Saved in:
Bibliographic Details
Main Authors: Fujita, Felipe, Takada, Hideyuki
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2511.12109
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917083183316992
author Fujita, Felipe
Takada, Hideyuki
author_facet Fujita, Felipe
Takada, Hideyuki
contents In this paper, we explore the effectiveness of combining fine-tuning and backtranslation on a small Japanese corpus for neural machine translation. Starting from a baseline English{\textrightarrow}Japanese model (COMET = 0.460), we first apply backtranslation (BT) using synthetic data generated from monolingual Japanese corpora, yielding a modest increase (COMET = 0.468). Next, we fine-tune (FT) the model on a genuine small parallel dataset drawn from diverse Japanese news and literary corpora, achieving a substantial jump to COMET = 0.589 when using Mistral 7B. Finally, we integrate both backtranslation and fine-tuning{ -- }first augmenting the small dataset with BT generated examples, then adapting via FT{ -- }which further boosts performance to COMET = 0.597. These results demonstrate that, even with limited training data, the synergistic use of backtranslation and targeted fine-tuning on Japanese corpora can significantly enhance translation quality, outperforming each technique in isolation. This approach offers a lightweight yet powerful strategy for improving low-resource language pairs.
format Preprint
id arxiv_https___arxiv_org_abs_2511_12109
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Exploring Parameter-Efficient Fine-Tuning and Backtranslation for the WMT 25 General Translation Task
Fujita, Felipe
Takada, Hideyuki
Computation and Language
In this paper, we explore the effectiveness of combining fine-tuning and backtranslation on a small Japanese corpus for neural machine translation. Starting from a baseline English{\textrightarrow}Japanese model (COMET = 0.460), we first apply backtranslation (BT) using synthetic data generated from monolingual Japanese corpora, yielding a modest increase (COMET = 0.468). Next, we fine-tune (FT) the model on a genuine small parallel dataset drawn from diverse Japanese news and literary corpora, achieving a substantial jump to COMET = 0.589 when using Mistral 7B. Finally, we integrate both backtranslation and fine-tuning{ -- }first augmenting the small dataset with BT generated examples, then adapting via FT{ -- }which further boosts performance to COMET = 0.597. These results demonstrate that, even with limited training data, the synergistic use of backtranslation and targeted fine-tuning on Japanese corpora can significantly enhance translation quality, outperforming each technique in isolation. This approach offers a lightweight yet powerful strategy for improving low-resource language pairs.
title Exploring Parameter-Efficient Fine-Tuning and Backtranslation for the WMT 25 General Translation Task
topic Computation and Language
url https://arxiv.org/abs/2511.12109