A Comparative Study of the Readability of English Reading Materials Generated by Different Large Language Models

DOI

10.2991/978-2-38476-601-7_45How to use a DOI?

Keywords

Large language models; English reading instruction; Readability; AI-assisted teaching

Abstract

Large language models are widely used in English language teaching, yet the yet differences in readability among reading materials generated by different large language models remain unclear. This study compared English reading passages generated by three mainstream models—DeepSeek, ChatGPT, and Gemini—across two genres: expository and narrative. Readability was assessed using Flesch Reading Ease and Flesch-Kincaid Grade Level, two standardized metrics widely adopted in educational research. The results revealed systematic differences in the readability of texts generated by the three models. DeepSeek consistently produced the more readable texts, Gemini produced the least readable, and ChatGPT ranked in between. Across all models, narrative passages were found to be significantly more readable than expository ones, indicating that text genre plays a crucial role in determining difficulty levels for language learners. These findings address a critical gap in model selection criteria for AI-assisted reading instruction. They provide empirical guidance for teachers and instructional designers in choosing appropriate AI tools based on target genre and desired difficulty level, contributing to more effective and informed integration of large language models into English language classrooms.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Xinghan Li
AU  - Shu Zhang
PY  - 2026
DA  - 2026/08/06
TI  - A Comparative Study of the Readability of English Reading Materials Generated by Different Large Language Models
BT  - Proceedings of the 2026 2nd International Conference on Education Reform, Ideology and Politics (ERIP 2026)
PB  - Atlantis Press
SP  - 375
EP  - 381
SN  - 2352-5398
UR  - https://doi.org/10.2991/978-2-38476-601-7_45
DO  - 10.2991/978-2-38476-601-7_45
ID  - Li2026
ER  -