model to find text that lets you reconstruct an image from the text encoding. basically, the latent space of a text model is expressive enough to serve as a compilation target for images "