AI Basics Part 3: Putting LLMs to good use
How can you put LLM's to good use? In this third of four presentations about Artificial Intelligence and how it can be used in practice, Pere Sivecas, Spend Matters Development PM and Software Architect, goes over the ways a company may modify a large language model (LLM) to its specific use case. After explaining why a company may want to modify the model, he details the advantages and disadvantages of fine tuning, low rank adaptation, retrieval-augmented generation (RAG) and agents. In the Q&A that follows, the team discusses the degree to which a generative AI model relies on an actual dataset and the presence of risk in a model failing to handle incorrect or contradictory data.