Elamur Research⌕
☰
home / papers / 2024 / Stepwise Verification and Remediation of Student Reasoning Errors with Large Language Model Tutors

Stepwise Verification and Remediation of Student Reasoning Errors with Large Language Model Tutors

Paper · 2024 · Nico Daheim, Jakub Macina, Manu Kapur et al.

Summary

LLM-based math tutors can solve word problems well, but they often fail to notice where a student went wrong. As a result, they may praise wrong answers or give feedback that misses the actual mistake. The authors split tutoring into two stages, mirroring how human teachers work. First, a verifier checks the student's solution. Then a separate generator writes the tutor's reply, conditioned on the verifier's output.

To support this, they extend the MathDial dialog tutoring corpus with about 1K stepwise student reasoning chains. In each chain, teachers marked the first erroneous step. They compare several verifiers, using both prompting and finetuning:

  • binary and step-level classifiers
  • free-text error descriptions
  • an alignment between student steps and a reference solution

Locating the first error turns out to be hard for current models. However, finetuning smaller LLMs on the new data lets them beat prompted state-of-the-art models.

In automatic metrics and in teacher-based human evaluation, verification-grounded replies were more targeted, more often correct, less hallucinated, and more actionable. The gains were largest when the verification itself was correct.

Key points

  • Dataset: about 1K math reasoning chains with teacher-annotated first error steps, built on MathDial and released under CC-BY-4.0. Code is at github.com/eth-lre/verify-then-generate.
  • Verifier types compared:
    • Overall Verification (binary)
    • Stepwise Verification (multi-class, with an iterative variant)
    • textual Error Description
    • Step Alignment against a reference solution
  • Baseline: conditioning on coarse error categories ("Error Reason," Wang et al., 2024b). Error Description and Step Alignment reduced hallucinations most clearly.
  • Finetuning result: smaller LLMs finetuned on the annotations outperformed prompted state-of-the-art LLMs at verification.
  • Limitations: English only; reasoning chains are LLM-simulated rather than written by real students; evaluation was done by teachers, not students.

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

 Institutions

🎓Technische Universität DarmstadtDarmstadt · Germany123 works · 36,745 citations🎓ETH ZurichZurich · Switzerland97 works · 9,612 citations🔬Hessian Center for Artificial IntelligenceDarmstadt · Germany17 works · 787 citations

 Fields, topics and concepts

📚 Shelves 5The field of science this work belongs to (OpenAlex fields).Computer Science · 19,503 worksArtificial Intelligence · 18,805 worksArts and Humanities · 17,959 worksLanguage and Linguistics · 17,789 worksInformation Systems · 1,888 works

🧭 Topics 2Research subjects and languages the work is about.Language models & pretraining · 4,145 worksReasoning · 1,948 works

Cite

Cite as 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. Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2024.emnlp-main.478 · arXiv 2407.09136

@inproceedings{daheim2024stepwise,
  title = {{Stepwise Verification and Remediation of Student Reasoning Errors with Large Language Model Tutors}},
  author = {Nico Daheim, Jakub Macina, Manu Kapur, Iryna Gurevych and Mrinmaya Sachan},
  booktitle = {Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing},
  year = {2024},
  doi = {10.18653/v1/2024.emnlp-main.478},
  eprint = {2407.09136},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.18653/v1/2024.emnlp-main.478}
}

