From the course: Foundations of AI and Machine Learning for Java Developers
Join today to access over 24,500 courses taught by industry experts.
Retrieval-augmented generation (RAG) - Java Tutorial
From the course: Foundations of AI and Machine Learning for Java Developers
Retrieval-augmented generation (RAG)
- [Instructor] While prompt techniques can work well, they rely purely on the training of the foundation model. It is extremely expensive to create a new foundation model that uses up-to-date information. In another video, we learned that providing context to a language model is critically important. There I manually copied and pasted additional information into the prompt, so the language model had your prompt plus additional context, which gave us a better result, but the process was awkward and clumsy. Also, with simple prompt techniques, there are distinct chances of made up responses, or what's called hallucinations, for the large language model. Instead of using a manual process, Meta AI research developed a technique called retrieval-augmented generation, or RAG, which was first used back in 2020. RAG solves a few problems by using just prompts. Instead of manually adding relevant context to the prompt, the RAG architecture uses an external data storage subsystem to store and…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.