# Large Language Models Approach Expert Pedagogical Quality in Math Tutoring but Differ in Instructional and Linguistic Profiles

arXiv: https://arxiv.org/abs/2512.20780 · HTML: https://mlsanity.com/paper/2512.20780/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 18 of its 41 references that are in the map.
- First posted on arXiv 23 Dec 2025.

## Builds on (most cited first, 18 of 18)

- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone](https://mlsanity.com/paper/2404.14219/) (2024)
- [MathDial: A Dialogue Tutoring Dataset with Rich Pedagogical Properties Grounded in Math Reasoning Problems](https://mlsanity.com/paper/2305.14536/) (2023)
- [Unifying AI Tutor Evaluation: An Evaluation Taxonomy for Pedagogical Ability Assessment of LLM-Powered AI Tutors](https://mlsanity.com/paper/2412.09416/) (2024)
- [Bridging the Novice-Expert Gap via Models of Decision-Making: A Case Study on Remediating Math Mistakes](https://mlsanity.com/paper/2310.10648/) (2023)
- [Evaluating Mathematical Reasoning of Large Language Models: A Focus on Error Identification and Correction](https://mlsanity.com/paper/2406.00755/) (2024)
- [Detection and Measurement of Syntactic Templates in Generated Text](https://mlsanity.com/paper/2407.00211/) (2024)
- [The TalkMoves Dataset: K-12 Mathematics Lesson Transcripts Annotated for Teacher and Student Discursive Moves](https://mlsanity.com/paper/2204.09652/) (2022)
- [AutoTutor meets Large Language Models: A Language Model Tutor with Rich Pedagogy and Guardrails](https://mlsanity.com/paper/2402.09216/) (2024)
- [TyDiP: A Dataset for Politeness Classification in Nine Typologically Diverse Languages](https://mlsanity.com/paper/2211.16496/) (2022)
- [Training Turn-by-Turn Verifiers for Dialogue Tutoring Agents: The Curious Case of LLMs as Your Coding Tutors](https://mlsanity.com/paper/2502.13311/) (2025)
- [Findings of the BEA 2025 Shared Task on Pedagogical Ability Assessment of AI-powered Tutors](https://mlsanity.com/paper/2507.10579/) (2025)
- [Novice Learner and Expert Tutor: Evaluating Math Reasoning Abilities of Large Language Models with Misconceptions](https://mlsanity.com/paper/2310.02439/) (2023)
- [QUDsim: Quantifying Discourse Similarities in LLM-Generated Text](https://mlsanity.com/paper/2504.09373/) (2025)
- [Using Large Language Models to Assess Tutors' Performance in Reacting to Students Making Math Errors](https://mlsanity.com/paper/2401.03238/) (2024)
- [Linguistic and Embedding-Based Profiling of Texts generated by Humans and Large Language Models](https://mlsanity.com/paper/2507.13614/) (2025)
- [Computational analysis of US Congressional speeches reveals a shift from evidence to intuition](https://mlsanity.com/paper/2405.07323/) (2024)
- [Collective moderation of hate, toxicity, and extremity in online discussions](https://mlsanity.com/paper/2303.00357/) (2023)

## Abstract (arXiv)

Recent work has explored the use of large language models (LLMs) to generate tutoring responses in mathematics, yet it remains unclear how closely their instructional behavior aligns with expert human practice. We analyze a dataset of math remediation dialogues in which expert tutors, novice tutors, and seven LLMs of varying sizes, comprising both open-weight and commercial models, respond to the same student errors. We examine instructional strategies and linguistic characteristics of tutoring responses, including uptake (restating and revoicing), pressing for accuracy and reasoning, lexical diversity, readability, politeness, and agency. We find that expert tutors produce higher-quality responses than novices, and that larger LLMs generally receive higher pedagogical quality ratings than smaller models, approaching expert performance on average. However, LLMs exhibit systematic differences in their instructional profiles: they underuse discursive strategies characteristic of expert tutors while generating longer, more lexically diverse, and more polite responses. Regression analyses show that pressing for accuracy and reasoning, restating and revoicing, and lexical diversity, are positively associated with perceived pedagogical quality, whereas higher levels of agentic and polite language are negatively associated. These findings highlight the importance of analyzing instructional strategies and linguistic characteristics when evaluating tutoring responses across human tutors and intelligent tutoring systems.