Improving the retriever step is the most effective way to increase the accuracy and relevance of a RAG system. You need a solid toolkit to build customized, highly efficient systems. This chapter introduces advanced retrieval techniques that go beyond a basic vector search. Figure 7-1 provides an overview of techniques for optimizing the pre-retrieval and retrieval steps.
Figure 7-1. Techniques to enhance the retrieval process
Real-world RAG workflows typically combine several of these techniques, depending on the use case. Figure 7-2 shows an example workflow that combines multiple steps and advanced retrieval techniques. The workflow illustrates three key stages: (1) breaking a complex query into focused subqueries, (2) routing each subquery to the appropriate tool or data source to retrieve relevant information, and (3) reasoning across the answers to synthesize a comprehensive response.
Figure 7-2. A multistep retrieval process
Think of the recipes in this chapter as building blocks that you can combine to improve your workflow and achieve better search results. You need a toolkit of advanced retrieval techniques that you can apply in the way that best fits your use case.
Table 7-1 summarizes seven techniques discussed in this chapter. You should understand ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month, and much more.
O’Reilly covers everything we've got, with content to help us build a world-class technology community, upgrade the capabilities and competencies of our teams, and improve overall team performance as well as their engagement.
Julian F.
Head of Cybersecurity
I wanted to learn C and C++, but it didn't click for me until I picked up an O'Reilly book. When I went on the O’Reilly platform, I was astonished to find all the books there, plus live events and sandboxes so you could play around with the technology.
Addison B.
Field Engineer
I’ve been on the O’Reilly platform for more than eight years. I use a couple of learning platforms, but I'm on O'Reilly more than anybody else. When you're there, you start learning. I'm never disappointed.
Amir M.
Data Platform Tech Lead
I'm always learning. So when I got on to O'Reilly, I was like a kid in a candy store. There are playlists. There are answers. There's on-demand training. It's worth its weight in gold, in terms of what it allows me to do.
O’Reilly covers everything we've got, with content to help us build a world-class technology community, upgrade the capabilities and competencies of our teams, and improve overall team performance as well as their engagement.
