Building Chatbots with Retrieval-Augmented Generation Techniques

Explore the intricacies of building chatbots using Retrieval-Augmented Generation (RAG) techniques in this conference talk from Conf42 LLMs 2024. Delve into the comparison between human doctors and AI-powered medical assistants, examining the simplest approach to chatbot creation before diving deep into the RAGDoctor methodology. Learn about crucial components such as patient case datasets, vector databases for efficient searching, bio-embedding models, and specialized biomedical domain language models. Discover advanced techniques like query rewriting and hybrid search to enhance chatbot performance. Gain valuable insights into the practical application of RAG in the medical field, with a step-by-step breakdown of the process from initial concept to sophisticated implementation.

intro
preamble
the average human doctor
simplest approach
ragdoctor approach
patient cases dataset
vector database - search
bio embedding models
biomedical domain llm
can we do event better?
query rewriting
hybrid search