Balakrishna Sreeram

Retrieval-Augmented Generation: Architecting Trustworthy Knowledge Systems on Top of Large Language Models

This review treats RAG as a systems-engineering discipline, and concludes that retrieval quality, not generator choice, is what governs whether a RAG system can be trusted.