I’m implementing a Retrieval-Augmented Generation (RAG) system for a chatbot that handles a variety of user queries. While RAG is primarily designed to provide informative responses by retrieving relevant documents and generating responses using a language model (LLM), I want to handle certain types of queries, like greetings and polite exchanges, without resorting to the LLM for efficiency and control.

Are there any tips or best practices to achieve this? I have tried few-shot prompting
are there any open source models which can perform this?
I have tried few-shot prompting it was not giving best results. Thank you in advance

i’m trying to understand first if this is possible or not and if so then how can i begin to work on this, so what’s i’m trying to do is there is one game that gives us 2 numbers every time and we select one of those and it then tells us if it’s right answer or not. so i have lots of it’s old data of which number was true 0 or 1 in last 2k some turns.