---
title: "Embracing Imperfection — paper by Wu et al. (2025)"
description: "The authors propose a training-free, three-stage pipeline: Cognitive prototype. 60 references, cited by 9 papers here."
authors: ["Tao Wu", "Jingyuan Chen", "Wang Lin", "Mengze Li", "Yumeng Zhu", "Ang Li", "Kun Kuang", "Fei Wu"]
year: 2025
venue: "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)"
source: "https://aclanthology.org/2025.acl-long.488/"
doi: "10.18653/v1/2025.acl-long.488"
licence: "CC BY 4.0 (ACL Anthology)"
licence_url: "https://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/wu-2025-embracing-imperfection-simulating-students-diverse/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/wu-2025-embracing-imperfection-simulating-students-diverse/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2025](https://md.research.elamur.ai/free/en/papers/2025/page/7/) / Embracing Imperfection: Simulating Students with Diverse Cognitive Levels Using LLM-based Agents

# Embracing Imperfection: Simulating Students with Diverse Cognitive Levels Using LLM-based Agents

**Authors:** [Tao Wu](https://md.research.elamur.ai/free/en/authors/tao-wu/), [Jingyuan Chen](https://md.research.elamur.ai/free/en/authors/jingyuan-chen/), [Wang Lin](https://md.research.elamur.ai/free/en/authors/wang-lin/), [Mengze Li](https://md.research.elamur.ai/free/en/authors/mengze-li/), Yumeng Zhu, [Ang Li](https://md.research.elamur.ai/free/en/authors/ang-li/), [Kun Kuang](https://md.research.elamur.ai/free/en/authors/kun-kuang/) and [Fei Wu](https://md.research.elamur.ai/free/en/authors/fei-qing-wu/) · **Year:** 2025 · **Venue:** [Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)](https://md.research.elamur.ai/free/en/venues/acl/) ([ACL 2025](https://md.research.elamur.ai/free/en/venues/acl/2025/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2025.acl-long.488/) · [PDF](https://aclanthology.org/2025.acl-long.488.pdf) · [DOI 10.18653/v1/2025.acl-long.488](https://doi.org/10.18653/v1/2025.acl-long.488) · [OpenAlex](https://openalex.org/W4412945501)

**Citations:** 4 citations (OpenAlex, 2026-10-14) · [9](https://md.research.elamur.ai/free/en/papers/wu-2025-embracing-imperfection-simulating-students-diverse/cited-by/) in our library, of the 149,439 papers whose bibliographies we read

## Summary
LLM agents are increasingly used to role-play learners, for example to test tutoring systems or teaching strategies. Because they are tuned to be helpful assistants, they tend to give answers that are too good for weaker students and miss the errors such students actually make. The authors propose a training-free, three-stage pipeline:

1. **Cognitive prototype.** A knowledge graph is built from a student's earlier work. It records concepts, how they relate, and how well the student has mastered each one.
2. **Behaviour prediction.** For a new task, the prototype is mapped onto the concepts the task involves. This predicts whether the student will succeed and which mistakes are likely. It replaces retrieval based on surface text similarity.
3. **Solution generation.** A beam-search self-refinement loop revises the generated solution until it matches the predicted behaviour, including the errors.

For evaluation they built Student_100, a Python programming dataset drawn from the PTA platform. The method beat the baselines in both LLM-based and human evaluation, with the abstract claiming roughly 100% gains in simulation accuracy and realism. This matters because realistic, imperfect simulated learners make AI-in-education experiments more credible, and the approach avoids fine-tuning models on erroneous knowledge.

## Key points
- **Dataset.** Student_100 covers 100 students and 5,000 annotated records. For every student, 40 records serve as prior learning history and 10 are held out for simulation, all completed within one week. Ten trained annotators added task descriptions and evaluations of each student's behaviour.
- **Contrast with fine-tuning.** Approaches like MalAlgoPy fine-tune on error-rich data, which risks degrading the model and ignores individual differences. This framework needs no training.
- **Human evaluation.** Ten undergraduates with solid Python skills rated solutions on a 1–5 scale, with two raters per solution. The method scored above the second-best baseline in every model setting.
- **Difficulty by student level.** Simulation quality correlates positively with student ability. Reproducing the plausible mistakes of weaker students is the harder problem.
- **Limitations.** Validation covers programming only, and the simulation uses text alone, with no multimodal signals.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

📚 Shelves (4): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 18,614 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 17,965 works · [🗄️ Information Systems](https://md.research.elamur.ai/free/en/shelves/computer-science/information-systems/) · 1,730 works · [🖥️ Computer Science Applications](https://md.research.elamur.ai/free/en/shelves/computer-science/computer-science-applications/) · 131 works

## Cite

Cite as: Tao Wu, Jingyuan Chen, Wang Lin, Mengze Li, Yumeng Zhu, Ang Li, Kun Kuang and Fei Wu. 2025. Embracing Imperfection: Simulating Students with Diverse Cognitive Levels Using LLM-based Agents. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers).

