Fine-tuning
Further training of an already trained model on a smaller, focused dataset to adapt it to a task or style.
Also: finetuning
Fine-tuning starts from existing weights and continues training on new examples, such as medical notes, legal contracts, a company's support chats or text in a particular language. Because the model already knows a great deal, a few thousand to a few million examples can change its behaviour noticeably. Full fine-tuning updates every parameter; parameter-efficient methods such as LoRA update a small add-on. Fine-tuning can also weaken a model's safety training, a concern raised for open-weight releases.