Towards Pedagogically Aligned LLM Tutors for Math Mistake Remediation

• URL: http://arxiv.org/abs/2606.21502v1
• Date: Fri, 19 Jun 2026 14:53:55 GMT
• Title: Towards Pedagogically Aligned LLM Tutors for Math Mistake Remediation
• Abstract summary: We study the application of a two-stage alignment pipeline for math mistake remediation.<n>We construct a dataset that integrates existing tutoring corpora with synthetic data generated along pedagogical dimensions.
• Score: 12.296644292843801
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language models have strong potential for use in intelligent tutoring systems, but they often fail to follow effective pedagogical strategies, such as guiding students without revealing final answers. We study the application of a two-stage alignment pipeline for math mistake remediation, combining supervised fine-tuning on tutoring dialogs with Direct Preference Optimization on synthetic preference pairs. We construct a dataset that integrates existing tutoring corpora with synthetic data generated along pedagogical dimensions, such as scaffolding and factuality, and study different input configurations that incorporate solution correctness and gold answers. Experiments show that this approach improves both factual accuracy and pedagogical quality over base models and existing tutoring models. Human evaluation further indicates that our best model is competitive with a strong proprietary baseline, while providing additional benefits in terms of openness, transparency, and reproducibility. Our results highlight the effectiveness of preference-based pedagogical alignment, while also revealing challenges in reliably evaluating tutoring quality.

Related papers

• Measuring Whether LLM Tutors Teach or Solve: A Diagnostic for Educational Impact [4.6840536642016195]
  We study whether public tutoring benchmarks distinguish learning-supportive behavior from mere answer production.<n>We propose a lightweight diagnostic rubric based on the gap between solving-oriented and pedagogy-oriented benchmark performance.
  arXiv  Detail & Related papers  (2026-06-15T04:32:18Z)
• PEARL: Training Socratic Tutors with Pedagogically Aligned Reinforcement Learning [18.478727410749205]
  Large Language Models (LLMs) have shown promise as educational tutors, yet effective tutoring requires more than solving problems.<n>We propose PEARL, a pedagogically aligned reinforcement learning framework for training Socratic tutoring agents.
  arXiv  Detail & Related papers  (2026-05-28T08:25:08Z)
• From Problem-Solving to Teaching Problem-Solving: Aligning LLMs with Pedagogy using Reinforcement Learning [82.50157695987558]
  Large language models (LLMs) can transform education, but their optimization for direct question-answering often undermines effective pedagogy.<n>We propose an online reinforcement learning (RL)-based alignment framework that can quickly adapt LLMs into effective tutors.
  arXiv  Detail & Related papers  (2025-05-21T15:00:07Z)
• Training LLM-based Tutors to Improve Student Learning Outcomes in Dialogues [46.60683274479208]
  We introduce an approach to train large language models (LLMs) to generate tutor utterances that maximize the likelihood of student correctness.<n>We show that tutor utterances generated by our model lead to significantly higher chances of correct student responses.
  arXiv  Detail & Related papers  (2025-03-09T03:38:55Z)
• MathTutorBench: A Benchmark for Measuring Open-ended Pedagogical Capabilities of LLM Tutors [82.91830877219822]
  We present MathTutorBench, an open-source benchmark for holistic tutoring model evaluation.<n>MathTutorBench contains datasets and metrics that broadly cover tutor abilities as defined by learning sciences research in dialog-based teaching.<n>We evaluate a wide set of closed- and open-weight models and find that subject expertise, indicated by solving ability, does not immediately translate to good teaching.
  arXiv  Detail & Related papers  (2025-02-26T08:43:47Z)
• Evaluating and Optimizing Educational Content with Large Language Model Judgments [52.33701672559594]
  We use Language Models (LMs) as educational experts to assess the impact of various instructions on learning outcomes. We introduce an instruction optimization approach in which one LM generates instructional materials using the judgments of another LM as a reward function. Human teachers' evaluations of these LM-generated worksheets show a significant alignment between the LM judgments and human teacher preferences.
  arXiv  Detail & Related papers  (2024-03-05T09:09:15Z)
• Improving the Validity of Automatically Generated Feedback via Reinforcement Learning [46.667783153759636]
  We propose a framework for feedback generation that optimize both correctness and alignment using reinforcement learning (RL)<n>Specifically, we use GPT-4's annotations to create preferences over feedback pairs in an augmented dataset for training via direct preference optimization (DPO)
  arXiv  Detail & Related papers  (2024-03-02T20:25:50Z)
• MathDial: A Dialogue Tutoring Dataset with Rich Pedagogical Properties Grounded in Math Reasoning Problems [74.73881579517055]
  We propose a framework to generate such dialogues by pairing human teachers with a Large Language Model prompted to represent common student errors. We describe how we use this framework to collect MathDial, a dataset of 3k one-to-one teacher-student tutoring dialogues.
  arXiv  Detail & Related papers  (2023-05-23T21:44:56Z)
• Opportunities and Challenges in Neural Dialog Tutoring [54.07241332881601]
  We rigorously analyze various generative language models on two dialog tutoring datasets for language learning. We find that although current approaches can model tutoring in constrained learning scenarios, they perform poorly in less constrained scenarios. Our human quality evaluation shows that both models and ground-truth annotations exhibit low performance in terms of equitable tutoring.
  arXiv  Detail & Related papers  (2023-01-24T11:00:17Z)
• Enhancing Dialogue Generation via Multi-Level Contrastive Learning [57.005432249952406]
  We propose a multi-level contrastive learning paradigm to model the fine-grained quality of the responses with respect to the query. A Rank-aware (RC) network is designed to construct the multi-level contrastive optimization objectives. We build a Knowledge Inference (KI) component to capture the keyword knowledge from the reference during training and exploit such information to encourage the generation of informative words.
  arXiv  Detail & Related papers  (2020-09-19T02:41:04Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.