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{wu2025embracing,
  title = {{Embracing Imperfection: Simulating Students with Diverse Cognitive Levels Using LLM-based Agents}},
  author = {Tao Wu, Jingyuan Chen, Wang Lin, Mengze Li, Yumeng Zhu, Ang Li, Kun Kuang and Fei Wu},
  booktitle = {Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2025},
  doi = {10.18653/v1/2025.acl-long.488},
  url = {https://doi.org/10.18653/v1/2025.acl-long.488}
}
```

- [← previous Efficient Tool Use with Chain-of-Abstraction Reasoning](https://md.research.elamur.ai/free/en/papers/gao-2025-efficient-tool-use-chain-abstraction/)
- [next → Enhancing Arabic Automated Essay Scoring with Synthetic Data and Error Injection](https://md.research.elamur.ai/free/en/papers/qwaider-2025-enhancing-arabic-automated-essay-scoring/)
- [↑ back to papers of 2025](https://md.research.elamur.ai/free/en/papers/2025/page/7/)

## References

60 references, 8 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Gati V. Aher, Rosa I. Arriaga, and Adam Tauman Kalai. 2023. Using large language models to simulate multiple humans and replicate human subject studies. In International Conference on Machine Learning, ICML 2023, 23-29 July 2023, Honolulu, Hawaii, USA, volume 202 of Proceedings of Machine Learning R… [find](https://scholar.google.com/scholar?q=Gati+V%2E+Aher%2C+Rosa+I%2E+Arriaga%2C+and+Adam+Tauman+Kalai%2E+2023%2E+Using+large+language+models+to+simulate+multiple+humans+and+replicate+human+subject+studies%2E+In+International+Conference+on+Machine+Learning)
2. Anthropic. 2024. The claude 3 model family: Opus, sonnet, haiku. [find](https://scholar.google.com/scholar?q=Anthropic%2E+2024%2E+The+claude+3+model+family%3A+Opus%2C+sonnet%2C+haiku%2E)
3. Luca Benedetto, Giovanni Aradelli, Antonia Donvito, Alberto Lucchetti, Andrea Cappelli, and Paula Buttery. 2024. Using llms to simulate students’ responses to exam questions. In Findings of the Association for Computational Linguistics: EMNLP 2024, Miami, Florida, USA, November 12-16, 2024, pages 11… [find](https://scholar.google.com/scholar?q=Luca+Benedetto%2C+Giovanni+Aradelli%2C+Antonia+Donvito%2C+Alberto+Lucchetti%2C+Andrea+Cappelli%2C+and+Paula+Buttery%2E+2024%2E+Using+llms+to+simulate+students%E2%80%99+responses+to+exam+questions%2E+In+Findings+of+the+Associ)
4. Guoxin Chen, Minpeng Liao, Chengxi Li, and Kai Fan. 2024. [Alphamath almost zero: process supervision without process](https://md.research.elamur.ai/free/en/papers/chen-2024-alphamath-almost-zero-process-supervision/). CoRR, abs/2405.03553.
5. Nico Daheim, Jakub Macina, Manu Kapur, Iryna Gurevych, and Mrinmaya Sachan. 2024. Stepwise verification and remediation of student reasoning errors with large language model tutors. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024, Miami, FL, USA,… [find](https://scholar.google.com/scholar?q=Nico+Daheim%2C+Jakub+Macina%2C+Manu+Kapur%2C+Iryna+Gurevych%2C+and+Mrinmaya+Sachan%2E+2024%2E+Stepwise+verification+and+remediation+of+student+reasoning+errors+with+large+language+model+tutors%2E+In+Proceedings+of+)
6. Zhenlong Dai, Bingrui Chen, Zhuoluo Zhao, Xiu Tang, Sai Wu, Chang Yao, Zhipeng Gao, and Jingyuan Chen. 2025. Less is more: Adaptive program repair with bug localization and preference learning. In AAAI-25, Sponsored by the Association for the Advancement of Artificial Intelligence, February 25 -Marc… [find](https://scholar.google.com/scholar?q=Zhenlong+Dai%2C+Bingrui+Chen%2C+Zhuoluo+Zhao%2C+Xiu+Tang%2C+Sai+Wu%2C+Chang+Yao%2C+Zhipeng+Gao%2C+and+Jingyuan+Chen%2E+2025%2E+Less+is+more%3A+Adaptive+program+repair+with+bug+localization+and+preference+learning%2E+In+AAA)
7. Zhenlong Dai, Chang Yao, WenKang Han, Yuanying Yuanying, Zhipeng Gao, and Jingyuan Chen. 2024. Mpcoder: Multi-user personalized code generator with explicit and implicit style representation learning. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume… [find](https://scholar.google.com/scholar?q=Zhenlong+Dai%2C+Chang+Yao%2C+WenKang+Han%2C+Yuanying+Yuanying%2C+Zhipeng+Gao%2C+and+Jingyuan+Chen%2E+2024%2E+Mpcoder%3A+Multi%2Duser+personalized+code+generator+with+explicit+and+implicit+style+representation+learning%2E)
8. Jimmy De La Torre. 2009. Dina model and parameter estimation: A didactic. Journal of educational and behavioral statistics, 34(1):115–130. [find](https://scholar.google.com/scholar?q=Jimmy+De+La+Torre%2E+2009%2E+Dina+model+and+parameter+estimation%3A+A+didactic%2E+Journal+of+educational+and+behavioral+statistics%2C+34%281%29%3A115%E2%80%93130%2E)
9. Chin-Yew Lin. 2004. [Rouge: A package for automatic evaluation of summaries](https://md.research.elamur.ai/free/en/papers/lin-2004-rouge-package-automatic-evaluation-summaries/). In Text summarization branches out, pages 74–81.
