Large Language Models as Evaluators in Education: Verification of Feedback Consistency and Accuracy
Source abstract
Abstract
The recent advancements in large language models (LLMs) have brought significant changes to the field of education, particularly in the generation and evaluation of feedback. LLMs are transforming education by streamlining tasks like content creation, feedback generation, and assessment, reducing teachers’ workload and improving online education efficiency. This study aimed to verify the consistency and reliability of LLMs as evaluators by conducting automated evaluations using various LLMs based on five educational evaluation criteria. The analysis revealed that while LLMs were capable of performing consistent evaluations under certain conditions, a lack of consistency was observed both among evaluators and across models for other criteria. Notably, low agreement among human evaluators correlated with reduced reliability in LLM evaluations. Furthermore, variations in evaluation results were influenced by factors such as prompt strategies and model architecture, highlighting the complexity of achieving reliable assessments using LLMs. These findings suggest that while LLMs have the potential to transform educational systems, careful selection and combination of models are essential to improve their consistency and align their performance with human evaluators in educational settings.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Topic Modeling · Natural Language Processing Techniques · Multimodal Machine Learning Applications
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Computer Science and Engineering, Chungnam National University, Daejeon 34134, Republic of Korea
Electronics and Telecommunications Research Institute, Daejeon 34129, Republic of Korea
Funding
Institute of Information & Communications Technology Planning & Evaluation (IITP) grant funded by the Korea government (MSIT) · National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) · Chungnam National University
References (41)
Show source-provided references
1. Gonçalves, J.A.d.C., Lima, J.L.S.d.M., Coelho, J.P., García-Peñalvo, F.J., and García-Holgado, A. (2024). Smart Learning Applications: Leveraging LLMs for Contextualized and Ethical Educational Technology. Proceedings of the TEEM 2023, Springer.10.1007/978-981-97-1814-6 ↗
2. Laato, S., Morschheuser, B., Hamari, J., and Björne, J. (2023, January 10–13). AI-Assisted Learning with ChatGPT and Large Language Models: Implications for Higher Education. Proceedings of the 2023 IEEE International Conference on Advanced Learning Technologies (ICALT), Orem, UT, USA.10.1109/ICALT58122.2023.00072 ↗
3. Azaiz, I., Kiesler, N., and Strickroth, S. (2024, January 8–10). Feedback-Generation for Programming Exercises with GPT-4. Proceedings of the 2024 on Innovation and Technology in Computer Science Education V. 1, Milan, Italy.10.1145/3649217.3653594 ↗
4. Comparing the quality of human and ChatGPT feedback of students’ writing10.1016/j.learninstruc.2024.101894 ↗
5. Education in the Era of Generative Artificial Intelligence (AI): Understanding the Potential Benefits of ChatGPT in Promoting Teaching and Learning10.61969/jai.1337500 ↗
6. Teachers’ Agency in the Era of LLM and Generative AI
7. Ku, L.W., Martins, A., and Srikumar, V. (2024, January 11–16). Ranking Large Language Models without Ground Truth. Proceedings of the Findings of the Association for Computational Linguistics ACL 2024, Bangkok, Thailand.
8. Chen, G.H., Chen, S., Liu, Z., Jiang, F., and Wang, B. (2024). Humans or llms as the judge? A study on judgement biases. arXiv.10.18653/v1/2024.emnlp-main.474 ↗
9. Gan, W., Qi, Z., Wu, J., and Lin, J.C.W. (2023, January 15–18). Large Language Models in Education: Vision and Opportunities. Proceedings of the 2023 IEEE International Conference on Big Data (BigData), Sorrento, Italy.10.1109/BigData59044.2023.10386291 ↗
10. Wang, S., Xu, T., Li, H., Zhang, C., Liang, J., Tang, J., Yu, P.S., and Wen, Q. (2024). Large language models for education: A survey and outlook. arXiv.
