ATD-Trans: A Geographically Grounded Japanese-English Travelogue Translation Dataset

TL;DR

This paper introduces ATD-Trans, a Japanese-English travelogue translation dataset that enables detailed evaluation of machine translation quality at geo-entity levels across regions.

Contribution

It provides a new geographically grounded translation dataset for Japanese-English travel texts, facilitating analysis of MT performance on geo-entities in different regions.

Findings

Japanese-enhanced models perform better in translation tasks.

Translating domestic-region geo-entities is more challenging.

The dataset supports evaluation at both overall and geo-entity levels.

Abstract

Geographic text, or textual data rich in geographic (geo-) information is a valuable source for various geographic applications, e.g., tourism management. Making such information accessible to speakers of other languages further enhances its utility; thus, accurate machine translation (MT) is essential for equity in multilingual geo-information access. To facilitate in-depth analysis for geographic text, we introduce ATD-Trans, a geographically grounded Japanese--English travelogue translation dataset, which enables evaluation of MT quality at both the overall and geo-entity levels across domestic (within Japan) and overseas regions. Our experiments on existing language models examine two factors: model language focus and geographic regions. The results highlight advantages of Japanese-enhanced models and greater difficulty in translating domestic-region geo-entities mentioned in travel blogs.