I assume this comes up time and time again when models release new context windows, but curious where RAG, vector dbs, etc comes into play with these attention focused models?
hey all this has probably been asked but I have a pdf and I want to extract information like address and author in metadata? thoughts on how to do this? TIA