Tailoring mt5 for the Uzbek Language: a Compact Model for NLP Applications

Адилова, Ф.Т., Давронов, Р.Р., Кушмуратов, С.И.

Рақамли технологияларнинг назарий ва амалий масалалари · 2023-yil

Annotatsiya

Despite being spoken by nearly 50 million individuals, the Uzbek language remains underrepresented in Natural Language Processing (NLP). One primary reason is the limited availability of Uzbek linguistic resources. With the rising prominence of the Transformer architecture in NLP, it has overtaken earlier methods like convolutional and recurrent neural networks. The T5 (Text-to-Text Transfer Transformer) standardizes linguistic tasks in English by converting them into a text-to-text format. The mT5, its multilingual version, has shown promising outcomes in various NLP tasks spanning multiple languages. However, the considerable dimensions of the mT5 pose challenges for applications focused on a singular language. In our study, we fine-tuned the mT5 specifically for Uzbek, leading to a more compact T5 model. We compared this tailored model's efficiency with the mT5 on Automatic Text Summarization (ATS) and Named Entity Recognition (NER) tasks using identical protocols and datasets. Our adapted model surpassed the performance of the mT5, indicating the feasibility of developing a more compact pre-trained model with nearly half the size, without compromising results. This streamlined model also benefits from reduced memory usage, faster startup, and swifter processing times. For access to this model, please reach out.

Maqola ma’lumotlari
MualliflarАдилова, Ф.Т., Давронов, Р.Р., Кушмуратов, С.И.
JurnalРақамли технологияларнинг назарий ва амалий масалалари
NashriyotТАТУ Самарқанд филиали
ISSN2181-3086, 2181-3094
Nashr sanasi2023-10-02
Jild5
Son3
Betlar7-16
TilIngliz
Ilmiy sohaTexnika
Indeksga qo‘shilgan2026-08-26

Maqolaning asl sahifasi

Kalit so‘zlar

model compression, transformer, pre-trained model, automatic text summarization, named entity recognition, сжатие модели, преобразователь, предварительно обученная модель, автоматическое суммирование текста, распознавание именованных объектов

Ilmiy soha

Texnika

Iqtibos keltirish

APAАдилова, Ф.Т., Давронов, Р.Р., & Кушмуратов, С.И. (2023). Tailoring mt5 for the Uzbek Language: a Compact Model for NLP Applications. Рақамли технологияларнинг назарий ва амалий масалалари, 5(3), 7-16. https://ijdt.uz/index.php/ijdt/article/view/104
GOST (OAK)Адилова, Ф.Т., Давронов, Р.Р., Кушмуратов, С.И. Tailoring mt5 for the Uzbek Language: a Compact Model for NLP Applications // Рақамли технологияларнинг назарий ва амалий масалалари. – 2023. – T. 5, № 3. – P. 7-16.
MLAАдилова, Ф.Т., Давронов, Р.Р., and Кушмуратов, С.И. “Tailoring mt5 for the Uzbek Language: a Compact Model for NLP Applications.” Рақамли технологияларнинг назарий ва амалий масалалари, 5(3), 7-16, 2023. https://ijdt.uz/index.php/ijdt/article/view/104
ChicagoАдилова, Ф.Т., Давронов, Р.Р., and Кушмуратов, С.И. “Tailoring mt5 for the Uzbek Language: a Compact Model for NLP Applications.” Рақамли технологияларнинг назарий ва амалий масалалари, 5(3), 7-16 (2023). https://ijdt.uz/index.php/ijdt/article/view/104
HarvardАдилова, Ф.Т., Давронов, Р.Р., and Кушмуратов, С.И (2023) ‘Tailoring mt5 for the Uzbek Language: a Compact Model for NLP Applications’, Рақамли технологияларнинг назарий ва амалий масалалари, 5(3), 7-16. https://ijdt.uz/index.php/ijdt/article/view/104
BibTeX
@article{article2023,
  title = {Tailoring mt5 for the Uzbek Language: a Compact Model for NLP Applications},
  author = {Адилова, Ф.Т. and Давронов, Р.Р. and Кушмуратов, С.И.},
  journal = {Рақамли технологияларнинг назарий ва амалий масалалари},
  year = {2023},
  volume = {5},
  number = {3},
  pages = {7-16},
  url = {https://ijdt.uz/index.php/ijdt/article/view/104}
}

Boshqa maqola uchun iqtibos tayyorlash

Рақамли технологияларнинг назарий ва амалий масалалари jurnalidan boshqa maqolalar

Рақамли технологияларнинг назарий ва амалий масалалари — barcha maqolalar · 2023-yil arxivi

Адилова, Ф.Т. — boshqa maqolalari

Texnika sohasidagi boshqa maqolalar

Texnika sohasidagi OAK jurnallari