Write a PREreview

Automated story writing has been a subject of study for over 60 years. While there are mechanisms capable of generating consistent and coherent stories, and even those that employ narratological theories or demonstrate a strong command of language, little attention has been given to evaluating these texts in terms of their literary value, particularly from an aesthetic perspective. In this paper, we address the challenge of evaluating literary microfictions and argue that this task requires consideration of literary criteria across various aspects of the text. To facilitate this, we present an evaluation protocol grounded in literary theory, specifically drawing from the communication approach, to offer an objective framework for assessing generated texts. Furthermore, we report the results of our validation of the evaluation protocol (GrAImes), as answered by literary professionals. This protocol will serve as a foundation for evaluating automatically generated microfictions and assessing their literary value.