Glossary term
Glossary term
Agentic Systems
A prompt having multiple correct responses. For example, the following prompt has no one right answer:
Tell me a funny joke about elephants.
Evaluating the responses to no one right answer prompts is usually far more subjective than evaluating prompts with one right answer. For example, evaluating an elephant joke requires a systematic way to determine how funny the joke is.
Created for this library
A creative writing assistant team labels its prompts as no-one-right-answer when evaluating subjective generation tasks like marketing copy.
An LLM evaluation team uses a no-one-right-answer prompt set to evaluate open-ended generation with several acceptable outputs.
A research team uses no-one-right-answer evaluation when scoring brainstorming outputs since several distinct responses can be equally good.
Definition source: Google for Developers Machine Learning Glossary | Creative Commons Attribution 4.0 License