ChatGPT as a Translation Engine: A Case Study on Japanese-English

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sutanto, Vincent Michael, De Giacomo, Giovanni Gatti, Nakazawa, Toshiaki, Yamada, Masaru
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918157225033728
author Sutanto, Vincent Michael
De Giacomo, Giovanni Gatti
Nakazawa, Toshiaki
Yamada, Masaru
author_facet Sutanto, Vincent Michael
De Giacomo, Giovanni Gatti
Nakazawa, Toshiaki
Yamada, Masaru
contents This study investigates ChatGPT for Japanese-English translation, exploring simple and enhanced prompts and comparing against commercially available translation engines. Performing both automatic and MQM-based human evaluations, we found that document-level translation outperforms sentence-level translation for ChatGPT. On the other hand, we were not able to determine if enhanced prompts performed better than simple prompts in our experiments. We also discovered that ChatGPT-3.5 was preferred by automatic evaluation, but a tradeoff exists between accuracy (ChatGPT-3.5) and fluency (ChatGPT-4). Lastly, ChatGPT yields competitive results against two widely-known translation systems.
format Preprint
id arxiv_https___arxiv_org_abs_2510_08042
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle ChatGPT as a Translation Engine: A Case Study on Japanese-English
Sutanto, Vincent Michael
De Giacomo, Giovanni Gatti
Nakazawa, Toshiaki
Yamada, Masaru
Computation and Language
This study investigates ChatGPT for Japanese-English translation, exploring simple and enhanced prompts and comparing against commercially available translation engines. Performing both automatic and MQM-based human evaluations, we found that document-level translation outperforms sentence-level translation for ChatGPT. On the other hand, we were not able to determine if enhanced prompts performed better than simple prompts in our experiments. We also discovered that ChatGPT-3.5 was preferred by automatic evaluation, but a tradeoff exists between accuracy (ChatGPT-3.5) and fluency (ChatGPT-4). Lastly, ChatGPT yields competitive results against two widely-known translation systems.
title ChatGPT as a Translation Engine: A Case Study on Japanese-English
topic Computation and Language
url https://arxiv.org/abs/2510.08042