10. Zhiang Dong, Jingyuan Chen, and Fei Wu. 2025a. Knowledge is power: Harnessing large language models for enhanced cognitive diagnosis. In AAAI-25, Sponsored by the Association for the Advancement of Artificial Intelligence, February 25 - March 4, 2025, Philadelphia, PA, USA, pages 164–172. AAAI Press… [find](https://scholar.google.com/scholar?q=Zhiang+Dong%2C+Jingyuan+Chen%2C+and+Fei+Wu%2E+2025a%2E+Knowledge+is+power%3A+Harnessing+large+language+models+for+enhanced+cognitive+diagnosis%2E+In+AAAI%2D25%2C+Sponsored+by+the+Association+for+the+Advancement+of+Ar)
11. Wang Lin, Yueying Feng, WenKang Han, Tao Jin, Zhou Zhao, Fei Wu, Chang Yao, and Jingyuan Chen. 2024. E3 : Exploring embodied emotion through A large-scale egocentric video dataset. In Advances in Neural Information Processing Systems 38: Annual Conference on Neural Information Processing Systems 202… [find](https://scholar.google.com/scholar?q=Wang+Lin%2C+Yueying+Feng%2C+WenKang+Han%2C+Tao+Jin%2C+Zhou+Zhao%2C+Fei+Wu%2C+Chang+Yao%2C+and+Jingyuan+Chen%2E+2024%2E+E3+%3A+Exploring+embodied+emotion+through+A+large%2Dscale+egocentric+video+dataset%2E+In+Advances+in+Neur)
12. Zhiang Dong, Liya Hu, Jingyuan Chen, Zhihua Wang, and Fei Wu. 2025b. Comprehend then predict: Prompting large language models for recommendation with semantic and collaborative data. ACM Transactions on Information Systems. [find](https://scholar.google.com/scholar?q=Zhiang+Dong%2C+Liya+Hu%2C+Jingyuan+Chen%2C+Zhihua+Wang%2C+and+Fei+Wu%2E+2025b%2E+Comprehend+then+predict%3A+Prompting+large+language+models+for+recommendation+with+semantic+and+collaborative+data%2E+ACM+Transactions+)
13. Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Amy Yang, Angela Fan, et al. 2024. [The llama 3 herd of models](https://md.research.elamur.ai/free/en/papers/grattafiori-2024-llama-3-herd-models/). CoRR, abs/2407.21783.
14. Wang Lin, Tao Jin, Wenwen Pan, Linjun Li, Xize Cheng, Ye Wang, and Zhou Zhao. 2023. TAVT: towards transferable audio-visual text generation. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), ACL 2023, Toronto, Canada, July 9-14, 2023,… [find](https://scholar.google.com/scholar?q=Wang+Lin%2C+Tao+Jin%2C+Wenwen+Pan%2C+Linjun+Li%2C+Xize+Cheng%2C+Ye+Wang%2C+and+Zhou+Zhao%2E+2023%2E+TAVT%3A+towards+transferable+audio%2Dvisual+text+generation%2E+In+Proceedings+of+the+61st+Annual+Meeting+of+the+Associatio)
15. Omar Erak, Omar Alhussein, Shimaa Naser, Nouf Al-abbasi, De Mi, and Sami Muhaidat. 2024. Large language model-driven curriculum design for mobile networks. CoRR, abs/2405.18039. — [arXiv:2405.18039](https://arxiv.org/abs/2405.18039) [find](https://scholar.google.com/scholar?q=Omar+Erak%2C+Omar+Alhussein%2C+Shimaa+Naser%2C+Nouf+Al%2Dabbasi%2C+De+Mi%2C+and+Sami+Muhaidat%2E+2024%2E+Large+language+model%2Ddriven+curriculum+design+for+mobile+networks%2E+CoRR%2C+abs%2F2405%2E18039%2E+%E2%80%94+%5BarXiv%3A2405%2E18039%5D%28h)
16. Jiayu Liu, Zhenya Huang, Tong Xiao, Jing Sha, Jinze Wu, Qi Liu, Shijin Wang, and Enhong Chen. 2024a. Socraticlm: Exploring socratic personalized teaching with large language models. In The Thirty-eighth Annual Conference on Neural Information Processing Systems. [find](https://scholar.google.com/scholar?q=Jiayu+Liu%2C+Zhenya+Huang%2C+Tong+Xiao%2C+Jing+Sha%2C+Jinze+Wu%2C+Qi+Liu%2C+Shijin+Wang%2C+and+Enhong+Chen%2E+2024a%2E+Socraticlm%3A+Exploring+socratic+personalized+teaching+with+large+language+models%2E+In+The+Thirty%2Deigh)
17. Joy He-Yueya, Noah D. Goodman, and Emma Brunskill. 2024. Evaluating and optimizing educational content with large language model judgments. In Proceedings of the 17th International Conference on Educational Data Mining, EDM 2024, Atlanta, Georgia, USA, July 14-17, 2024. International Educational Dat… [find](https://scholar.google.com/scholar?q=Joy+He%2DYueya%2C+Noah+D%2E+Goodman%2C+and+Emma+Brunskill%2E+2024%2E+Evaluating+and+optimizing+educational+content+with+large+language+model+judgments%2E+In+Proceedings+of+the+17th+International+Conference+on+Educa)
18. Zhengyuan Liu, Stella Xin Yin, Geyu Lin, and Nancy Chen. 2024b. Personality-aware student simulation for conversational intelligent tutoring systems. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024, Miami, FL, USA, November 12-16, 2024, pages 626… [find](https://scholar.google.com/scholar?q=Zhengyuan+Liu%2C+Stella+Xin+Yin%2C+Geyu+Lin%2C+and+Nancy+Chen%2E+2024b%2E+Personality%2Daware+student+simulation+for+conversational+intelligent+tutoring+systems%2E+In+Proceedings+of+the+2024+Conference+on+Empirical)
