语言模型
90.54%
领先 Gemma 3 27B +4.47
- 01Airun LLM90.54
- 02Gemma 3 27B86.07
- 03DeepSeek-R1 32B74.99
- 04Qwen3 30B69.56
回答准确率,%。吉尔吉斯语任务。
五项评测,五项第一。
对比模型Gemma 3 27BDeepSeek-R1 32BQwen3 30BKani TTSAitil.kgClaude Opus 4.6Aitil.kg TranslateGPT-5.4Grok 4.20
90.54%
领先 Gemma 3 27B +4.47
回答准确率,%。吉尔吉斯语任务。
6.39%
最接近的竞争对手的错误率是我们的 3.2 倍
WER:每 100 个词中的错误数。仅限西里尔字母。
52.9
领先 Claude Opus 4.6 +6.9
BLEU:与参考译文的接近程度。
我们研究 AI 如何理解吉尔吉斯语 以及中亚的其他语言。
发表于MeLLMMRLLoResMTClimateNLParXiv
MeLLM 2026·ACL Anthology2026.mellm-1.5
Zarina Uvalieva, Bektemir Kumarbai Uulu, Adilet Metinov, Tynchtykbek Tashbaltaev, Nurtilek Alibekov
Abstract
Text normalization, the task of converting noisy, informal text into a standardized form - is a fundamental preprocessing step for many NLP applications. Despite the growing need for Kyrgyz language processing tools, to the best of our knowledge, no prior work has addressed automatic text normalization for Kyrgyz, a morphologically rich, low-resource Turkic language. In this paper, we present the first systematic study of Kyrgyz text normalization. We collect a dataset of 1.67 million noisy–clean text pairs sourced from YouTube comments, Instagram posts, and Telegram channels, where users frequently write without punctuation, capitalization, or standard spelling. Pairs were annotated with Gemini 3 Pro; the 1,000-example test set was fully verified by two native Kyrgyz speakers with adjudication, and a random subset of the training data was spot-checked, while the full 1.67M training set was not verified exhaustively. For continual pre-training, we additionally use a 538 MB Kyrgyz corpus compiled from news portals and books. We evaluate five systems: a rule-based baseline, zero-shot mT5, a fine-tuned mT5-small model, a continually pre-trained mT5-small followed by fine-tuning, and zero-shot Gemma 4. Our experiments show that fine-tuned mT5-small achieves a CER of 0.0796, outperforming the rule-based baseline (CER 0.2029), zero-shot mT5 (CER 0.9887), and zero-shot Gemma 4 (CER 0.1620), a roughly 32× larger model in a fine-tuned vs. zero-shot setting. Human evaluation by two native Kyrgyz speakers confirms these results, with fine-tuned mT5-small rated as correct in 99.8% of cases. We further analyze why continual pre-training with span corruption does not improve over direct fine-tuning, finding hallucination in 35/40 of the inspected failure cases (87.5%, 95% Wilson CI [74%, 95%]).
Proceedings of the 1st Workshop on Multilinguality in the Era of Large Language Models (MeLLM 2026), pages 52–62July 2026 ©2026 Association for Computational Linguistics52
首个系统性研究,基于 167 万对社交媒体文本: 神经网络的错误比规则方法少 2.5 倍。
出行、购物、文档和法律: 城市生活所需的一切。
Pro仅限 Airun Pro 订阅
全国都已接入 Airun
名单中银行 2政府 3教育 8企业 9
银行
Bakai Bank
银行
Берекет Банк
政府
Жогорку Кенеш
政府
Министерство юстиции
政府
Мэрия Бишкека
教育
Университет Манас
教育
Университет Ала-Тоо
教育
КГМА
教育
КРСУ
教育
МУК
教育
AUCA
教育
Compass College
教育
Nomad College
企业
mTicket.kg
企业
Etnomedia
企业
1.1 studio
企业
OnClinic
企业
ArtWin
企业
Devolution
企业
Profile.kg
企业
Bakai Store
企业
Bakai Travel
报道过我们的媒体
The Times of Central AsiaSputnik КыргызстанToday.kgEconomist.kgBanks.kgReporter.kgKaktus.mediaTHE TECHАкчабарРамблер/новости
自己用:在浏览器和手机上。 做产品:通过 API 调用 Airun 模型。