Retrieval Augmented Generation (RAG)
Retrieval Augmented Generation (RAG) is a technique that grants generative artificial intelligence models information retrieval capabilities. It modifies interactions with a large language model (LLM) so that the model responds to user queries with reference to a specified set of documents, using this information to augment information drawn from its own vast, static training data. This allows LLMs to use domain-specific and/or updated information. Use cases include providing chatbot access to internal company data or giving factual information only from an authoritative source.
See:
Resources
- Retrieval-augmented generation - Wikipedia
- https://github.com/hymie122/RAG-Survey
- https://github.com/NirDiamant/RAG_Techniques
- RAG Survey & Available Research. A Survey on Retrieval-Augmented Text… | by Cobus Greyling | Medium
- Advanced Retriever Techniques to Improve Your RAGs | Towards Data Science
- Retrieval Augmented Generation (RAG) for LLMs | Prompt Engineering Guide (promptingguide.ai)
- RAG | 🦜️🔗 Langchain
- Q&A with RAG | 🦜️🔗 Langchain
- Compare Top 12 Retrieval Augmented Generation (RAG) Tools / Software
- Implementing a Retrieval-Augmented Generation (RAG) System with OpenAI’s API using LangChain | by ScaleX Innovation | Medium
- Training Your Own Dataset in Llama2 using RAG LangChain | by dmitri yanno mahayana | Medium
- langchain/cookbook/Semi_Structured_RAG.ipynb at master · langchain-ai/langchain (github.com)
- RAG and generative AI - Azure AI Search | Microsoft Learn
- ¿Qué es RAG?: explicación de la generación aumentada por recuperación: AWS (amazon.com)
- A RAG from scratch to query the scikit-learn documentation
Bases de datos vectoriales
- Chroma (trychroma.com)
- The vector database to build knowledgeable AI | Pinecone
- Welcome to Faiss Documentation — Faiss documentation
- https://www.pinecone.io/
- Vector stores | 🦜️🔗 Langchain
Agentic RAG
Graph RAG
See Graph RAG
Evaluation
See LLMs evaluation - RAG evaluation
Courses
- #COURSE Implement Retrieval Augmented Generation (RAG) with Azure OpenAI Service - Training | Microsoft Learn
Code
- #CODE AutoRAG - RAG AutoML tool for automatically finding an optimal RAG pipeline for your data
- #CODE Raglite - 🥤 RAGLite is a Python toolkit for Retrieval-Augmented Generation (RAG) with PostgreSQL or SQLite
- #CODE PGai - A suite of tools to develop RAG, semantic search, and other AI applications more easily with PostgreSQL