19. Liya Hu, Zhiang Dong, Jingyuan Chen, Guifeng Wang, Zhihua Wang, Zhou Zhao, and Fei Wu. 2023. Ptadisc: A cross-course dataset supporting personalized learning in cold-start scenarios. In Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2… [find](https://scholar.google.com/scholar?q=Liya+Hu%2C+Zhiang+Dong%2C+Jingyuan+Chen%2C+Guifeng+Wang%2C+Zhihua+Wang%2C+Zhou+Zhao%2C+and+Fei+Wu%2E+2023%2E+Ptadisc%3A+A+cross%2Dcourse+dataset+supporting+personalized+learning+in+cold%2Dstart+scenarios%2E+In+Advances+in+Ne)
20. Zitao Liu, Qiongqiong Liu, Jiahao Chen, Shuyan Huang, and Weiqi Luo. 2023. simplekt: A simple but tough-to-beat baseline for knowledge tracing. In The Eleventh International Conference on Learning Representations, ICLR 2023, Kigali, Rwanda, May 1-5, 2023. OpenReview.net. [find](https://scholar.google.com/scholar?q=Zitao+Liu%2C+Qiongqiong+Liu%2C+Jiahao+Chen%2C+Shuyan+Huang%2C+and+Weiqi+Luo%2E+2023%2E+simplekt%3A+A+simple+but+tough%2Dto%2Dbeat+baseline+for+knowledge+tracing%2E+In+The+Eleventh+International+Conference+on+Learning+Rep)
21. Zihan Huang, Tao Wu, Wang Lin, Shengyu Zhang, Jingyuan Chen, and Fei Wu. 2025. Autogeo: Automating geometric image dataset creation for enhanced geometry understanding. IEEE Transactions on Multimedia. [find](https://scholar.google.com/scholar?q=Zihan+Huang%2C+Tao+Wu%2C+Wang+Lin%2C+Shengyu+Zhang%2C+Jingyuan+Chen%2C+and+Fei+Wu%2E+2025%2E+Autogeo%3A+Automating+geometric+image+dataset+creation+for+enhanced+geometry+understanding%2E+IEEE+Transactions+on+Multimedia)
22. Bo Jiang, Xinya Li, Shuhao Yang, Yaqi Kong, Wei Cheng, Chuanyan Hao, and Qiaomin Lin. 2022. Data-driven personalized learning path planning based on cognitive diagnostic assessments in moocs. Applied Sciences, 12(8):3982. [find](https://scholar.google.com/scholar?q=Bo+Jiang%2C+Xinya+Li%2C+Shuhao+Yang%2C+Yaqi+Kong%2C+Wei+Cheng%2C+Chuanyan+Hao%2C+and+Qiaomin+Lin%2E+2022%2E+Data%2Ddriven+personalized+learning+path+planning+based+on+cognitive+diagnostic+assessments+in+moocs%2E+Applied+)
23. Soonwoo Kwon, Sojung Kim, Minju Park, Seunghyun Lee, and Kyuseok Kim. 2024. BIPED: pedagogically informed tutoring system for ESL education. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), ACL 2024, Bangkok, Thailand, August 11-16,… [find](https://scholar.google.com/scholar?q=Soonwoo+Kwon%2C+Sojung+Kim%2C+Minju+Park%2C+Seunghyun+Lee%2C+and+Kyuseok+Kim%2E+2024%2E+BIPED%3A+pedagogically+informed+tutoring+system+for+ESL+education%2E+In+Proceedings+of+the+62nd+Annual+Meeting+of+the+Associatio)
24. Haohao Luo, Yang Deng, Ying Shen, See-Kiong Ng, and Tat-Seng Chua. 2024. Chain-of-exemplar: Enhancing distractor generation for multimodal educational question generation. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), ACL 2024, Ba… [find](https://scholar.google.com/scholar?q=Haohao+Luo%2C+Yang+Deng%2C+Ying+Shen%2C+See%2DKiong+Ng%2C+and+Tat%2DSeng+Chua%2E+2024%2E+Chain%2Dof%2Dexemplar%3A+Enhancing+distractor+generation+for+multimodal+educational+question+generation%2E+In+Proceedings+of+the+62nd+A)
25. Xiangwei Lv, Guifeng Wang, Jingyuan Chen, Hejian Su, Zhiang Dong, Yumeng Zhu, Beishui Liao, and Fei Wu. 2025. Debiased cognition representation learning for knowledge tracing. ACM Transactions on Information Systems. [find](https://scholar.google.com/scholar?q=Xiangwei+Lv%2C+Guifeng+Wang%2C+Jingyuan+Chen%2C+Hejian+Su%2C+Zhiang+Dong%2C+Yumeng+Zhu%2C+Beishui+Liao%2C+and+Fei+Wu%2E+2025%2E+Debiased+cognition+representation+learning+for+knowledge+tracing%2E+ACM+Transactions+on+Info)
26. Julia M. Markel, Steven G. Opferman, James A. Landay, and Chris Piech. 2023. Gpteach: Interactive TA training with gpt-based students. In Proceedings of the Tenth ACM Conference on Learning @ Scale, Copenhagen, Denmark, July 20-22, 2023, pages 226–236. ACM. [find](https://scholar.google.com/scholar?q=Julia+M%2E+Markel%2C+Steven+G%2E+Opferman%2C+James+A%2E+Landay%2C+and+Chris+Piech%2E+2023%2E+Gpteach%3A+Interactive+TA+training+with+gpt%2Dbased+students%2E+In+Proceedings+of+the+Tenth+ACM+Conference+on+Learning+%40+Scale%2C+C)
27. OpenAI. 2023. [GPT-4 technical report](https://md.research.elamur.ai/free/en/papers/openai-2023-gpt-4-technical-report/). abs/2303.08774.
