View source for Being-Creative-with-AI/C4/Rag-and-its-Uses/English

'''Title of the Script: RAG and its Uses.'''

'''Author: EduPyramids'''

'''Keywords: RAG, Retrieval, Generation, Chatbot, AI, EduPyramids, video tutorial.'''

{|border=1
|-
|| '''Visual Cue'''
|| '''Narration'''
|-
|| '''Slide 1 '''

'''Title Slide'''
|| Welcome to this '''Spoken Tutorial''' on '''RAG and its Uses'''.
|-
|| '''Slide 2 '''

'''Learning Objectives'''
|| In this tutorial, we will learn: 
* How an '''AI''' model answers questions using additional information
* What is '''Retrieval-Augmented Generation (RAG)'''
* How '''RAG''' helps improve the accuracy of answers
|-
|| '''Slide 3'''

'''Disclaimer Slide'''

As '''AI''' tools constantly evolve, if you are unable to locate any icon or encounter difficulty at any step, you may use any conversational '''AI''' '''Chatbot''' for guidance.
|| As '''AI''' tools constantly evolve, if you are unable to locate any icon or encounter difficulty at any step, you may use any conversational '''AI''' '''Chatbot''' for guidance.
|-
|| '''Slide 4'''

'''System Requirements'''
|| To record this tutorial, I am using: 
*'''Ubuntu 24.04 LTS, '''and
*'''Firefox '''version''' 148.0.2'''

Learners will also need a working internet connection
|-
|| '''Slide 5'''

'''Prerequisites'''
|| To follow this tutorial, 
* Learners should know basic computer and internet usage. 
* No prior '''AI''' knowledge is required. 
|-
|| 
|| Let us get started.
|-
|| '''Show grocery app order (1 kg onions)'''
|| Imagine this: you ordered one kilogram of onions from a grocery app.
|-
|| '''Show wrong delivery (tomatoes)'''
|| But you receive one kilogram of tomatoes''' '''instead.
|-
|| '''Show support chat opening'''
|| To resolve this, you open the support section of the app.

You ask the support chatbot: “I got the wrong item. Can I return it?”
|-
|| '''Show chatbot response'''
|| The '''chatbot''' powered by '''AI''' gives you an answer.
|-
|| '''Pause on screen (AI thinking?)'''
|| But how does it know what to say?
|-
|| '''Show AI model trained on large data'''
|| The '''AI model''' is trained on a large amount of general text data.
|-
|| '''Show How AI is helping to understand languages and generating responses as well'''
|| This helps it understand language and generate responses.
|-
|| '''Highlight limitation'''
|| However, this training does not include the app’s return policy.

It is unique to each company
|-
|| '''Show return policy document'''
|| The return policy exists as a separate document on the company’s server.
|-
|| '''Show secure storage'''
|| This document is stored securely and is not part of the '''model’s''' training.
|-
|| '''Show search inside documents'''
|| When you ask the question, the system first searches the policy document.
|-
|| '''Highlight relevant lines'''
|| It finds the most relevant information related to your query.
|-
|| '''Show info + question sent to AI'''
|| This information is sent along with your question to the '''AI model.'''
|-
|| '''Show final response generation'''
|| The '''AI model''' then generates an answer based on this information.
|-
|| '''Highlight term RAG'''
|| This process is called '''Retrieval-Augmented Generation''', or '''RAG'''.
|-
|| '''Show chatbot without policy'''
|| Let us see what happens without a policy document for the '''chatbot'''.

When a user asks “I got the wrong item in a grocery app. Can I return it?”

Without specific information, the '''chatbot''' gives a general response.

The response may not match the company’s actual return policy.
|-
|| '''Show policy document'''
|| Now imagine the same question is asked with a policy document.
|-
|| '''Display policy points'''
|| The policy states that items can be returned within 24 hours. 

Perishable items cannot be returned.

Refunds take 3 to 5 business days.
|-
|| '''Ask same question again '''
|| We ask the same question again.
|-
|| '''Show improved answer '''
|| Observe the new response.
|-
|| '''Highlight improvement'''
|| The answer now clearly follows the given policy.
|-
|| '''Compare both responses'''
|| The first response is general. The second response is accurate and specific.
|-
|| '''Embedding / vector visual'''
|| The question is converted into a machine-readable form.
|-
|| '''Search visual'''
|| The system searches for similar information in the document.
|-
|| '''Highlight retrieval'''
|| The most relevant part is retrieved.
|-
|| '''Send to model '''
|| This retrieved content is sent to the '''AI model.'''
|-
|| '''Final answer generation.'''
|| The '''model''' generates the final answer using this information.
|-
|| '''Slide 6'''

'''Key Idea behind RAG'''
* '''RAG helps the AI retrieve relevant information and use it to generate accurate answers.'''
* '''Question → Retrieve → Generate Answer'''
|| This is the key idea behind '''RAG'''.

'''RAG''' lets '''AI''' fetch relevant data and generate more accurate answers.
|-
|| 
|| With this, we come to the end of this tutorial.
|-
|| '''Slide 7'''

'''Summary'''

In this tutorial, we learnt to: 
* How an '''AI''' model answers questions using additional information,
* What is '''Retrieval-Augmented Generation(RAG)''', and 
* How '''RAG''' helps improve the accuracy of answers.
|| In this tutorial, we learnt to: 
* How an '''AI''' model answers questions using additional information,
* What is '''Retrieval-Augmented Generation (RAG)''', 
* How '''RAG''' helps improve the accuracy of answers.
|-
|| '''Slide 8'''

'''Assignment'''

'''As an Assignment,'''
* '''Create your own return policy document with at least three rules.'''
* '''Ask a chatbot the same question with and without the document.'''
* '''Compare the responses and observe the difference.'''
|| We encourage you to do this assignment.
|-
|| '''Slide 9'''

'''Acknowledgement'''

'''Domain Inputs: Bhavani Shankar R and Saisudha Sugavanam '''

'''Script Writer: Ketki Naina'''

'''Admin Reviewer: Arthi Varadarajan'''

'''Quality Reviewer: Sakina Sidhwa'''

'''Novice Reviewer: Misbah Samir'''

'''AI Narration: Debosmita Mukherjee'''

'''AI Graphics: Arvind Pillai'''

'''Video Editor: Arvind Pillai'''

'''Web Developer: Ankita Singhal'''
|| 

Thank you for watching.
|-
|| '''Slide 10'''

'''Acknowledgement'''

This Spoken Tutorial is brought to you by EduPyramids Educational Services Private Limited at SINE, IIT Bombay. 
|| 
|-
|}