Sergey Sosnovsky s.a.sosnovsky@uu.nl 2 Peter Brusilovsky peterb@pitt.edu 1 Andrew S. Lan andrewlan@cs.umass.edu 0 University of Massachutsetts Amherst , 140 Governors Dr., Amherst, MA 01003 , USA University of Pittsburgh , 135 North Bellefield Ave., Pittsburgh, PA. 15260 , USA Utrecht University , Princetonplein 5, Utrecht 3584 CC , the Netherlands

This year, the workshop on Intelligent Textbooks is organized for the fourth time. It builds on the success of the three previous workshops conducted as parts of the satellite programs of the International Conferences on Artificial Intelligence in Education in 20191, 20202 and 20213. At the first workshop in 2019, most accepted submissions have focused on various aspects of making textbooks adaptive through navigation, recommendation, or problem-solving support. Other popular topics were integration of interactive content, orchestration of learning around digital textbooks and automated analysis of the textbook content for various purposes. In 2020, adaptivity and interactivity remained important aspects of intelligent textbooks. However, the trend on leveraging machine learning, natural language processing and semantic technologies to automate processing or construction of textbook content became much more prevalent. Several papers and demos have presented approaches for textbook generation, transformation, linking to external content and extraction of knowledge from textbooks. The third workshop explored a variety of topics. Two new trends that separated it from its predecessors were: demonstration-based papers presenting prototypes of domain-oriented textbook applications and projects exploring automated approaches to extract from textbooks different kinds of learning objects. This year, the accepted contributions have been broadly categorized into the two main groups. Group 1 focuses on the usage of the textbooks' content to generate various forms of Assessment and Interactive material. Group 2 covers a broad set of topics that explore textbooks enrichment with external content and the use of such enriched textbooks in real educational contexts. Overall, the workshop has received 13 submissions (7 full papers, 4 short papers, and 2 demo papers). 9 submissions have been accepted (5 full papers, 3 short papers, 1 http://ml4ed.cc/2019-AIED-workshop/ 2 https://intextbooks.science.uu.nl/workshop2020/ 3 https://intextbooks.science.uu.nl/workshop2021/

 and 1 demo paper); 1 full paper submission has been downgraded to a short paper . The workshop program consists of two sessions that cover its two main themes: 1. Assessment and Interactivity Generation - Introduction - Full paper: Reading Comprehension Quiz Generation using Generative Pre-trained Transformers . Ramon Dijkstra, Zülküf Genç, Subhradeep Kayal, and Jaap Kamps. - Short paper: YAI4Edu: an Explanatory AI to Generate Interactive eBooks for Education . Francesco Sovrano , Kevin Ashley, Peter Brusilovsky , Fabio Vitali. 2. Enriched Textbooks : from Linking to Learning - Full paper: Towards a Pedagogical Framework for Designing and Developing iTextbooks . Chaohua Ou , Ashok Goel, David Joyner. The workshop takes place in Durham, UK and online on 27 July , 2022 . The workshop website (https://intextbooks.science.uu.nl/workshop2022/) provides iTextbook'2022 that helped prepare the workshop program: - Jordan Barria-Pineda , University of Pittsburgh; - Paulo Carvalho , Carnegie Mellon University; - Vinay Chaudhri ; - Paul Denny , The University of Auckland; - Brendan Flanagan , Kyoto University; - Reva Freedman , Northern Illinois University; - Debshila Mallick , OpenStax, Rice University;