Research on the Paths and Challenges of Large Language Models in Promoting Educational Equity
DOI: https://doi.org/10.62381/ACS.SSFS2025.01
Author(s)
Yuming Sun*
Affiliation(s)
Experimental School Affiliated to Chinese Academy of Sciences, Beijing, China
*Corresponding author.
Abstract
Large Language Models (LLMs), as a significant breakthrough in the field of artificial intelligence, are reshaping the allocation of educational resources. This paper explores how LLMs can promote educational equity through personalized learning support, multilingual education coverage, and low-cost resource provision, from the dual perspectives of technology empowerment and ethical risks. It also analyzes the challenges they face, such as the widening digital divide, embedded algorithmic bias, and inadequate cultural adaptability. The study proposes a three-dimensional governance framework comprising policy regulation, technological optimization, and social collaboration, providing a theoretical reference for the application of LLMs in promoting educational equity.
Keywords
Allocation of Educational Resources; Multilingual Education; Policy Regulation; Social Collaboration
References
[1] UNESCO. (2022). Global Education Monitoring Report: Technology in education—A tool on whose terms. Paris: UNESCO.
[2] Crystal, D. (2018). English as a global language. Cambridge University Press.
[3] Rawls, J. (1971). A theory of justice. Harvard University Press.
[4] Sen, A. (1999). Development as freedom. Anchor.
[5] OpenAI. (2023). GPT-4 technical report. OpenAI.
[6] Bender, E. M., Gebru, T., McMillan-Major, A., & Shmitchell, S. (2021). On the dangers of stochastic parrots: Can language models be too big? In Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency (pp. 610-623). ACM.
[7] Cotton, C. J., Robson, C., & McAndrew, P. (2023). An exploratory investigation into the use of ChatGPT in higher education assignments. Computers & Education, 182, 100516.
[8] EU. (2023). Proposal for a Regulation of the European Parliament and of the Council on Artificial Intelligence, and amending certain Union legal acts (Artificial Intelligence Act).
[9] MOE Singapore. (2022). AI-Enhanced Teaching Competency Framework. Ministry of Education White Paper.