Sejoon Kim


2024

pdf bib
Efficient Terminology Integration for LLM-based Translation in Specialized Domains
Sejoon Kim | Mingi Sung | Jeonghwan Lee | Hyunkuk Lim | Jorge Gimenez Perez
Proceedings of the Ninth Conference on Machine Translation

Traditional machine translation methods typically involve training models directly on large parallel corpora, with limited emphasis on specialized terminology. However, In specialized fields such as patents, finance, biomedical domains, terminology is crucial for translation, with many terminologies that should not be translated based on semantics of the sentence but should be translated following agreed-upon conventions. In this paper we introduce a methodology that efficiently trains models with a smaller amount of data while preserving the accuracy of terminology translation. The terminology extraction model generates a glossary from existing training datasets and further refines the LLM by instructing it to effectively incorporate these terms into translations. We achieve this through a systematic process of term extraction and glossary creation using the Trie Tree algorithm, followed by data reconstruction to teach the LLM how to integrate these specialized terms. This methodology enhances the model’s ability to handle specialized terminology and ensures high-quality translations, particularly in fields where term consistency is crucial. Our approach has demonstrated exceptional performance, achieving the highest translation score among participants in the WMT patent task to date, showcasing its effectiveness and broad applicability in specialized translation domains where general methods often fall short.

pdf bib
Context-Aware LLM Translation System Using Conversation Summarization and Dialogue History
Mingi Sung | Seungmin Lee | Jiwon Kim | Sejoon Kim
Proceedings of the Ninth Conference on Machine Translation

Translating conversational text, particularly in customer support contexts, presents unique challenges due to its informal and unstructured nature. We propose a context-aware LLM translation system that leverages conversation summarization and dialogue history to enhance translation quality for the English-Korean language pair. Our approach incorporates the two most recent dialogues as raw data and a summary of earlier conversations to manage context length effectively. We demonstrate that this method significantly improves translation accuracy, maintaining coherence and consistency across conversations. This system offers a practical solution for customer support translation tasks, addressing the complexities of conversational text.