3 ms·
Word2Vec-style vector arithmetic on docs embeddings
- antirez 1y agoIt works with image embeddings too: https://youtu.be/r6TJfGUhv6s?si=_LC0d4Mwyw18c53B https://youtu.be/r6TJfGUhv6s?si=_LC0d4Mwyw18c53B
- londons_explore 1y agoYou can probably make jointly trained decoder to turn a vector back into a new document which most closely matches. Would be cool to add together the vectors for harry potter and lord of the rings and then decode that into a new book about Frodo going to wizard school to collect the ring to help push Voldemort into mount doom.
- klysm 1y agoIsn’t that an auto encoder?
- Xx_crazy420_xX 1y agoLook at https://github.com/vec2text/vec2text https://github.com/vec2text/vec2text
- aDyslecticCrow 1y agoPrevious discussion https://news.ycombinator.com/item?id=45784455 https://news.ycombinator.com/item?id=45784455 https://news.ycombinator.com/item?id=45756599 https://news.ycombinator.com/item?id=45756599
- Xx_crazy420_xX 1y agoThis is really interesting! I've experimented with similar idea, but with time series forecasting on the sentence embeddings - https://github.com/Srakai/embcaster https://github.com/Srakai/embcaster. It turns out you can tokenise arbitrary information into constant vector which is really useful for later processing. The vec2text (https://github.com/vec2text/vec2text https://github.com/vec2text/vec2text) is an excellent asset if you want to reverse the embeddings back to text. This allows you to encode arbitrary data into standarized vectors, and all the way back.