arXiv:2512.20780v3 Announce Type: replace 
Abstract: 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.

أظهرت الأبحاث الحديثة أن نماذج اللغة الكبيرة (LLMs) يمكن أن تولد استجابات تعليمية في الرياضيات تقترب من الجودة التربوية للمعلمين البشريين الخبراء، على الرغم من أنها تظهر ملفات تعريف تعليمية ولغوية مميزة. قامت الدراسة بتحليل حوارات بين معلمين خبراء ومبتدئين بالإضافة إلى نماذج LLMs المختلفة، كاشفة أن المعلمين الخبراء يقدمون استجابات ذات جودة أعلى بشكل مستمر.

Investigaciones recientes han demostrado que los modelos de lenguaje grandes (LLMs) pueden generar respuestas de tutoría en matemáticas que se acercan a la calidad pedagógica de los tutores humanos expertos, aunque exhiben perfiles de instrucción y lingüísticos distintos. El estudio analizó diálogos de tutores expertos y novatos junto con varios LLMs, revelando que los tutores expertos proporcionan respuestas de mayor calidad de manera consistente.

Des recherches récentes ont montré que les grands modèles de langage (LLMs) peuvent générer des réponses de tutorat en mathématiques qui approchent la qualité pédagogique des tuteurs humains experts, bien qu'ils présentent des profils d'instruction et linguistiques distincts. L'étude a analysé des dialogues entre des tuteurs experts et novices ainsi que divers LLMs, révélant que les tuteurs experts fournissent systématiquement des réponses de meilleure qualité.

Recent research has shown that large language models (LLMs) can generate tutoring responses in mathematics that approach the pedagogical quality of expert human tutors, although they exhibit distinct instructional and linguistic profiles. The study analyzed dialogues from expert and novice tutors alongside various LLMs, revealing that expert tutors consistently provide higher-quality responses.