Prompt University

More Than Completion

This article is intended to correct some very common misrepresentaions of LLMs and virtual assistants.

When you read a book or course about Prompt Engineering that says that LLMs simnply do prompt completion, you might think that they do more. And the ones you likely encounter likely do.

Most books and courses teach that LLMs just complete prompts by guessing what is likely to follow a prompt provided to it but that is an oversimplification. Base LLMs do predict sequences statistically, yet the LLMs in virtual assistants you interact with daily are Instruction-tuned LLMs that go far beyond that fundamental capability. Virtual assistants add much more to what base LLMs do. A major feature of virtual assistants is determination of intent. They also decompose complex requests into manageable parts. Responses are then composed to be appropriate for the prompt, not just a dump of material found elsewhere.