References

48 references, 23 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Leonard Adolphs, Kurt Shuster, Jack Urbanek, Arthur Szlam, and Jason Weston. 2022. Reason first, then respond: Modular generation for knowledge-infused dialogue. In Findings of the Association for Computational Linguistics: EMNLP 2022, pages 7112– 7132, Abu Dhabi, United Arab Emirates. Association f… find
  2. Vincent Aleven, Bruce M McLaren, Jonathan Sewall, Martin Van Velsen, Octav Popescu, Sandra Demi, Michael Ringenberg, and Kenneth R Koedinger. 2016. Example-tracing tutors: Intelligent tutor development for non-programmers. International Journal of Artificial Intelligence in Education, 26:224–269. find
  3. John R Anderson, C Franklin Boyle, and Brian J Reiser. 1985. Intelligent tutoring systems. Science, 228(4698):456–462. find
  4. Julia Anghileri. 2006. Scaffolding practices that enhance mathematics learning. Journal of Mathematics Teacher Education, 9:33–52. find
  5. Rishi Bommasani, Drew A. Hudson, Ehsan Adeli, Russ Altman, Simran Arora, Sydney von Arx, Michael S. Bernstein, Jeannette Bohg, Antoine Bosselut, Emma Brunskill, Erik Brynjolfsson, S. Buch, Dallas Card, Rodrigo Castellon, Niladri S. Chatterji, Annie S. Chen, Kathleen A. Creel, Jared Davis, Dora Demsz… — On the Opportunities and Risks of Foundation Models
  6. Andrew Caines, Helen Yannakoudakis, Helen Allen, Pascual Pérez-Paredes, Bill Byrne, and Paula Buttery. 2022. The teacher-student chatroom corpus version 2: more lessons, new annotation, automatic detection of sequence shifts. In Swedish Language Technology Conference and NLP4CALL, pages 23–35. find
  7. Andrew Caines, Helen Yannakoudakis, Helena Edmondson, Helen Allen, Pascual Pérez-Paredes, Bill Byrne, and Paula Buttery. 2020. The teacher-student chatroom corpus. In Proceedings of the 9th Workshop on NLP for Computer Assisted Language Learning, pages 10–20.
  8. Alexis Chevalier, Jiayi Geng, Alexander Wettig, Howard Chen, Sebastian Mizera, Toni Annala, Max Aragon, Arturo Rodriguez Fanlo, Simon Frieder, Simon Machado, Akshara Prabhakar, Ellie Thieu, Jiachen T. Wang, Zirui Wang, Xindi Wu, Mengzhou Xia, Wenhan Xia, Jiatong Yu, Junjie Zhu, Zhiyong Ren, Sanjeev… — Language Models as Science Tutors
  9. Karl Cobbe, Vineet Kosaraju, Mohammad Bavarian, Mark Chen, Heewoo Jun, Lukasz Kaiser, Matthias Plappert, Jerry Tworek, Jacob Hilton, Reiichiro Nakano, et al. 2021. Training verifiers to solve math word problems. arXiv preprint arXiv:2110.14168.
  10. Jacob Cohen. 1960. A coefficient of agreement for nominal scales. Educational and psychological measurement, 20(1):37–46. find
  11. Nico Daheim, Nouha Dziri, Mrinmaya Sachan, Iryna Gurevych, and Edoardo Ponti. 2024. Elastic weight removal for faithful and abstractive dialogue generation. In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologi… find
  12. Olga Golovneva, Moya Peng Chen, Spencer Poff, Martin Corredor, Luke Zettlemoyer, Maryam Fazel-Zarandi, and Asli Celikyilmaz. 2023. ROSCOE: A suite of metrics for scoring step-by-step reasoning. In The Eleventh International Conference on Learning Representations.
  13. Dan Hendrycks, Collin Burns, Saurav Kadavath, Akul Arora, Steven Basart, Eric Tang, Dawn Song, and Jacob Steinhardt. 2021. Measuring mathematical problem solving with the MATH dataset. In Thirty-fifth Conference on Neural Information Processing Systems Datasets and Benchmarks Track (Round 2).
  14. Jie Huang, Xinyun Chen, Swaroop Mishra, Huaixiu Steven Zheng, Adams Wei Yu, Xinying Song, and Denny Zhou. 2024. Large language models cannot self-correct reasoning yet. In The Twelfth International Conference on Learning Representations.
  15. Sachan. 2023b. Opportunities and challenges in neural dialog tutoring. In Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics, pages 2357–2372, Dubrovnik, Croatia. Association for Computational Linguistics.
  16. Aman Madaan, Niket Tandon, Prakhar Gupta, Skyler Hallinan, Luyu Gao, Sarah Wiegreffe, Uri Alon, Nouha Dziri, Shrimai Prabhumoye, Yiming Yang, Shashank Gupta, Bodhisattwa Prasad Majumder, Katherine Hermann, Sean Welleck, Amir Yazdanbakhsh, and Peter Clark. 2023. Self-refine: Iterative refinement with… find
