Exploring RAG's

What is the primary goal of Retrieval-Augmented Generation (RAG)?

• To generate text without external data

• To retrieve relevant documents and use them for better response generation

• To fine-tune language models with additional parameters

• To replace traditional machine learning models

Which of the following is NOT a component of a RAG system?

• Large Language Model (LLM)

• Vector Database

• Text-to-Speech Converter

• Embedding Model

What is the role of embeddings in RAG?

• They convert text into numerical vectors for efficient retrieval

• They directly generate answers for user queries

• They store raw text data for later use

• They improve the speed of response generation

Which library is commonly used to generate text embeddings in Python?

• pandas

• sentence-transformers

• matplotlib

• seaborn

In ChromaDB, what is stored in a vector database?

• Raw text data

• Numerical vector representations of documents

• Audio files

• Image data only

What does RecursiveCharacterTextSplitter do in RAG pipelines?

• Converts documents into embeddings

• Splits large documents into smaller, retrievable chunks

• Directly generates responses from documents

• Compresses documents to reduce storage

What is the purpose of using chunking in RAG?

• To improve document retrieval by creating smaller sections

• To increase response time by adding more data

• To prevent language models from accessing external data

• To replace embeddings with direct text search

What does top_k=5 represent in a retrieval query?

• The number of queries sent to the model

• The number of most relevant documents retrieved

• The number of words generated in the final response

• The number of times the query is repeated

Which API is used in your code to generate responses?

• OpenAI GPT

• Google Gemini

• Hugging Face Transformers

• Facebook LLaMA

Why is ChromaDB used in RAG-based applications?

• It provides a way to fine-tune LLMs

• It serves as a vector database to efficiently store and retrieve document embeddings

• It enhances the speed of text-to-speech conversions

• It translates documents into different languages