At a time when everyone is talking Generative AI, it's easy to get lost in the acronyms and technical detail. We're here to help remove the confusion with these definitions of key Generative AI concepts.
FM: Foundation Models
Foundation Models are a form of artificial intelligence that is pre-trained on vast data in various domains, allowing them to develop a wide range of capabilities. These models are not limited to linguistic tasks, but can also include image recognition, sound processing and code generation.
LLM: Large Language Models
Large Language Models are a subset of foundation models specifically designed to process and generate human language. They are trained on large textual data sets and can perform tasks such as translating, summarising content and answering questions.
RAG: Retrieval Augmented Generation
Retrieval Augmented Generation is the process of optimising the output of a Large Language Model. It's about feeding the model with a reliable knowledge base, such as product and customer data and information, to train it before generating a response. Large Language Models are trained on large volumes of data and use billions of parameters to generate original results for tasks such as answering questions, translating languages and completing sentences.
The Retrieval Augmented Generation extends the already powerful capabilities of Large Language Models to specific domains or and organisation's internal knowledge base, all without the need to retrain the model.
Simply put, without RAG the LLM creates a response based on the smaller amount of information it has been trained on, or what it already knows. These responses can sometimes not be very on point to begin with and require more time to give better responses.
With RAG, the LLM uses its already integrated knowledge base to create better and accurate answers.
Would you like to find out more or speak to an expert at User JournAI? Contact us!