arXiv 2411.112952024-11-18cs.CLcs.AI

跨越语言边界：利用LLMs进行低资源语言翻译

Transcending Language Boundaries: Harnessing LLMs for Low-Resource Language Translation

School of Computing, The University of Georgia（佐治亚大学计算学院）；Second Language Acquisition and Teaching, University of Arizona（亚利桑那大学第二语言习得与教学系）；University of Alberta（阿尔伯塔大学）；Department of History, The University of Georgia（佐治亚大学历史系）；National GENIUS Center（国家GENIUS中心）；AI4STEM Education Center, University of Georgia（佐治亚大学AI4STEM教育中心）；Department of Mathematics, Science, and Social Studies Education, University of Georgia（佐治亚大学数学、科学与社会研究教育系）

Peng Shu, Junhao Chen, Zhengliang Liu, Hui Wang, Zihao Wu, Tianyang Zhong, Yiwei Li, Huaqin Zhao, Hanqi Jiang, Yi Pan, Yifan Zhou, Constance Owl, Xiaoming Zhai,… 展开作者

Peng Shu, Junhao Chen, Zhengliang Liu, Hui Wang, Zihao Wu, Tianyang Zhong, Yiwei Li, Huaqin Zhao, Hanqi Jiang, Yi Pan, Yifan Zhou, Constance Owl, Xiaoming Zhai, Ninghao Liu, Claudio Saunt, Tianming Liu

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出一种基于检索的方法，通过聚焦关键术语并从现有数据检索示例，有效提升了LLMs将英语翻译成Cherokee、Tibetan和Manchu等低资源语言的单词级准确率和语义理解。

摘要：

大型语言模型（LLMs）在广泛的任务和领域中展现了卓越的成功。然而，它们在低资源语言翻译中的表现，尤其是翻译成这些语言时，仍未被充分探索。这一空白带来了重大挑战，因为语言障碍阻碍了少数民族社区的文化保护与发展。为解决这一问题，本文提出了一种新颖的基于检索的方法，通过聚焦关键术语来提升低资源语言的翻译质量，这涉及翻译关键词并从现有数据中检索相应的示例。为评估该方法的有效性，我们进行了从英语翻译成三种低资源语言的实验：Cherokee（切罗基语），一种极度濒危的北美原住民语言；Tibetan（藏语），一种在亚洲具有历史和文化意义的语言；以及Manchu（满语），一种仅存少数使用者的语言。我们与GPT-4o和LLaMA 3.1 405B的零样本表现进行了比较，突显了这些模型在翻译成低资源语言时面临的重大挑战。相比之下，我们基于检索的方法通过更有效地利用现有资源，在提升单词级准确率和整体语义理解方面显示出了前景。

英文摘要：

Large Language Models (LLMs) have demonstrated remarkable success across a wide range of tasks and domains. However, their performance in low-resource language translation, particularly when translating into these languages, remains underexplored. This gap poses significant challenges, as linguistic barriers hinder the cultural preservation and development of minority communities. To address this issue, this paper introduces a novel retrieval-based method that enhances translation quality for low-resource languages by focusing on key terms, which involves translating keywords and retrieving corresponding examples from existing data. To evaluate the effectiveness of this method, we conducted experiments translating from English into three low-resource languages: Cherokee, a critically endangered indigenous language of North America; Tibetan, a historically and culturally significant language in Asia; and Manchu, a language with few remaining speakers. Our comparison with the zero-shot performance of GPT-4o and LLaMA 3.1 405B, highlights the significant challenges these models face when translating into low-resource languages. In contrast, our retrieval-based method shows promise in improving both word-level accuracy and overall semantic understanding by leveraging existing resources more effectively.

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。