What are the algorithms used in RAG?

RAG employs several algorithms and techniques to enhance the performance of large language models (LLMs) by integrating external information retrieval capabilities. 

The key algorithms and components involved in RAG are dense vector representations, document retrieval algorithms, reciprocal rank fusion (RRF), prompt engineering techniques, and re-ranking algorithms.

The RAG process typically involves four main stages: indexing, retrieval, augmentation and generation.