By Rosana Montes, Natalia Díaz-Rodríguez, Carlos Peláez Gonzalez y Carlos Núñez Molina
Large Language Models are capable of performing various language-related tasks, such as completing sentences, translating from one language to another, summarizing texts and even writing original texts on their own.
When these models are given an initial text as seed, they are able to generate several texts that are consistent with the initial text. Because these models are not perfect, it is sometimes necessary to give them as initial text a different one even if it conveys the same idea in order to obtain a satisfactory result. The people who are in charge of analyzing the behavior of these models to find the best answers or limitations of the models are known as prompt engineers.