11. Performance of the pre-trained large language model GPT-4 on automated short answer grading10.1007/s44163-024-00147-y ↗
12. Kochmar, E., Bexte, M., Burstein, J., Horbach, A., Laarmann-Quante, R., Tack, A., Yaneva, V., and Yuan, Z. (2024, January 6). Exploring LLM Prompting Strategies for Joint Essay Scoring and Feedback Generation. Proceedings of the 19th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2024), Mexico City, Mexico.
13. Automatic question generation: A review of methodologies, datasets, evaluation metrics, and applications10.1007/s13748-023-00295-9 ↗
14. Duh, K., Gomez, H., and Bethard, S. (2024, January 16–21). Exploring Automated Distractor Generation for Math Multiple-choice Questions via Large Language Models. Proceedings of the Findings of the Association for Computational Linguistics: NAACL 2024, Mexico City, Mexico.
15. Lee, J., Smith, D., Woodhead, S., and Lan, A. (2024). Math Multiple Choice Question Generation via Human-Large Language Model Collaboration. arXiv.
16. Dan, Y., Lei, Z., Gu, Y., Li, Y., Yin, J., Lin, J., Ye, L., Tie, Z., Zhou, Y., and Wang, Y. (2023). Educhat: A large-scale language model-based chatbot system for intelligent education. arXiv.
17. Using LLMs to bring evidence-based feedback into the classroom: AI-generated feedback increases secondary students’ text revision, motivation, and positive emotions10.1016/j.caeai.2023.100199 ↗
18. Dai, W., Lin, J., Jin, H., Li, T., Tsai, Y.S., Gašević, D., and Chen, G. (2023, January 10–13). Can Large Language Models Provide Feedback to Students? A Case Study on ChatGPT. Proceedings of the 2023 IEEE International Conference on Advanced Learning Technologies (ICALT), Orem, UT, USA.10.1109/ICALT58122.2023.00100 ↗
19. Scott, D., Bel, N., and Zong, C. (2020, January 8–13). Personalized Multimodal Feedback Generation in Education. Proceedings of the 28th International Conference on Computational Linguistics, Barcelona, Spain.
20. Koutcheme, C., Dainese, N., Hellas, A., Sarsa, S., Leinonen, J., Ashraf, S., and Denny, P. (2024). Evaluating Language Models for Generating and Judging Programming Feedback. arXiv.
21. Olney, A.M., Chounta, I.A., Liu, Z., Santos, O.C., and Bittencourt, I.I. (2024). Improving the Validity of Automatically Generated Feedback via Reinforcement Learning. Artificial Intelligence in Education, Springer.
22. Guo, S., Latif, E., Zhou, Y., Huang, X., and Zhai, X. (2024). Using Generative AI and Multi-Agents to Provide Automatic Feedback. arXiv.
23. Lagakis, P., and Demetriadis, S. (2024, January 10–13). EvaAI: A Multi-agent Framework Leveraging Large Language Models for Enhanced Automated Grading. Proceedings of the International Conference on Intelligent Tutoring Systems, Thessaloniki, Greece.10.1007/978-3-031-63028-6_32 ↗
24. Wu, Q., Bansal, G., Zhang, J., Wu, Y., Zhang, S., Zhu, E., Li, B., Jiang, L., Zhang, X., and Wang, C. (2023). Autogen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation Framework. arXiv.
25. Zhang, Z., Zhang-Li, D., Yu, J., Gong, L., Zhou, J., Liu, Z., Hou, L., and Li, J. (2024). Simulating Classroom Education with LLM-Empowered Agents. arXiv.
26. Kochmar, E., Bexte, M., Burstein, J., Horbach, A., Laarmann-Quante, R., Tack, A., Yaneva, V., and Yuan, Z. (2024, January 20). Using Program Repair as a Proxy for Language Models’ Feedback Ability in Programming Education. Proceedings of the 19th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2024), Mexico City, Mexico.
