Examples is another name less frequently used for few-shot learning. It is known by the following equivalent terms: Few-shot tuning, Few-shot learning, Instruction fine tuning or In-context learning.
Examples help the model understand what an appropriate model response looks like. You can write your own example input and output or use the Test section to save a real response as an example. You can also add a prefix which will be appended to every example (for instance, “question” and “answer”).
Using demonstrations to show how to perform a task is often called “few-shot learning.”
One strategy, known as instruction fine tuning, is particularly good at improving a model’s performance on a variety of tasks. Instruction fine tuning trains the model using examples that demonstrate how it should respond to a specific instruction.
See Also: Incontext Learning