🔮The Codex
RAG (Retrieval-Augmented Generation)
A technique that gives AI access to external knowledge to improve accuracy.
📖 Apprentice Explanation
RAG is like giving an AI a reference library. Instead of relying only on what it learned during training, it can look up current, accurate information before answering.
🧙 Archmage Notes
RAG systems combine a retriever (often using vector embeddings and similarity search) with a generator (LLM). Key components include chunking strategies, embedding models, vector databases, and reranking. Advanced RAG includes query decomposition and multi-hop retrieval.