@inproceedings{gu-etal-2025-llm,
    title = "An {LLM}-based Framework for Domain-Specific Information Extraction: A Case Study in Computer Science and Chemistry",
    author = "Gu, Xungang  and
      Tian, Yangjie  and
      Li, Ning  and
      Liu, Meng  and
      Xu, Ruohua  and
      Zhang, He  and
      Liu, Hanqiu  and
      Sheng, Yongpan  and
      Liu, Ming",
    editor = "Kummerfeld, Jonathan K.  and
      Joshi, Aditya  and
      Dras, Mark",
    booktitle = "Proceedings of The 23rd Annual Workshop of the Australasian Language Technology Association",
    month = nov,
    year = "2025",
    address = "Sydney, Australia",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-alta/2025.alta-main.8/",
    pages = "101--111",
    ISBN = "1834-7037",
    abstract = "Information extraction (IE) in specialized domains like computer science and chemistry is challenged by the poor generalization of traditional models and the knowledge deficits of general-purpose Large Language Models (LLMs). We introduce a robust, LLM-based framework featuring two core contributions: an end-to-end training and inference paradigm that combines continual pre-training (CPT) for knowledge injection, supervised fine-tuning (SFT) for task alignment, and retrieval-augmented generation (RAG) for inference-time enhancement; and a novel LLM-assisted data annotation pipeline for the efficient creation of high-quality training data. Comprehensive experiments demonstrate that while fine-tuning alone yields strong in-domain performance, our complete framework exhibits superior robustness and generalization. It consistently achieves state-of-the-art results in challenging domain-shift and novel-schema scenarios, validating our integrated approach for building adaptable and high-performance domain-specific IE systems."
}