Fine-Tuning
What is Fine-Tuning?
Fine-tuning is the process of further training an existing AI model on a smaller, task-specific set of examples so that it performs better at a particular job, style or domain.
Fine-tuning changes the model itself, whereas prompting and retrieval-augmented generation change only what the model is given at the moment it answers. It is useful for teaching a model a consistent format or tone, a specialist vocabulary or a narrow, repeated task such as classifying contact reasons, and it is one way small language models are made good at specific contact-center work.
Fine-tuning is not the best way to keep an AI up to date with facts that change, such as prices or policies; retrieval handles that better. It also raises data questions: which examples were used, whether they contained customer data, and who owns the resulting model. Ask any AI vendor whether your data is used to train or tune its models.
Related: Small language model (SLM), Retrieval-augmented generation (RAG), Machine learning.