Kamus AI
RAG
A way to give a model relevant document passages before it answers, so the answer rests on sources.
RAG (retrieval-augmented generation) works in two steps: find the most relevant document passages, then send them with the question to the model. Results are easier to check because you know the source. RAG reduces hallucination but does not remove it: the model can still misread a source or add things that are not there.
Also called: RAG, retrieval, retrieval-augmented generation