GlossaryAdapter
Adapter
In the context of AI, adapters are small neural network modules that are inserted into a pre-trained model. By training only the adapter and leaving the rest of the model's weights frozen, the model can be quickly and efficiently adapted to a new task. This is a form of parameter-efficient fine-tuning (PEFT).
Adapters are a cost-effective way to customize large language models for specific domains or tasks, as they require significantly less computational resources than full fine-tuning.