03 · How Models Learn
Fine-tuning
Adapting a trained model
Additional training on examples chosen for a narrower task, style, domain, or behavior. Fine-tuning changes model weights; prompting and retrieval change the inputs instead. It can improve consistency but does not automatically add fresh factual knowledge.
Concrete example
Fine-tune a model on labeled support replies to match a house style, while still using retrieval for current product facts.