Translating knowledge-intensive and entity-rich text between English and Korean requires transcreation to preserve language-specific and cultural nuances beyond literal, phonetic or word-for-word conversion. We evaluate 13 models (LLMs and MT models) using automatic metrics and human assessment by bilingual annotators. Our findings show LLMs outperform traditional MT systems but struggle with entity translation requiring cultural adaptation. By constructing an error taxonomy, we identify incorrect responses and entity name errors as key issues, with performance varying by entity type and popularity level. This work exposes gaps in automatic evaluation metrics and hope to enable future work in completing culturally-nuanced machine translation.
View on arXiv@article{lee2025_2504.20451,
title={ Team ACK at SemEval-2025 Task 2: Beyond Word-for-Word Machine Translation for English-Korean Pairs },
author={ Daniel Lee and Harsh Sharma and Jieun Han and Sunny Jeong and Alice Oh and Vered Shwartz },
journal={arXiv preprint arXiv:2504.20451},
year={ 2025 }
}