Exploiting Domain-Specific Parallel Data on Multilingual Language Models for Low-resource Language Translation
Year: 2024
Type: preprint
Abstract: Neural Machine Translation (NMT) systems built on multilingual sequence-to-sequence Language Models (msLMs) fail to deliver expected results when the amount of parallel data for a language, as well as the language's representation in the model are limited. This restricts the capabilities of domain-specific NMT systems for low-resource languages (LRLs). As a solution, parallel data from auxiliary domains can be used either to fine-tune or to further pre-train the msLM. We present an evaluation of the effectiveness of these two techniques in the context of domain-specific LRL-NMT. We also explore the impact of domain divergence on NMT model performance. We recommend several strategies for utilizing auxiliary parallel data in building domain-specific NMT models for LRLs.
Source: arXiv (Cornell University)
Authors: Ranathungaa, Surangika, Nayak, Shravan, Huang, Shih-Ting Cindy, Mao, Yanke, Su, Tong, Chan, Yun-Hsiang Ray, Yuan, Songchen, Rinaldi, Anthony, Lee, Annie En-Shiun
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4405903293