RAG

Artificial Intelligence/retrieval-augmented-generative-rag.md

@@ -0,0 +1,7 @@
+# Retrieval-Augmented Generation (RAG)
+
+Currently LLMs have these 2 problems:
+1. They lack valid source while generating text.
+2. They are often out-of date.
+
+The RAG framework is a new approach to solve these problems. In this appraoch a LLM first finds relevant passages from a large corpus (content store like the Internet) and then generates text based on user prompt. This way it solves both the problems mentioned above.

README.md

@@ -27,7 +27,7 @@
 ## Categories
 * [AWS](#aws) [**`1`**]
 * [Android](#android) [**`1`**]
-* [Artificial Intelligence](#artificial-intelligence) [**`1`**]
+* [Artificial Intelligence](#artificial-intelligence) [**`2`**]
 * [Career](#career) [**`7`**]
 * [Coding Practices](#coding-practices) [**`5`**]
 * [Community Building](#community-building) [**`3`**]
@@ -75,6 +75,7 @@
 
 <ul>
 <li><a target="_blank" href="https://github.com/Bhupesh-V/til/blob/master/Artificial Intelligence/fundamentals-of-llms.md">Fundamentals of LLMs</a>
+<li><a target="_blank" href="https://github.com/Bhupesh-V/til/blob/master/Artificial Intelligence/retrieval-augmented-generative-rag.md">Retrieval-Augmented Generation (RAG)</a>
 </ul>
 
 

SUMMARY.md

@@ -20,6 +20,7 @@
 
 <ul>
 <li><a href="Artificial Intelligence/fundamentals-of-llms.md">Fundamentals of LLMs</a>
+<li><a href="Artificial Intelligence/retrieval-augmented-generative-rag.md">Retrieval-Augmented Generation (RAG)</a>
 </ul>
 
 ## Career

count.json

@@ -1,3 +1,3 @@
 {
- "count": 142
+ "count": 143
 }