• An approach in which a large language model (LLM) is asked to perform a task without being given any specific examples, relying solely on its pre-trained knowledge and the task description provided in the prompt. This technique tests the model’s ability to understand and execute tasks based purely on natural language instructions.

1. Related terms: One-shot prompting · Few-shot prompting · Prompt engineering · Large language model
2. Previous term: Zero-indexed
3. Next term: Zeroconf
4. Random term: Function as a service (webglossary.info/random 🎲)