Core Concepts
Zero-Shot
Prompting a model to perform a task without providing any examples.
Zero-shot prompting relies on the model's pre-trained knowledge to interpret and complete a task from instructions alone. For example: 'Classify the sentiment of this review: ...' with no examples. Zero-shot works well for models trained with strong instruction-following but may need few-shot examples for complex or domain-specific tasks.
Related Terms
Few-Shot
Providing a small number of input-output examples in the prompt to guide the model.
Prompt Engineering
The practice of designing inputs to elicit optimal outputs from language models.
Chain of Thought
A prompting technique that instructs the model to reason step-by-step before answering.
Instruction Tuning
Fine-tuning a model on examples of instructions paired with ideal responses.