Benny G. Johnson, Jeffrey S. Dittel, Oscar J. Ortiz, Rodrigo Bistolfi, Michelle W. Clark, Bill Jerome, Richard Benton, Rachel Van Campenhout. LLM feedback isn't automatically better: Static scaffolds outperform dynamic feedback in textbook-embedded practice. In Sergey A. Sosnovsky, Peter Brusilovsky, Andrew S. Lan, Isaac Alpizar Chacon, editors, Proceedings of the Seventh International Workshop on Intelligent Textbooks 2026 co-located with the 27th International Conference on Artificial Intelligence in Education (AIED 2026), Seoul, Republic of Korea, June 28, 2026. Volume 4231 of CEUR Workshop Proceedings, pages 4-18, CEUR-WS.org, 2026. [doi]
Abstract is missing.