I asked them to show me their RAG pipeline...

RAG (Retrieval-Augmented Generation) is a widely adopted technique that gives LLMs access to external documents. We briefly discuss its roots in Information Retrieval and the transition from sparse to dense retrieval. Continua AI shares how they leverage HyDE (a query augmentation technique) to enhance their RAG pipeline in the context of group conversations / social AI. 📖 HyDE paper: https://arxiv.org/abs/2212.10496 🔗 Continua AI: https://continua.ai/ 🎤 Interview with David Petrou, CEO of Continua: https://www.patreon.com/posts/interview-with-143106016 🔗 Olga Dorabiala on LinkedIn: https://www.linkedin.com/in/olga-dorabiala-140930151/ 00:00 What is RAG? 02:02 RAG is rooted in Information Retrieval 03:50 RAG challenges 05:44 Continua AI uses HyDE for their RAG pipeline