Fine-tuning takes a foundation model trained on broad data and specializes it for particular use cases. Methods include: full fine-tuning (updating all parameters, expensive but thorough), LoRA/QLoRA (Low-Rank Adaptation - training small adapter layers, efficient and popular), prompt tuning (learning soft prompts), and RLHF (Reinforcement Learning from Human Feedback, used to align models with human preferences). Fine-tuning requires less data and compute than training from scratch while achieving task-specific performance. Common applications include domain adaptation (legal, medical), instruction following, and custom assistant behavior.
🧠 AI & LLMs intermediate
Fine-tuning
Adapting a pre-trained model to specific tasks or domains by training on specialized data.
53
views
</> Related Terms
Transformer
Neural network architecture using self-attention mechanisms, the foundation of modern LLMs like GPT and Claude.
Prompt Engineering
The practice of designing and optimizing inputs to AI models to achieve desired outputs.
LLM (Large Language Model)
AI models trained on massive text datasets to understand and generate human-like text.