@inproceedings{oh-etal-2025-incorporating,
    title = "Incorporating Domain Knowledge into Materials Tokenization",
    author = "Oh, Yerim  and
      Park, Jun-Hyung  and
      Kim, Junho  and
      Kim, SungHo  and
      Lee, SangKeun",
    editor = "Che, Wanxiang  and
      Nabende, Joyce  and
      Shutova, Ekaterina  and
      Pilehvar, Mohammad Taher",
    booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingestion-acl-25/2025.acl-long.474/",
    pages = "9623--9644",
    ISBN = "979-8-89176-251-0",
    abstract = "While language models are increasingly utilized in materials science, typical models rely on frequency-centric tokenization methods originally developed for natural language processing. However, these methods frequently produce excessive fragmentation and semantic loss, failing to maintain the structural and semantic integrity of material concepts. To address this issue, we propose MATTER, a novel tokenization approach that integrates material knowledge into tokenization. Based on MatDetector trained on our materials knowledge base and re-ranking method prioritizing material terms in token merging, MATTER maintains the structural integrity of identified materials concepts and prevents fragmentation during tokenization, ensuring their semantic meaning remains intact. The experimental results demonstrate that MATTER outperforms existing tokenization methods, achieving an average performance gain of 4{\%} and 2{\%} in the generation and classification tasks, respectively. These results underscore the importance of domain knowledge for tokenization strategies in scientific text processing."
}