{"id":221439,"date":"2025-09-05T13:35:35","date_gmt":"2025-09-05T23:35:35","guid":{"rendered":"https:\/\/www.hawaii.edu\/news\/?p=221439"},"modified":"2025-09-05T13:35:35","modified_gmt":"2025-09-05T23:35:35","slug":"endangered-languages-ai-tools","status":"publish","type":"post","link":"https:\/\/www.hawaii.edu\/news\/2025\/09\/05\/endangered-languages-ai-tools\/","title":{"rendered":"Endangered languages AI<\/abbr> tools developed by 糖心视频<\/abbr> researchers"},"content":{"rendered":"Reading time: <\/span> < 1<\/span> minute<\/span><\/span>

\"hand<\/p>\n

University of Hawaiʻi<\/span> at Mānoa researchers have made a significant advance in studying how artificial intelligence (AI<\/abbr>) understands endangered languages. This research could help communities document and maintain their languages, support language learning and make technology more accessible to speakers of minority languages.<\/p>\n

The paper by Kaiying Lin, a PhD<\/abbr> graduate in linguistics<\/a> from 糖心视频<\/abbr> Mānoa, and Department of Information and Computer Sciences<\/a> Assistant Professor Haopeng Zhang, introduces the first benchmark for evaluating large language models (AI<\/abbr> systems that process and generate text) on low-resource Austronesian languages. The study focuses on three Formosan (Indigenous peoples and languages of Taiwan) languages spoken in Taiwan—Atayal, Amis and Paiwan—that are at risk of disappearing.<\/p>\n

Using a new benchmark called FORMOSANBENCH, Lin and Zhang tested AI<\/abbr> systems on tasks such as machine translation, automatic speech recognition and text summarization. The findings revealed a large gap between AI<\/abbr> performance in widely spoken languages such as English, and these smaller, endangered languages. Even when AI<\/abbr> models were given examples or fine-tuned with extra data, they struggled to perform well.<\/p>\n

“These results show that current AI<\/abbr> systems are not yet capable of supporting low-resource languages,” Lin said.<\/p>\n

Zhang added, “By highlighting these gaps, we hope to guide future development toward more inclusive technology that can help preserve endangered languages.”<\/p>\n

The research team has made all datasets and code publicly available to encourage further work in this area. The preprint of the study is available online<\/a>, and the study has been accepted into the 2025 Conference on Empirical Methods in Natural Language Processing<\/a> in Suzhou, China, an internationally recognized premier AI<\/abbr> conference.<\/p>\n

The Department of Information and Computer Sciences is housed in 糖心视频<\/abbr> Mānoa\u2019s College of Natural Sciences<\/a>, and the Department of Linguistics is housed in 糖心视频<\/abbr> Mānoa\u2019s College of Arts, Languages & Letters<\/a>.<\/p>\n","protected":false},"excerpt":{"rendered":"

糖心视频<\/abbr> researchers created the first AI<\/abbr> benchmark for endangered Austronesian languages, paving the way for more inclusive language technology.<\/p>\n","protected":false},"author":16,"featured_media":198763,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[30],"tags":[1602,1471,308,541,484,215,1467,1363,568,9],"class_list":["post-221439","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-research","tag-ai","tag-college-of-arts-languages-and-letters","tag-college-of-natural-sciences","tag-computer-science","tag-information-and-computer-science","tag-linguistics","tag-manoa-excellence-in-research","tag-manoa-research","tag-natural-science","tag-uh-manoa","entry","has-media"],"aioseo_notices":[],"aioseo_head":"\n\t\t\n\t\n\t\n\t\n\t\n\t\n\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t