AWS Certified AI Practitioner AIF-C01 Practice Question

Retrieval Augmented Generation (RAG) combines information retrieval with text generation. The assistant can embed and index the policy documents, retrieve the most relevant passages for each user question, and pass those passages to a foundation model to compose an answer. Parameter-efficient fine-tuning adapts a model to new tasks but does not automatically search external content at inference. Increasing temperature only changes randomness in token selection and does not enable retrieval. A batch transform job performs offline inference on static data rather than dynamic question-answering with document lookup.