Represent tokens, words, documents, or images as high-dimensional vectors and store them in a vector database so an LLM can efficiently traverse and search curated data.
