When it comes to enterprise use cases, getting actual value from an LLM usually involves feeding it custom data. There are many methods for equipping an LLM with data that don’t involve doing custom model training. In this session, we will demonstrate two of the most common methods: context windows and retrieval augmented generation (RAG). You’ll see the impact that adding custom data has to LLM results and how these methods are approachable for not just data scientists but a wide range of information workers.
Agenda:
- Introduction
- Data Augmentation vs. Custom Training
- What is a Context Window?
- Context Window demo
- What is Retrieval Augmented Generation (RAG)?
- RAG demo
- Q&A
