AI
Fine-Tuning
Fine-tuning continues training a pretrained model on a curated dataset so it adapts to a specific task, format or domain, updating model weights rather than input context.
Parameter-efficient methods such as LoRA update a small adapter rather than the full network, cutting cost and storage dramatically while retaining most of the benefit.
Fine-tuning teaches behaviour and format reliably; it is a poor mechanism for teaching facts, which retrieval handles better and updates more cheaply.