Retrieval-Augmented Generation

An AI technique that enhances the responses of large language models (LLMs) by first retrieving relevant information from an external knowledge base or document collection, then providing that information as context for the model to generate its response. In accessibility applications, RAG enables AI assistants to answer questions grounded in specific content—such as a lecture transcript or course material—rather than relying solely on the model's pre-trained knowledge, which reduces hallucinations and improves factual accuracy. For example, a lecture accessibility tool might use RAG to allow students to ask questions about specific lecture content by searching the transcript for relevant passages and generating contextual answers.