A company wants to enhance multiple ML models. Select the appropriate technique for each use case. Each technique can be selected once or not at all. <Dropdown blanks={[{"id":"external_sources","options":["Few-shot learning","Fine-tuning","Retrieval Augmented Generation (RAG)","Zero-shot learning"],"answer":"Retrieval Augmented Generation (RAG)","label":"Enhancing the capabilities of a large language model (LLM) by using external sources"},{"id":"unseen_tasks","options":["Few-shot learning","Fine-tuning","Retrieval Augmented Generation (RAG)","Zero-shot learning"],"answer":"Zero-shot learning","label":"Querying a model to generalize and make predictions on unseen tasks"},{"id":"limited_data","options":["Few-shot learning","Fine-tuning","Retrieval Augmented Generation (RAG)","Zero-shot learning"],"answer":"Few-shot learning","label":"Querying a model with a limited amount of data for new tasks"}]} explanation={"RAG retrieves relevant external data and supplies it as context to an LLM at inference time. Zero-shot learning performs a task without task-specific examples, whereas few-shot learning relies on a small number of examples. Fine-tuning instead updates a pretrained model using additional training data.\n\n**Learn more:** [Augment large language models with retrieval-augmented generation or fine-tuning](https://learn.microsoft.com/en-us/azure/developer/ai/augment-llm-rag-fine-tuning)"} />