[GSoC] Feedback Flow for Chatbot Answers

[danylo-boiko]

Feature description

It is crucial to evaluate the model's performance when working with Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG). The project aims to implement a feedback flow for chatbot answers, allowing users to provide feedback using "Good" and "Bad" options. These responses will be collected and analyzed for continuous evaluation.

Expected outcomes

Users can provide feedback for each chatbot answer using “Good” and “Bad” options (the system must store responses in the database).

[wonderbee77]

A reinforcement learning-based solution to provide multiple answers would be great.