

I don’t think it will make enough difference, but RAG stands for Retrieval Augmented Generation.
There’s a few ways to do it, but basically it’s a way add extra information to the conversation. By default the model only knows what it generates, plus what is in the conversation. RAG adds extra information to the mix.
The simplest approach is to scan the conversation for keywords and add information based on them.
So you ask “what is the capital of France” and instead of the model answering/hallucinating by itself, your app could send the full Wikipedia page for France along with your question, and the model will almost always return the correct answer from the Wikipedia page and hallucinate much less. In practice it gets a lot more complicated and I’m not up to date on recent RAG but the idea is the same.













Part of the book is that everything is slower, it took them weeks to get to the planet from where they met, and Rocky had already spent time retro fitting graces ship to let him live there. (they did not go to the planet with the intention of sampling it, they were just going to sample the line and orbit the planet as an easy way to get close to the line)
It’s been a long time since I read the book, but I think they considered the danger relatively low for the ship with their plan to stay high up and were more focused on “maybe we’ll lose the bucket risks” and didn’t want to go back to rocky ship until they were done because travel time and stuff.