  17. Alon Jacovi, Yonatan Bitton, Bernd Bohnet, Jonathan Herzig, Or Honovich, Michael Tseng, Michael Collins, Roee Aharoni, and Mor Geva. 2024. A chain-of-thought is as strong as its weakest link: A benchmark for verifiers of reasoning chains. In Proceedings of the 62nd Annual Meeting of the Association…
  18. Saul B Needleman and Christian D Wunsch. 1970. A general method applicable to the search for similarities in the amino acid sequence of two proteins. Journal of molecular biology, 48(3):443–453. find
  19. Irina Jurenka, Markus Kunesch, Kevin R McKee, Daniel Gillick, Shaojian Zhu, Sara Wiltberger, Shubham Milind Phal, Katherine Hermann, Daniel Kasenberg, Avishkar Bhoopchand, et al. 2024. Towards responsible development of generative ai for education: An evaluation-driven approach. arXiv preprint arXiv…
  20. Andreas Opedal, Alessandro Stolfo, Haruki Shirakami, Ying Jiao, Ryan Cotterell, Bernhard Schölkopf, Abulhair Saparov, and Mrinmaya Sachan. 2024. Do language models exhibit the same cognitive biases in problem solving as human learners? In Forty-first International Conference on Machine Learning. find
  21. Manu Kapur. 2016. Examining productive failure, productive success, unproductive failure, and unproductive success in learning. Educational Psychologist, 51(2):289–299. find
  22. Sankalan Pal Chowdhury, Vilém Zouhar, and Mrinmaya Sachan. 2024. Autotutor meets large language models: A language model tutor with rich pedagogy and guardrails. In Proceedings of the Eleventh ACM Conference on Learning@ Scale, pages 5–15.
  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), pages 3389–3414, Bangkok, Thailand. Associa…
  24. Giovanni Paolini, Ben Athiwaratkun, Jason Krone, Jie Ma, Alessandro Achille, Rishita Anubhai, Cicero Nogueira dos Santos, Bing Xiang, and Stefano Soatto. 2021. Structured prediction as translation between augmented natural languages. In International Conference on Learning Representations. find
  25. Mark R. Lepper and Maria Woolverton. 2002. Chapter 7 - the wisdom of practice: Lessons learned from the study of highly effective tutors. In Joshua Aronson, editor, Improving Academic Achievement, Educational Psychology, pages 135–158. Academic Press, San Diego. find
  26. Jakub Macina, Nico Daheim, Sankalan Chowdhury, Tanmay Sinha, Manu Kapur, Iryna Gurevych, and Mrinmaya Sachan. 2023a. MathDial: A dialogue tutoring dataset with rich pedagogical properties grounded in math reasoning problems. In Findings of the Association for Computational Linguistics: EMNLP 2023, p…
  27. Jakub Macina, Nico Daheim, Lingzhi Wang, Tanmay Sinha, Manu Kapur, Iryna Gurevych, and Mrinmaya Benjamin D Nye, Arthur C Graesser, and Xiangen Hu. 2014. Autotutor and family: A review of 17 years of natural language tutoring. International Journal of Artificial Intelligence in Education, 24:427–469. find
  28. Matt Post. 2018. A call for clarity in reporting BLEU scores. In Proceedings of the Third Conference on Machine Translation: Research Papers, pages 186– 191, Brussels, Belgium. Association for Computational Linguistics.
  29. Ofir Press, Muru Zhang, Sewon Min, Ludwig Schmidt, Noah Smith, and Mike Lewis. 2023. Measuring and narrowing the compositionality gap in language models. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 5687–5711, Singapore. Association for Computational Linguistics. find
  30. Nils Reimers and Iryna Gurevych. 2019. Sentence-bert: Sentence embeddings using siamese bert-networks. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing. Association for Computational Linguistics.
  31. Brian J. Reiser. 2004. Scaffolding complex learning: The mechanisms of structuring and problematizing student work. Journal of the Learning Sciences, 13(3):273–304. find
  32. 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), pages 13283– 13310, Bangkok, Thailand. Association for Computational Lin… find
