Word2vec-Style Vector Arithmetic on Docs Embeddings

You 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.

It works with image embeddings too: https://youtu.be/r6TJfGUhv6s?si=_LC0d4Mwyw18c53B

Previous discussion

https://news.ycombinator.com/item?id=45784455

https://news.ycombinator.com/item?id=45756599

This is really interesting! I've experimented with similar idea, but with time series forecasting on the sentence embeddings - 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) 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.