28. CoRR, Long Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida, Carroll L. Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, John Schulman, Jacob Hilton, Fraser Kelton, Luke Miller, Maddie Simens, Amanda Askell, Peter Welinder, Paul F. Christiano, Jan Leike, and Ryan Lowe.… — [Training language models to follow instructions with human feedback](https://md.research.elamur.ai/free/en/papers/ouyang-2022-training-language-models-follow-instructions/)
29. Xia Sun, Xu Zhao, Bo Li, Yuan Ma, Richard F. E. Sutcliffe, and Jun Feng. 2022. Dynamic key-value memory networks with rich features for knowledge tracing. IEEE Trans. Cybern., 52(8):8239–8245. [find](https://scholar.google.com/scholar?q=Xia+Sun%2C+Xu+Zhao%2C+Bo+Li%2C+Yuan+Ma%2C+Richard+F%2E+E%2E+Sutcliffe%2C+and+Jun+Feng%2E+2022%2E+Dynamic+key%2Dvalue+memory+networks+with+rich+features+for+knowledge+tracing%2E+IEEE+Trans%2E+Cybern%2E%2C+52%288%29%3A8239%E2%80%938245%2E)
30. Xiaoyu Tan, Shaojie Shi, Xihe Qiu, Chao Qu, Zhenting Qi, Yinghui Xu, and Yuan Qi. 2023. Self-criticism: Aligning large language models with their understanding of helpfulness, honesty, and harmlessness. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing: EMNLP… [find](https://scholar.google.com/scholar?q=Xiaoyu+Tan%2C+Shaojie+Shi%2C+Xihe+Qiu%2C+Chao+Qu%2C+Zhenting+Qi%2C+Yinghui+Xu%2C+and+Yuan+Qi%2E+2023%2E+Self%2Dcriticism%3A+Aligning+large+language+models+with+their+understanding+of+helpfulness%2C+honesty%2C+and+harmlessnes)
31. Kishore Papineni, Salim Roukos, Todd Ward, and Wei-Jing Zhu. 2002. [Bleu: a method for automatic evaluation of machine translation](https://md.research.elamur.ai/free/en/papers/papineni-2002-bleu-method-automatic-evaluation-machine/). In Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics, July 6-12, 2002, Philadelphia, PA, USA, pages 311–318. ACL.
32. Dongsheng Wang, Jiequan Cui, Miaoge Li, Wang Lin, Bo Chen, and Hanwang Zhang. 2024a. Instruction tuning-free visual token complement for multimodal llms. In Computer Vision - ECCV 2024 -18th European Conference, Milan, Italy, September 29-October 4, 2024, Proceedings, Part LXXXI, volume 15139 of Lec… [find](https://scholar.google.com/scholar?q=Dongsheng+Wang%2C+Jiequan+Cui%2C+Miaoge+Li%2C+Wang+Lin%2C+Bo+Chen%2C+and+Hanwang+Zhang%2E+2024a%2E+Instruction+tuning%2Dfree+visual+token+complement+for+multimodal+llms%2E+In+Computer+Vision+%2D+ECCV+2024+%2D18th+European+)
33. Ruchir Puri, David S. Kung, Geert Janssen, Wei Zhang, Giacomo Domeniconi, Vladimir Zolotov, Julian Dolby, Jie Chen, Mihir R. Choudhury, Lindsey Decker, Veronika Thost, Luca Buratti, Saurabh Pujar, Shyam Ramji, Ulrich Finkler, Susan Malaika, and Frederick Reiss. 2021. [Codenet: A large-scale AI for co…](https://md.research.elamur.ai/free/en/papers/puri-2021-codenet-large-scale-ai-code/)
34. Rose E. Wang, Qingyang Zhang, Carly Robinson, Susanna Loeb, and Dorottya Demszky. 2024b. Bridging the novice-expert gap via models of decision-making: A case study on remediating math mistakes. In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational L… [find](https://scholar.google.com/scholar?q=Rose+E%2E+Wang%2C+Qingyang+Zhang%2C+Carly+Robinson%2C+Susanna+Loeb%2C+and+Dorottya+Demszky%2E+2024b%2E+Bridging+the+novice%2Dexpert+gap+via+models+of+decision%2Dmaking%3A+A+case+study+on+remediating+math+mistakes%2E+In+Pro)
35. Alexis Ross and Jacob Andreas. 2024. Toward in-context teaching: Adapting examples to students’ misconceptions. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), ACL 2024, Bangkok, Thailand, August 11-16, 2024, pages 13283–13310. Asso… [find](https://scholar.google.com/scholar?q=Alexis+Ross+and+Jacob+Andreas%2E+2024%2E+Toward+in%2Dcontext+teaching%3A+Adapting+examples+to+students%E2%80%99+misconceptions%2E+In+Proceedings+of+the+62nd+Annual+Meeting+of+the+Association+for+Computational+Linguisti)
36. Shashank Sonkar, Xinghe Chen, Naiming Liu, Richard G. Baraniuk, and Mrinmaya Sachan. 2024. Llm-based cognitive models of students with misconceptions. CoRR, abs/2410.12294. — [arXiv:2410.12294](https://arxiv.org/abs/2410.12294) [find](https://scholar.google.com/scholar?q=Shashank+Sonkar%2C+Xinghe+Chen%2C+Naiming+Liu%2C+Richard+G%2E+Baraniuk%2C+and+Mrinmaya+Sachan%2E+2024%2E+Llm%2Dbased+cognitive+models+of+students+with+misconceptions%2E+CoRR%2C+abs%2F2410%2E12294%2E+%E2%80%94+%5BarXiv%3A2410%2E12294%5D%28https%3A)
