Can Large Language Models Provide Feedback to Students? A Case Study on ChatGPT
Open accessAbstract
Educational feedback has been widely acknowledged as an effective approach to improving student learning. However, scaling effective practices can be laborious and costly, which motivated researchers to work on automated feedback systems (AFS). Inspired by the recent advancements in the pre-trained language models (e.g., ChatGPT), we posit that such models might advance the existing knowledge of textual feedback generation in AFS because of their capability to offer natural-sounding and detailed responses. Therefore, we aimed to investigate the feasibility of using ChatGPT to provide students with feedback to help them learn better. Specifically, we first examined the readability of ChatGPT-generated feedback. Then, we measured the agreement between ChatGPT and the instructor when assessing students' assignments according to the marking rubric. Finally, we used a well-known theoretical feedback framework to further investigate the effectiveness of the feedback generated by ChatGPT. Our results show that i) ChatGPT is capable of generating more detailed feedback that fluently and coherently summarizes students' performance than human instructors; ii) ChatGPT achieved high agreement with the instructor when assessing the topic of students' assignments; and iii) ChatGPT could provide feedback on the process of students completing the task, which benefits students developing learning skills.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The instant grading provided on assessments ensures that students receive timely and constructive feedback on their performance.
“In addition, the instant grading provided on assessments ensures that students receive timely and constructive feedback on their performance [44,50].”
From New Era of Artificial Intelligence in Education: Towards a Sustainable Multifaceted Revolution (2023) · cites this paper as [44]
Self-correction was made easy.
“Fifth, self-correction was made easy (Dai et al., 2023)”
From Impact of ChatGPT on ESL students’ academic writing skills: a mixed methods intervention study (2024) · cites this paper as Dai et al., 2023
The promotion of learner autonomy and peer collaboration are similar to findings in the AWE literature.
“On the other hand, the promotion of learner autonomy and peer collaboration are similar to findings in the AWE literature (Dai et al., 2023;Rudolph et al., 2023).”
From Impact of ChatGPT on ESL students’ academic writing skills: a mixed methods intervention study (2024) · cites this paper as Dai et al., 2023
Cites
Rethinking models of feedback for learning: the challenge of design
Boud, D., & Molloy, E. (2013) · Assessment & Evaluation in Higher Education · doi:10.1080/02602938.2012.691462 · Cited by 1,636 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
A Survey on Evaluation of Large Language Models
Chang, Y., Wang, X., Wang, J., Wu, Y., Yang, L., Zhu, K., Chen, H., Yi, X., Wang, C., Wang, Y., et al. (2024) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3641289 · Cited by 2,842 · see citing works
New Era of Artificial Intelligence in Education: Towards a Sustainable Multifaceted Revolution
Kamalov, F., Santandreu Calonge, D., & Gurrib, I. (2023) · Sustainability · doi:10.3390/su151612451 · Cited by 1,089 · see citing works
AI-generated feedback on writing: insights into efficacy and ENL student preference
Escalante, J., Pack, A., & Barrett, A. (2023) · International Journal of Educational Technology in Higher Education · doi:10.1186/s41239-023-00425-2 · Cited by 589 · see citing works
Promises and challenges of generative artificial intelligence for human learning
Yan, L., Greiff, S., Teuber, Z., & Gašević, D. (2024) · Nature Human Behaviour · doi:10.1038/s41562-024-02004-5 · Cited by 499 · see citing works
Self-citationJin, Y., Yan, L., Echeverria, V., Gašević, D., & Martinez-Maldonado, R. (2025) · Computers and Education Artificial Intelligence · doi:10.1016/j.caeai.2024.100348 · Cited by 452 · see citing works
Impact of ChatGPT on ESL students’ academic writing skills: a mixed methods intervention study
Mahapatra, S. (2024) · Smart Learning Environments · doi:10.1186/s40561-024-00295-9 · Cited by 398 · see citing works
Chang, D. H., Lin, M. P.-C., Hajian, S., & Wang, Q. Q. (2023) · Sustainability · doi:10.3390/su151712921 · Cited by 370 · see citing works
Luo (Jess), J. (2024) · Assessment & Evaluation in Higher Education · doi:10.1080/02602938.2024.2309963 · Cited by 366 · see citing works
Meyer, J., Jansen, T., Schiller, R., Liebenow, L. W., Steinbach, M., Horbach, A., & Fleckenstein, J. (2024) · Computers and Education: Artificial Intelligence · doi:10.1016/j.caeai.2023.100199 · Cited by 311 · see citing works
Feedback sources in essay writing: peer-generated or AI-generated feedback?
Banihashem, S. K., Kerman, N. T., Noroozi, O., Moon, J., & Drachsler, H. (2024) · International Journal of Educational Technology in Higher Education · doi:10.1186/s41239-024-00455-4 · Cited by 282 · see citing works
Mapping out a research agenda for generative artificial intelligence in tertiary education
Lodge, J. M., Thompson, K., & Corrin, L. (2023) · Australasian Journal of Educational Technology · doi:10.14742/ajet.8695 · Cited by 209 · see citing works
The grass is not always greener: Teacher vs. GPT-assisted written corrective feedback
Lin, S., & Crosthwaite, P. (2024) · System · doi:10.1016/j.system.2024.103529 · Cited by 86 · see citing works
Bewersdorff, A., Seßler, K., Baur, A., Kasneci, E., & Nerdel, C. (2023) · Computers and Education: Artificial Intelligence · doi:10.1016/j.caeai.2023.100177 · Cited by 76 · see citing works
Yilmaz, M., Temur, H. B., Emmungil, L., Çelik, E., Gauthier, A., & Cukurova, M. (2026) · International Journal of Educational Technology in Higher Education · doi:10.1186/s41239-026-00592-y · Cited by 10 · see citing works
Register alignment of ChatGPT-generated academic texts
Demir, N. Y., & Egbert, J. (2026) · Applied Corpus Linguistics · doi:10.1016/j.acorp.2025.100174 · Cited by 3 · see citing works
- Authors
- Dai, W., Lin, J., Jin, F., Li, T., Tsai, Y.-S., Gasevic, D., & Chen, G.
- Affiliations
- Monash University (AU)
- Venue
- IEEE International Conference on Advanced Learning Technologies (ICALT)
- Publisher
- Center for Open Science
- Published
- 2023-04-13
- Pages
- 323-325
- OpenAlex
- W4387162877
- Cited by
- 395
- References
- 1
- Citations by year
- 2026: 97 · 2025: 146 · 2024: 122 · 2023: 30
- FWCI
- 69.93
- Type
- conference paper
- Language
- en
- Open access
- green
- doi.org
- Licence
- Locations
- IEEE International Conference on Advanced Learning Technologies (ICALT) (publishedVersion)
- HAL (submittedVersion)
- The HKU Scholars Hub (University of Hong Kong) (submittedVersion)
- EdArXiv (OSF Preprints) (submittedVersion, cc-by-nc-nd)
- EdArXiv (OSF Preprints) (submittedVersion)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Intelligent Tutoring Systems and Adaptive Learning
- Physical Sciences › Computer Science › Information Systems › Artificial Intelligence in Education
- Social Sciences › Education › Student Assessment and Feedback
- Keywords
- automated feedback systems, ChatGPT, large language models, instructional feedback, student assignments, study skills, natural language generation
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
