Task Specification by Memetic Proxy

[…]

This simple example is meant to frame a question central to the Motivation of Prompt Programming: what Prompt will result in the intended Behavior and only the intended behavior?

REYNOLDS, Laria and MCDONELL, Kyle, 2021. Prompt Programming for Large Language Models: Beyond the Few-Shot Paradigm. In: Extended Abstracts of the 2021 CHI Conference on Human Factors in Computing Systems. New York, NY, USA: Association for Computing Machinery. 2021. p. 1–7. [Accessed 29 January 2023]. CHI EA ’21. ISBN 978-1-4503-8095-9. DOI 10.1145/3411763.3451760.

This statement suggests that the choice of prompt in programming can impact the likelihood of achieving the desired outcome. When the prompt consists of many instances of a function, it is more likely that the continuation will also be an instance of that function. However, with a small number of examples, it becomes more plausible for the continuation to deviate from the pattern. To ensure the desired outcome, it is important to carefully choose the prompt to balance the number of examples with the intended behavior, while minimizing the Risk of unintended behavior.

⇒ What is a Memetic Proxy? ⇒ Establish Rules ⇒ Real Constraint ⇒ Constraint ⇒ Patent