What does RAG stand for?	Retrieval Augmented Generation
What is a major limitation of putting an entire large document into a single prompt?	There is a hard limit on the amount of text that can be fed into the model.
How does RAG address the problem of large documents?	It breaks the document into small chunks and retrieves only the most relevant chunks for the user's question.
What is a key advantage of RAG regarding document size?	It can scale up to handle very large documents and multiple documents.
How does RAG improve efficiency and cost compared to feeding the whole document?	It leads to much smaller prompts, which reduces processing time and cost.
What is a primary technical challenge introduced by RAG?	It requires a complex pre-processing step (chunk creation) and a search mechanism.
What two main steps are involved in the RAG process?	Chunking the document and retrieving relevant chunks based on the user's question.
What is a risk when using RAG to find relevant chunks?	The retrieved chunks may not contain all the necessary context for the model to answer the question fully.
What must be decided when implementing RAG regarding text division?	How to define and split the text (e.g., equal portions or based on headers).
Compared to the naive approach, what is the overall nature of RAG implementation?	It is significantly more complex and requires more technical work.