Classroom and lab-based research have shown the advantages of exposing students to a variety of problems with formatdifferences between them, compared to giving students problem sets with a single problem format. The rapid developmentof technologies such as intelligent tutoring systems (ITS) in education affords the opportunity to automatically generateand adapt problem content for practice and assessment purposes. In this paper, we investigate whether this approach canbe effectively deployed to an ITS, conducting a randomized controlled trial to compare students who practiced problemsbased on a single template and those who were exposed to problems based on multiple templates, both in the same ITS.Results show no statistically significant difference in the two conditions on students post-test performance and hint requestbehavior. However, students who saw multiple templates spent more time to answer the practice items compared tostudents who solved problems of a single structure.