37. William Stout. 2007. Skills diagnosis using irt-based continuous latent trait models. Journal of Educational Measurement, 44(4):313–324. [find](https://scholar.google.com/scholar?q=William+Stout%2E+2007%2E+Skills+diagnosis+using+irt%2Dbased+continuous+latent+trait+models%2E+Journal+of+Educational+Measurement%2C+44%284%29%3A313%E2%80%93324%2E)
38. Yu Su, Qingwen Liu, Qi Liu, Zhenya Huang, Yu Yin, Enhong Chen, Chris H. Q. Ding, Si Wei, and Guoping Hu. 2018. Exercise-enhanced sequential modeling for student performance prediction. In Proceedings of the Thirty-Second AAAI Conference on Artificial Intelligence, (AAAI-18), the 30th innovative Appl… [find](https://scholar.google.com/scholar?q=Yu+Su%2C+Qingwen+Liu%2C+Qi+Liu%2C+Zhenya+Huang%2C+Yu+Yin%2C+Enhong+Chen%2C+Chris+H%2E+Q%2E+Ding%2C+Si+Wei%2C+and+Guoping+Hu%2E+2018%2E+Exercise%2Denhanced+sequential+modeling+for+student+performance+prediction%2E+In+Proceedings+)
39. Shulei Wang, Wang Lin, Hai Huang, Hanting Wang, Sihang Cai, WenKang Han, Tao Jin, Jingyuan Chen, Jiacheng Sun, Jieming Zhu, and Zhou Zhao. 2025. Towards transformer-based aligned generation with self-coherence guidance. CoRR, abs/2503.17675. — [arXiv:2503.17675](https://arxiv.org/abs/2503.17675) [find](https://scholar.google.com/scholar?q=Shulei+Wang%2C+Wang+Lin%2C+Hai+Huang%2C+Hanting+Wang%2C+Sihang+Cai%2C+WenKang+Han%2C+Tao+Jin%2C+Jingyuan+Chen%2C+Jiacheng+Sun%2C+Jieming+Zhu%2C+and+Zhou+Zhao%2E+2025%2E+Towards+transformer%2Dbased+aligned+generation+with+self%2D)
40. Wenjie Wang, Fuli Feng, Xiangnan He, Liqiang Nie, and Tat-Seng Chua. 2021a. Denoising implicit feedback for recommendation. In WSDM ’21, The Fourteenth ACM International Conference on Web Search and Data Mining, Virtual Event, Israel, March 8-12, 2021, pages 373–381. ACM. [find](https://scholar.google.com/scholar?q=Wenjie+Wang%2C+Fuli+Feng%2C+Xiangnan+He%2C+Liqiang+Nie%2C+and+Tat%2DSeng+Chua%2E+2021a%2E+Denoising+implicit+feedback+for+recommendation%2E+In+WSDM+%E2%80%9921%2C+The+Fourteenth+ACM+International+Conference+on+Web+Search+and+D)
41. Wenjie Wang, Fuli Feng, Xiangnan He, Hanwang Zhang, and Tat-Seng Chua. 2021b. Clicks can be cheating: Counterfactual recommendation for mitigating clickbait issue. In SIGIR ’21: The 44th International ACM SIGIR Conference on Research and Development in Information Retrieval, Virtual Event, Canada, J… [find](https://scholar.google.com/scholar?q=Wenjie+Wang%2C+Fuli+Feng%2C+Xiangnan+He%2C+Hanwang+Zhang%2C+and+Tat%2DSeng+Chua%2E+2021b%2E+Clicks+can+be+cheating%3A+Counterfactual+recommendation+for+mitigating+clickbait+issue%2E+In+SIGIR+%E2%80%9921%3A+The+44th+International)
42. Wenjie Wang, Xinyu Lin, Fuli Feng, Xiangnan He, Min Lin, and Tat-Seng Chua. 2022. Causal representation learning for out-of-distribution recommendation. In WWW ’22: The ACM Web Conference 2022, Virtual Event, Lyon, France, April 25 - 29, 2022, pages 3562– 3571. ACM. [find](https://scholar.google.com/scholar?q=Wenjie+Wang%2C+Xinyu+Lin%2C+Fuli+Feng%2C+Xiangnan+He%2C+Min+Lin%2C+and+Tat%2DSeng+Chua%2E+2022%2E+Causal+representation+learning+for+out%2Dof%2Ddistribution+recommendation%2E+In+WWW+%E2%80%9922%3A+The+ACM+Web+Conference+2022%2C+Virtua)
43. Ye Wang, Wang Lin, Shengyu Zhang, Tao Jin, Linjun Li, Xize Cheng, and Zhou Zhao. 2023a. Weakly-supervised spoken video grounding via semantic interaction learning. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), ACL 2023, Toronto, C… [find](https://scholar.google.com/scholar?q=Ye+Wang%2C+Wang+Lin%2C+Shengyu+Zhang%2C+Tao+Jin%2C+Linjun+Li%2C+Xize+Cheng%2C+and+Zhou+Zhao%2E+2023a%2E+Weakly%2Dsupervised+spoken+video+grounding+via+semantic+interaction+learning%2E+In+Proceedings+of+the+61st+Annual+Me)
44. Zhifeng Wang, Wenxing Yan, Chunyan Zeng, Yuan Tian, and Shi Dong. 2023b. A unified interpretable intelligent learning diagnosis framework for learning performance prediction in intelligent tutoring systems. International Journal of Intelligent Systems, 2023(1):4468025. [find](https://scholar.google.com/scholar?q=Zhifeng+Wang%2C+Wenxing+Yan%2C+Chunyan+Zeng%2C+Yuan+Tian%2C+and+Shi+Dong%2E+2023b%2E+A+unified+interpretable+intelligent+learning+diagnosis+framework+for+learning+performance+prediction+in+intelligent+tutoring+sy)
45. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Brian Ichter, Fei Xia, Ed H. Chi, Quoc V. Le, and Denny Zhou. 2022. [Chain-of-thought prompting elicits reasoning in large language models](https://md.research.elamur.ai/free/en/papers/wei-2022-chain-thought-prompting-elicits-reasoning/). In Advances in Neural Information Processing Systems 35: Annual Conference on Neural Information Processi…
46. Tao Wu, Mengze Li, Jingyuan Chen, Wei Ji, Wang Lin, Jinyang Gao, Kun Kuang, Zhou Zhao, and Fei Wu. 2024. Semantic alignment for multimodal large language models. In Proceedings of the 32nd ACM International Conference on Multimedia, MM 2024, Melbourne, VIC, Australia, 28 October 2024 - 1 November 20… [find](https://scholar.google.com/scholar?q=Tao+Wu%2C+Mengze+Li%2C+Jingyuan+Chen%2C+Wei+Ji%2C+Wang+Lin%2C+Jinyang+Gao%2C+Kun+Kuang%2C+Zhou+Zhao%2C+and+Fei+Wu%2E+2024%2E+Semantic+alignment+for+multimodal+large+language+models%2E+In+Proceedings+of+the+32nd+ACM+Interna)
47. Di Zhang, Xiaoshui Huang, Dongzhan Zhou, Yuqiang Li, and Wanli Ouyang. 2024. Accessing GPT-4 level mathematical olympiad solutions via monte carlo tree self-refine with llama-3 8b. CoRR, abs/2406.07394. [find](https://scholar.google.com/scholar?q=Di+Zhang%2C+Xiaoshui+Huang%2C+Dongzhan+Zhou%2C+Yuqiang+Li%2C+and+Wanli+Ouyang%2E+2024%2E+Accessing+GPT%2D4+level+mathematical+olympiad+solutions+via+monte+carlo+tree+self%2Drefine+with+llama%2D3+8b%2E+CoRR%2C+abs%2F2406%2E0739)
48. Yiyun Zhou, WenKang Han, and Jingyuan Chen. 2025a. Revisiting applicable and comprehensive knowledge tracing in large-scale data. CoRR, abs/2501.14256. — [arXiv:2501.14256](https://arxiv.org/abs/2501.14256) [find](https://scholar.google.com/scholar?q=Yiyun+Zhou%2C+WenKang+Han%2C+and+Jingyuan+Chen%2E+2025a%2E+Revisiting+applicable+and+comprehensive+knowledge+tracing+in+large%2Dscale+data%2E+CoRR%2C+abs%2F2501%2E14256%2E+%E2%80%94+%5BarXiv%3A2501%2E14256%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2501%2E)
49. Yiyun Zhou, Zheqi Lv, Shengyu Zhang, and Jingyuan Chen. 2024. Cuff-KT: Tackling learners’ real-time learning pattern adjustment via tuning-free knowledge state-guided model updating. [find](https://scholar.google.com/scholar?q=Yiyun+Zhou%2C+Zheqi+Lv%2C+Shengyu+Zhang%2C+and+Jingyuan+Chen%2E+2024%2E+Cuff%2DKT%3A+Tackling+learners%E2%80%99+real%2Dtime+learning+pattern+adjustment+via+tuning%2Dfree+knowledge+state%2Dguided+model+updating%2E)
50. Yiyun Zhou, Zheqi Lv, Shengyu Zhang, and Jingyuan Chen. 2025b. Disentangled knowledge tracing for alleviating cognitive bias. In Proceedings of the ACM on Web Conference 2025, WWW 2025, Sydney, NSW, Australia, 28 April 2025- 2 May 2025, pages 2633– 2645. ACM. [find](https://scholar.google.com/scholar?q=Yiyun+Zhou%2C+Zheqi+Lv%2C+Shengyu+Zhang%2C+and+Jingyuan+Chen%2E+2025b%2E+Disentangled+knowledge+tracing+for+alleviating+cognitive+bias%2E+In+Proceedings+of+the+ACM+on+Web+Conference+2025%2C+WWW+2025%2C+Sydney%2C+NSW%2C+A)
51. In this appendix, we present the following content: A Dataset Statistics and Details A.1 Privacy Protection . . . . . . . . . A.2 Dataset Statistics . . . . . . . . . A.3 Significance of Programming Task for Student Simulation . . . . . . A.4 Data Generalizability . . . . . . . 14 14 14 B Details on… [find](https://scholar.google.com/scholar?q=In+this+appendix%2C+we+present+the+following+content%3A+A+Dataset+Statistics+and+Details+A%2E1+Privacy+Protection+%2E+%2E+%2E+%2E+%2E+%2E+%2E+%2E+%2E+A%2E2+Dataset+Statistics+%2E+%2E+%2E+%2E+%2E+%2E+%2E+%2E+%2E+A%2E3+Significance+of+Programming+T)
52. Weicai Yan, Wang Lin, Zirun Guo, Ye Wang, Fangming Feng, Xiaoda Yang, Zehan Wang, and Tao Jin. 2025. Diff-prompt: Diffusion-driven prompt generator with mask supervision. In The Thirteenth International Conference on Learning Representations, ICLR 2025, Singapore, April 24-28, 2025. OpenReview.net. [find](https://scholar.google.com/scholar?q=Weicai+Yan%2C+Wang+Lin%2C+Zirun+Guo%2C+Ye+Wang%2C+Fangming+Feng%2C+Xiaoda+Yang%2C+Zehan+Wang%2C+and+Tao+Jin%2E+2025%2E+Diff%2Dprompt%3A+Diffusion%2Ddriven+prompt+generator+with+mask+supervision%2E+In+The+Thirteenth+Internation)
