Search⌘ K
AI Features

Creating Augmented Queries Using LangChain

Explore how to create and verify augmented queries in LangChain to refine the input for generative models. This lesson helps you understand the augmentation process, ensuring more accurate and context-rich responses in retrieval-augmented generation systems.

Now that we’ve covered how LangChain and ChromaDB handle indexing and retrieval, let’s dive into the next crucial step: augmentation. Imagine you’re on a treasure hunt, and you’ve just found a map. The map is a bit faded, and you need to enhance it to see all the details clearly. In the world of RAG systems, our “map” is the retrieved information, and augmentation is the process of removing that blurriness.

Think of it this way: when we ask a question, our system embarks on a computational adventure, fetching bits of information from various sources, as we saw in the previous lesson. But before this information is handed over to the generator component—which is responsible for the final response—it needs to undergo preprocessing or, in our context, augmentation.

The relationship between prompt template and knowledge base is the augmented query
The relationship between prompt template and knowledge base is the augmented query

In this lesson, we’ll explore how augmentation works in ...