  33. Swarnadeep Saha, Peter Hase, and Mohit Bansal. 2023. Can language models teach? teacher explanations improve student performance via personalization. In Thirty-seventh Conference on Neural Information Processing Systems. find
  34. Meghan Shaughnessy, Rosalie DeFino, Erin Pfaff, and Merrie Blunk. 2021. I think i made a mistake: How do prospective teachers elicit the thinking of a student who has made a mistake? Journal of Mathematics Teacher Education, 24:335–359. find
  35. Noah Shinn, Federico Cassano, Ashwin Gopinath, Karthik Narasimhan, and Shunyu Yao. 2024. Reflexion: Language agents with verbal reinforcement learning. Advances in Neural Information Processing Systems, 36.
  36. Kumar Shridhar, Jakub Macina, Mennatallah El-Assady, Tanmay Sinha, Manu Kapur, and Mrinmaya Sachan. 2022. Automatic generation of socratic subquestions for teaching math word problems. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 4136–4149, Abu Dha…
  37. Tanmay Sinha and Manu Kapur. 2021. When problem solving followed by instruction works: Evidence for productive failure. Review of Educational Research, 91(5):761–798. find
  38. Shashank Sonkar, Naiming Liu, Debshila Mallick, and Richard Baraniuk. 2023. Class: A design framework for building intelligent tutoring systems based on learning science principles. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 1941–1961.
  39. Katherine Stasaski, Kimberly Kao, and Marti A. Hearst. 2020. CIMA: A large open access dialogue dataset for tutoring. In Proceedings of the Fifteenth Workshop on Innovative Use of NLP for Building Educational Applications, pages 52–64, Seattle, WA, USA → Online. Association for Computational Linguis…
  40. Romal Thoppilan, Daniel De Freitas, Jamie Hall, Noam Shazeer, Apoorv Kulshreshtha, Heng-Tze Cheng, Alicia Jin, Taylor Bos, Leslie Baker, Yu Du, et al. 2022. Lamda: Language models for dialog applications. arXiv preprint arXiv:2201.08239.
  41. Gladys Tyen, Hassan Mansoor, Victor Carbune, Peter Chen, and Tony Mak. 2024. LLMs cannot find reasoning errors, but can correct them given the error location. In Findings of the Association for Computational Linguistics ACL 2024, pages 13894–13908, Bangkok, Thailand and virtual meeting. Association…
  42. Junling Wang, Jakub Macina, Nico Daheim, Sankalan Pal Chowdhury, and Mrinmaya Sachan. 2024a. Book2Dial: Generating teacher student interactions from textbooks for cost-effective development of educational chatbots. In Findings of the Association for Computational Linguistics ACL 2024, pages 9707– 97… find
  43. 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
  44. Thomas Wolf, Lysandre Debut, Victor Sanh, Julien Chaumond, Clement Delangue, Anthony Moi, Pier-ric Cistac, Tim Rault, Remi Louf, Morgan Funtow-icz, Joe Davison, Sam Shleifer, Patrick von Platen, Clara Ma, Yacine Jernite, Julien Plu, Canwen Xu, Teven Le Scao, Sylvain Gugger, Mariama Drame, Quentin Lh… — Transformers: State-of-the-Art Natural Language Processing
  45. Sebastian Wollny, Jan Schneider, Daniele Di Mitri, Joshua Weidlich, Marc Rittberger, and Hendrik Drachsler. 2021. Are we there yet?-a systematic literature review on chatbots in education. Frontiers in artificial intelligence, 4. find
  46. An-Zi Yen and Wei-Ling Hsu. 2023. Three questions concerning the use of large language models to facilitate mathematics learning. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 3055–3069, Singapore. Association for Computational Linguistics. find
  47. Tianyi Zhang, Varsha Kishore, Felix Wu, Kilian Q. Weinberger, and Yoav Artzi. 2020. Bertscore: Evaluating text generation with bert. In International Conference on Learning Representations. find
  48. Lianmin Zheng, Wei-Lin Chiang, Ying Sheng, Siyuan Zhuang, Zhanghao Wu, Yonghao Zhuang, Zi Lin, Zhuohan Li, Dacheng Li, Eric Xing, et al. 2024. Judging llm-as-a-judge with mt-bench and chatbot arena. Advances in Neural Information Processing Systems, 36. *Figure 3: Dataset Distribution. The index of…

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/daheim-2024-stepwise-verification-remediation-student-reasoning/