adaptation

Definition

Adjusting an already-trained model to a new task, domain, or user. It is the economic foundation of the foundation-model era: adaptation costs orders of magnitude less than training from scratch, so one base model branches into many specialized ones. The term is broader than fine-tuning, covering full-parameter updates as well as parameter-efficient methods like LoRA, adapters, and prompt tuning.

Explainers using this term

• Paper Explained: Retrieval-Augmented Skill Optimization via Cross-Harness Adaptation — Borrow Someone Else's Skill, Rewrite It for Your Own HarnessRetrieval-Augmented Skill Optimization via Cross-Harness Adaptation
• QLoRA — Fine-Tuning a 65B Model on One GPULoRA: Low-Rank Adaptation of Large Language Models
• LoRA from Scratch — Fine-Tuning a Giant by Training Two Thin MatricesLoRA: Low-Rank Adaptation of Large Language Models