Fine-Tuning

Fine-tuning is further training of a pre-trained model on a smaller, specialised dataset to adapt it to a particular task, domain or style.

What is Fine-Tuning?

Organisations fine-tune models on legal, medical or customer-service data, for example. Under the EU AI Act, significant fine-tuning of a general-purpose model can make the modifier a provider with its own obligations.

How it works

Training continues from the pre-trained weights using a lower learning rate and task-specific data.

Related terms

• Pre-Training
• Transfer Learning
• Parameter-Efficient Fine-Tuning (PEFT)
• Reinforcement Learning from Human Feedback (RLHF)
• Substantial Modification under the EU AI Act

← All Artificial Intelligence & Superintelligence definitions · ← Back to the A–Z definitions