arXiv 2604.209962026-05-26cs.CL

AFRILANGTUTOR：利用大语言模型推进低资源语言的语言辅导与文化教育

AFRILANGTUTOR: Advancing Language Tutoring and Culture Education in Low-Resource Languages with Large Language Models

Instituto Politécnico Nacional（墨西哥政治技术学院）；University of South Florida（佛罗里达州立大学）；Saarland University（萨尔兰大学）；Imperial College London（伦敦帝国理工学院）；University of Hamburg（汉堡大学）

Tadesse Destaw Belay, Shahriar Kabir Nahin, Israel Abebe Azime, Ocean Monjur, Marek Rei, Chris Biemann, Shamsuddeen Hassan Muhammad, Seid Muhie Yimam, Anshuman Chhabra

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对低资源语言缺乏训练数据的问题，提出AFRILANGDICT词典资源并构建AFRILANGEDU数据集，通过监督微调和直接偏好优化训练AFRILANGTUTOR模型，在10种非洲语言上显著提升辅导性能。

摘要：

如何为缺乏足够训练资源的语言开发语言学习系统？这一挑战日益被非洲大陆的开发者所面临，他们旨在构建能够理解并用当地语言回应的AI系统。为弥补这一差距，我们引入AFRILANGDICT，一个包含19.47万条非洲语言-英语词典条目的集合，作为生成语言学习材料的种子资源，使我们能够自动构建大规模、多样且可验证的学生-导师问答交互，适用于训练AI辅助语言导师。利用AFRILANGDICT，我们构建了AFRILANGEDU，一个包含7.89万个多轮训练示例的数据集，用于监督微调（SFT）和直接偏好优化（DPO）。使用AFRILANGEDU，我们训练了统称为AFRILANGTUTOR的语言辅导模型。我们在AFRILANGEDU上对两个多语言LLM：Llama-3-8B-IT和Gemma-3-12B-IT进行了微调，覆盖10种非洲语言，并评估了它们的性能。结果表明，在AFRILANGEDU上训练的模型始终优于其基础版本，且结合SFT和DPO带来了显著改进，在LLM作为评判者的评估中，四项指标的提升范围从1.8%到15.5%。为促进低资源语言的进一步研究，所有资源均可在https://huggingface.co/afrilang-edu获取。

英文摘要：

How can language learning systems be developed for languages that lack sufficient training resources? This challenge is increasingly faced by developers across the African continent who aim to build AI systems capable of understanding and responding in local languages. To address this gap, we introduce AFRILANGDICT, a collection of 194.7K African language-English dictionary entries designed as seed resources for generating language-learning materials, enabling us to automatically construct large-scale, diverse, and verifiable student-tutor question-answer interactions suitable for training AI-assisted language tutors. Using AFRILANGDICT, we build AFRILANGEDU, a dataset of 78.9K multi-turn training examples for Supervised Fine-Tuning (SFT) and Direct Preference Optimization (DPO). Using AFRILANGEDU, we train language tutoring models collectively referred to as AFRILANGTUTOR. We fine-tune two multilingual LLMs: Llama-3-8B-IT and Gemma-3-12B-IT on AFRILANGEDU across 10 African languages and evaluate their performance. Our results show that models trained on AFRILANGEDU consistently outperform their base counterparts, and combining SFT and DPO yields substantial improvements, with gains ranging from 1.8% to 15.5% under LLM-as-a-judge evaluations across four criteria. To facilitate further research on low-resource languages, all resources are available at https://huggingface.co/afrilang-edu.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。