A technique where a pre-trained model is further trained on a specific dataset or task to adapt its capabilities. Fine-tuning leverages existing learned representations while specializing the model for particular domains, improving performance while requiring less computational resources than training from scratch.
