RAG-based chatbot
With Heisenware, you can create apps that allow app users to interact with previously provided knowledge — e.g., PDF documents or specific websites — via chat. For this purpose, RAG is used.
What is RAG?
RAG stands for Retrieval Augmented Generation. It is an architectural approach that enhances large language models (LLMs) by incorporating relevant external data into their responses.
Key Components
Retrieval: RAG retrieves pertinent information from external data sources based on the user's query.
Augmentation: The retrieved data is used to augment or supplement the LLM's knowledge.
Generation: The LLM then generates a response using both its pre-trained knowledge and the retrieved information.
RAG effectively bridges the gap between an LLM's broad knowledge and the need for precise, contextual, and current information, making it valuable for applications like question-answering systems and chatbots.
Getting started
To build a RAG-based chatbot as part of an app, there are two things required:
Knowledge base: The documents and other knowledge your chatbot should be able to make statements about and give answers to.
Chat instance: The connection to an LLM (by OpenAI) that you like to use to generate response.
This is currently a beta feature.
Please get in touch to make it work. We'll support you with this.
Last updated