Researchers have introduced ATD-Trans, a new dataset designed to evaluate machine translation quality for Japanese-English travelogues, with a focus on geographic information. The dataset allows for assessment at both overall and geo-entity levels, covering domestic and international regions. Experiments using ATD-Trans revealed that Japanese-enhanced language models perform better and that translating geographic entities within Japanese travel blogs presents a greater challenge. AI
Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →
IMPACT This dataset could improve the accuracy of machine translation for travel-related content, making geographic information more accessible across languages.
RANK_REASON The cluster describes a new academic dataset for evaluating machine translation quality, which falls under research. [lever_c_demoted from research: ic=1 ai=1.0]