@inproceedings{zhong-etal-2025-odda,
    title = "{ODDA}: An {OODA}-Driven Diverse Data Augmentation Framework for Low-Resource Relation Extraction",
    author = "Zhong, Yijie  and
      Gao, Yunfan  and
      Zhang, Xiaolian  and
      Wang, Haofen",
    editor = "Che, Wanxiang  and
      Nabende, Joyce  and
      Shutova, Ekaterina  and
      Pilehvar, Mohammad Taher",
    booktitle = "Findings of the Association for Computational Linguistics: ACL 2025",
    month = jul,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/landing_page/2025.findings-acl.15/",
    pages = "267--285",
    ISBN = "979-8-89176-256-5",
    abstract = "Data Augmentation (DA) has emerged as a promising solution to address the scarcity of high-quality annotated data in low-resource relation extraction (LRE). Leveraging large language models (LLMs), DA has significantly improved the performance of RE models with considerably fewer parameters. However, existing DA methods struggle with diversity misalignments, as they neglect the diversity required by the model and generate homogeneous augmentations that do not cover the inter-sample and inter-relation variability, leading to suboptimal performance. Inspired by the Observe-Orient-Decide-Act (OODA) framework, which provides a robust theoretical foundation for iterative decision-making under dynamic conditions, we propose an OODA-driven Diverse DA method (ODDA), guiding the data generation and selection process. DDA first $\mathrm{observes}$ the RE model{'}s behavior to select effective demonstrations for LLMs. Next, it $\mathrm{orients}$ LLMs towards generating diverse data by replacing schema constraints with attribute constraints. Then ODDA $\mathrm{decides}$ on the final augmented dataset with overall diversity from a global search and finally $\mathrm{acts}$ to train the RE model. Extensive experiments on three widely-used benchmarks demonstrate that ODDA consistently outperforms state-of-the-art baselines, achieving average F1 improvements of 3.1{\%} across various LRE scenarios while maintaining enhanced model stability."
}