arXiv：cs.AI· Junior Cedric Tonga, KV Aditya Srivatsa, Kaushal Kumar Maurya, Fajri Koto, Ekaterina Kochmar·· 16 小时前
LLM 多语言数学辅导模拟研究：英语提示通常更受偏好
Simulating Student-Tutor Interactions with LLMs for Multilingual Math Feedback
AI 导读
研究者用 LLM 模拟多语言学生-导师互动，由强模型扮演导师生成提示、弱模型扮演学生，覆盖 11 种语言、4 个 SOTA LLM 和 352 组实验设置，考察学生输入语言、导师反馈语言、模型选择与语言资源水平如何共同影响学习效果。结果显示英语提示普遍更受偏好，即便对多语言学生 LLM 也是如此，仅少数例外能带来更好的学习表现。
正文
Abstract:Large language models (LLMs) have demonstrated the ability to generate formative feedback and instructional hints in English, making them increasingly relevant for AI-assisted education. However, their ability to provide effective instructional support across different languages, especially for mathematically grounded reasoning tasks, remains largely unexamined. In this work, we present the first large-scale simulation of multilingual student-tutor interactions using LLMs. A stronger model plays the role of the tutor, generating feedback in the form of hints, while a weaker model simulates the student. We explore 352 experimental settings across 11 typologically diverse languages, four state-of-the-art LLMs, and multiple prompting strategies to assess whether language-specific feedback leads to measurable learning gains. Our study examines how student input language, tutor feedback language, model choice, and language resource level jointly influence performance. Results show that English hints are generally preferred, even by multilingual student LLMs, with few exceptions that lead to improved student learning outcomes. These findings offer practical insights for developing multilingual, LLM-based educational tools that are both effective and inclusive.
| Comments: | MathNLP EMNLP 2026 |
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2506.04920 [cs.CL] |
| (or arXiv:2506.04920v2 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2506.04920 arXiv-issued DOI via DataCite |
Submission history
From: Junior Cedric Tonga [view email]
[v1]
Thu, 5 Jun 2025 11:53:04 UTC (553 KB)
[v2]
Sat, 10 Oct 2026 10:01:20 UTC (540 KB)
来源：arXiv：cs.AI · arxiv.org