Academic Bibliography

 Search 200 years of publications by Ghent University researchers.
Advanced search
1 file | 585.88 KB
Add to list
  1. Search results
  2. Predicting machine translation perfor...

Predicting machine translation performance on low-resource languages : the role of domain similarity

Eric Khiu, Hasti Toossi, Jinyu Liu, Jiaxu Li, David Anugraha, Juan Flores, Leandro Roman, A. Seza Doğruöz (UGent) and En-Shiun Lee
(2024) FINDINGS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS : EACL 2024. p.1474-1486
Author
Organization
Abstract
Fine-tuning and testing a multilingual large language model is a challenge for low-resource languages (LRLs) since it is an expensive process. While previous studies have predicted the performance of natural language processing (NLP) tasks using machine learning methods, they primarily focus on high-resource languages, overlooking LRLs and shifts across domains. Focusing on LRLs, we investigate three factors (the size of the fine-tuning corpus, domain similarity between fine-tuning and testing corpora, and language similarity between source and target languages), which can potentially impact the model performance by using classical regression models. Our results indicate that domain similarity has the most important impact on predicting the performance of Machine Translation models.

Downloads

  • Download
    2024.findings-eacl.100v2.pdf
    • full text (Published version)
    • |
    • open access
    • |
    • PDF
    • |
    • 585.88 KB

Citation

Please use this url to cite or link to this publication:

MLA
Khiu, Eric, et al. “Predicting Machine Translation Performance on Low-Resource Languages : The Role of Domain Similarity.” FINDINGS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS : EACL 2024, edited by Yvette Graham and Matthew Purver, Association for Computational Linguistics (ACL), 2024, pp. 1474–86.
APA
Khiu, E., Toossi, H., Liu, J., Li, J., Anugraha, D., Flores, J., … Lee, E.-S. (2024). Predicting machine translation performance on low-resource languages : the role of domain similarity. In Y. Graham & M. Purver (Eds.), FINDINGS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS : EACL 2024 (pp. 1474–1486). Association for Computational Linguistics (ACL).
Chicago author-date
Khiu, Eric, Hasti Toossi, Jinyu Liu, Jiaxu Li, David Anugraha, Juan Flores, Leandro Roman, A. Seza Doğruöz, and En-Shiun Lee. 2024. “Predicting Machine Translation Performance on Low-Resource Languages : The Role of Domain Similarity.” In FINDINGS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS : EACL 2024, edited by Yvette Graham and Matthew Purver, 1474–86. Association for Computational Linguistics (ACL).
Chicago author-date (all authors)
Khiu, Eric, Hasti Toossi, Jinyu Liu, Jiaxu Li, David Anugraha, Juan Flores, Leandro Roman, A. Seza Doğruöz, and En-Shiun Lee. 2024. “Predicting Machine Translation Performance on Low-Resource Languages : The Role of Domain Similarity.” In FINDINGS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS : EACL 2024, ed by. Yvette Graham and Matthew Purver, 1474–1486. Association for Computational Linguistics (ACL).
Vancouver
1.
Khiu E, Toossi H, Liu J, Li J, Anugraha D, Flores J, et al. Predicting machine translation performance on low-resource languages : the role of domain similarity. In: Graham Y, Purver M, editors. FINDINGS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS : EACL 2024. Association for Computational Linguistics (ACL); 2024. p. 1474–86.
IEEE
[1]
E. Khiu et al., “Predicting machine translation performance on low-resource languages : the role of domain similarity,” in FINDINGS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS : EACL 2024, St. Julian’s, Malta, 2024, pp. 1474–1486.
@inproceedings{01HT7AS5NTMR82R901RCXVR71H,
  abstract     = {{Fine-tuning and testing a multilingual large language model is a challenge for low-resource languages (LRLs) since it is an expensive process. While previous studies have predicted the performance of natural language processing (NLP) tasks using machine learning methods, they primarily focus on high-resource languages, overlooking LRLs and shifts across domains. Focusing on LRLs, we investigate three factors (the size of the fine-tuning corpus, domain similarity between fine-tuning and testing corpora, and language similarity between source and target languages), which can potentially impact the model performance by using classical regression models. Our results indicate that domain similarity has the most important impact on predicting the performance of Machine Translation models.}},
  author       = {{Khiu, Eric and Toossi, Hasti and Liu, Jinyu and Li, Jiaxu and Anugraha, David and Flores, Juan and Roman, Leandro and Doğruöz, A. Seza and Lee, En-Shiun}},
  booktitle    = {{FINDINGS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS : EACL 2024}},
  editor       = {{Graham, Yvette and Purver, Matthew}},
  isbn         = {{9798891760936}},
  language     = {{eng}},
  location     = {{St. Julian’s, Malta}},
  pages        = {{1474--1486}},
  publisher    = {{Association for Computational Linguistics (ACL)}},
  title        = {{Predicting machine translation performance on low-resource languages : the role of domain similarity}},
  url          = {{https://aclanthology.org/2024.findings-eacl.100}},
  year         = {{2024}},
}
Web of Science
Times cited: