Introduction
Retrieval-Augmented Generation (RAG) represents a powerful paradigm in AI that bridges the gap between static language models and dynamic knowledge systems. Rather than a single technique, RAG encompasses a family of methods that share a common philosophy: enhancing language model outputs by incorporating external knowledge retrieved at inference time.
This hybrid architecture creates a flexible framework that can be adapted to various applications and domains. The RAG approach has spawned numerous implementations and variations, each optimizing different aspects of the retrieval-generation pipeline.
π¦ Auto-RAG
π¦ Corrective RAG
π¦ FLARE / Active RAG
π¦ GraphRAG
π¦ HybridRAG
π¦ InFO-RAG
π¦ Multi-Fusion Retrieval Augmented Generation (MoRAG)
π¦ R^2AG
π¦ Retrieval-Augmented Generation (RAG)
π¦ Reliability-Aware RAG (RA-RAG)
π¦ Self-RAG
π¦ Speculative RAG
Footnotes
-
Lewis, P., Perez, E., Piktus, A., Petroni, F., Karpukhin, V., Goyal, N., KΓΌttler, H., Lewis, M., tau Wen-Yih, RocktΓ€schel, T., Riedel, S., & Kiela, D. (2021). Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks. https://arxiv.org/abs/2005.11401 β©
Valeriia Kuka
Valeriia Kuka, Head of Content at Learn Prompting, is passionate about making AI and ML accessible. Valeriia previously grew a 60K+ follower AI-focused social media account, earning reposts from Stanford NLP, Amazon Research, Hugging Face, and AI researchers. She has also worked with AI/ML newsletters and global communities with 100K+ members and authored clear and concise explainers and historical articles.