53. C Method Details C.1 Details on Concept Relationships . C.2 Edge Category Conflict Resolving C.3 Scalability and Computational Feasibility . . . . . . . . . . . . . . . C.4 Error Attribution and Traceability 15 15 16 [find](https://scholar.google.com/scholar?q=C+Method+Details+C%2E1+Details+on+Concept+Relationships+%2E+C%2E2+Edge+Category+Conflict+Resolving+C%2E3+Scalability+and+Computational+Feasibility+%2E+%2E+%2E+%2E+%2E+%2E+%2E+%2E+%2E+%2E+%2E+%2E+%2E+%2E+%2E+C%2E4+Error+Attribution+and+Trace)
54. Rui Yang, Boming Yang, Sixun Ouyang, Tianwei She, Aosong Feng, Yuang Jiang, Freddy Lécué, Jinghui Lu, and Irene Li. 2024. Graphusion: Leveraging large language models for scientific knowledge graph fusion and construction in NLP education. CoRR, abs/2407.10794. — [arXiv:2407.10794](https://arxiv.org/abs/2407.10794) [find](https://scholar.google.com/scholar?q=Rui+Yang%2C+Boming+Yang%2C+Sixun+Ouyang%2C+Tianwei+She%2C+Aosong+Feng%2C+Yuang+Jiang%2C+Freddy+L%C3%A9cu%C3%A9%2C+Jinghui+Lu%2C+and+Irene+Li%2E+2024%2E+Graphusion%3A+Leveraging+large+language+models+for+scientific+knowledge+graph+fu)
55. D Experiment Details D.1 Model Details . . . . . . . . . . . D.2 Baseline Details . . . . . . . . . . D.3 Metric Details . . . . . . . . . . . 17 17 18 19 [find](https://scholar.google.com/scholar?q=D+Experiment+Details+D%2E1+Model+Details+%2E+%2E+%2E+%2E+%2E+%2E+%2E+%2E+%2E+%2E+%2E+D%2E2+Baseline+Details+%2E+%2E+%2E+%2E+%2E+%2E+%2E+%2E+%2E+%2E+D%2E3+Metric+Details+%2E+%2E+%2E+%2E+%2E+%2E+%2E+%2E+%2E+%2E+%2E+17+17+18+19)
56. Hanyi Xu, Wensheng Gan, Zhenlian Qi, Jiayang Wu, and Philip S. Yu. 2024a. Large language models for education: A survey. CoRR, abs/2405.13001. — [arXiv:2405.13001](https://arxiv.org/abs/2405.13001) [find](https://scholar.google.com/scholar?q=Hanyi+Xu%2C+Wensheng+Gan%2C+Zhenlian+Qi%2C+Jiayang+Wu%2C+and+Philip+S%2E+Yu%2E+2024a%2E+Large+language+models+for+education%3A+A+survey%2E+CoRR%2C+abs%2F2405%2E13001%2E+%E2%80%94+%5BarXiv%3A2405%2E13001%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2405%2E13001%29)
57. Songlin Xu, Xinyu Zhang, and Lianhui Qin. 2024b. Eduagent: Generative student agents in learning. CoRR, abs/2404.07963. — [arXiv:2404.07963](https://arxiv.org/abs/2404.07963) [find](https://scholar.google.com/scholar?q=Songlin+Xu%2C+Xinyu+Zhang%2C+and+Lianhui+Qin%2E+2024b%2E+Eduagent%3A+Generative+student+agents+in+learning%2E+CoRR%2C+abs%2F2404%2E07963%2E+%E2%80%94+%5BarXiv%3A2404%2E07963%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2404%2E07963%29)
58. Shunyu Yao, Dian Yu, Jeffrey Zhao, Izhak Shafran, Tom Griffiths, Yuan Cao, and Karthik Narasimhan. 2023. Tree of thoughts: Deliberate problem solving with large language models. In Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023,… [find](https://scholar.google.com/scholar?q=Shunyu+Yao%2C+Dian+Yu%2C+Jeffrey+Zhao%2C+Izhak+Shafran%2C+Tom+Griffiths%2C+Yuan+Cao%2C+and+Karthik+Narasimhan%2E+2023%2E+Tree+of+thoughts%3A+Deliberate+problem+solving+with+large+language+models%2E+In+Advances+in+Neural+)
59. Murong Yue, Wijdane Mifdal, Yixuan Zhang, Jennifer Suh, and Ziyu Yao. 2024. Mathvc: An llm-simulated multi-character virtual classroom for mathematics education. CoRR, abs/2404.06711. 14 15 16 17 — [arXiv:2404.06711](https://arxiv.org/abs/2404.06711) [find](https://scholar.google.com/scholar?q=Murong+Yue%2C+Wijdane+Mifdal%2C+Yixuan+Zhang%2C+Jennifer+Suh%2C+and+Ziyu+Yao%2E+2024%2E+Mathvc%3A+An+llm%2Dsimulated+multi%2Dcharacter+virtual+classroom+for+mathematics+education%2E+CoRR%2C+abs%2F2404%2E06711%2E+14+15+16+17+%E2%80%94+%5Ba)
60. E More Experimental Results and Analysis 19 E.1 End-to-end Comparison on Student_100 . . . . . . . . . . . 19 E.2 Results for Captioning Metrics . . 20 E.3 Results on Java and C++ programming . . . . . . . . . . . . . . . . 20 E.4 In-Depth Analysis Details . . . . . 20 E.5 Bad Case Analysis . . . .… [find](https://scholar.google.com/scholar?q=E+More+Experimental+Results+and+Analysis+19+E%2E1+End%2Dto%2Dend+Comparison+on+Student%5F100+%2E+%2E+%2E+%2E+%2E+%2E+%2E+%2E+%2E+%2E+%2E+19+E%2E2+Results+for+Captioning+Metrics+%2E+%2E+20+E%2E3+Results+on+Java+and+C%2B%2B+programming+%2E+%2E+%2E+%2E+)

*Updated 2 November 2026*