27. Zheng, L., Chiang, W.L., Sheng, Y., Zhuang, S., Wu, Z., Zhuang, Y., Lin, Z., Li, Z., Li, D., and Xing, E. (2023, January 10–16). Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena. Proceedings of the Advances in Neural Information Processing Systems, New Orleans, LA, USA.
28. Rogers, A., Boyd-Graber, J., and Okazaki, N. (2023, January 9–14). Can Large Language Models Be an Alternative to Human Evaluations?. Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Toronto, ON, Canada.
29. Bouamor, H., Pino, J., and Bali, K. (2023, January 6–10). G-Eval: NLG Evaluation using Gpt-4 with Better Human Alignment. Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, Singapore.
30. Bavaresco, A., Bernardi, R., Bertolazzi, L., Elliott, D., Fernández, R., Gatt, A., Ghaleb, E., Giulianelli, M., Hanna, M., and Koller, A. (2024). LLMs instead of Human Judges? A Large Scale Empirical Study across 20 NLP Evaluation Tasks. arXiv.
31. Verga, P., Hofstatter, S., Althammer, S., Su, Y., Piktus, A., Arkhangorodsky, A., Xu, M., White, N., and Lewis, P. (2024). Replacing Judges with Juries: Evaluating LLM Generations with a Panel of Diverse Models. arXiv.
32. Thakur, A.S., Choudhary, K., Ramayapally, V.S., Vaidyanathan, S., and Hupkes, D. (2024). Judging the Judges: Evaluating Alignment and Vulnerabilities in LLMs-as-Judges. arXiv.
33. Jung, J., Brahman, F., and Choi, Y. (2024). Trust or Escalate: LLM Judges with Provable Guarantees for Human Agreement. arXiv.
34. Kenton, Z., Siegel, N.Y., Kramár, J., Brown-Cohen, J., Albanie, S., Bulian, J., Agarwal, R., Lindner, D., Tang, Y., and Goodman, N.D. (2024). On scalable oversight with weak LLMs judging strong LLMs. arXiv.
35. Yarowsky, D., Baldwin, T., Korhonen, A., Livescu, K., and Bethard, S. (2013, January 18–21). MCTest: A Challenge Dataset for the Open-Domain Machine Comprehension of Text. Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing, Seattle, WA, USA.
36. Jia, Q., Young, M., Xiao, Y., Cui, J., Liu, C., Rashid, P., and Gehringer, E. (2022, January 24–27). Insta-Reviewer: A Data-Driven Approach for Generating Instant Feedback on Students’ Project Reports. Proceedings of the International Educational Data Mining Society, Durham, UK.
37. Achiam, J., Adler, S., Agarwal, S., Ahmad, L., Akkaya, I., Aleman, F.L., Almeida, D., Altenschmidt, J., Altman, S., and Anadkat, S. (2023). Gpt-4 technical report. arXiv.
38. Touvron, H., Lavril, T., Izacard, G., Martinet, X., Lachaux, M.A., Lacroix, T., Rozière, B., Goyal, N., Hambro, E., and Azhar, F. (2023). LLaMA: Open and Efficient Foundation Language Models. arXiv.
39. Team, G., Mesnard, T., Hardin, C., Dadashi, R., Bhupatiraju, S., Pathak, S., Sifre, L., Rivière, M., Kale, M.S., and Love, J. (2024). Gemma: Open models based on gemini research and technology. arXiv.
40. Wolf, T., Debut, L., Sanh, V., Chaumond, J., Delangue, C., Moi, A., Cistac, P., Rault, T., Louf, R., and Funtowicz, M. (2019). HuggingFace’s Transformers: State-of-the-art natural language processing. arXiv.10.18653/v1/2020.emnlp-demos.6 ↗
41. Hu, E.J., Shen, Y., Wallis, P., Allen-Zhu, Z., Li, Y., Wang, S., Wang, L., and Chen, W. (2021). Lora: Low-rank adaptation of large language models. arXiv.