@inproceedings{boulanger-lavergne-and-sophie-rosset-2022-generating,
    title = "Generating unlabelled data for a tri-training approach in a low resourced {NER} task",
    author = "Boulanger, Hugo  and
      Lavergne, Thomas  and
      Rosset, Sophie",
    editor = "Cherry, Colin  and
      Fan, Angela  and
      Foster, George  and
      Haffari, Gholamreza (Reza)  and
      Khadivi, Shahram  and
      Peng, Nanyun (Violet)  and
      Ren, Xiang  and
      Shareghi, Ehsan  and
      Swayamdipta, Swabha",
    booktitle = "Proceedings of the Third Workshop on Deep Learning for Low-Resource Natural Language Processing",
    month = jul,
    year = "2022",
    address = "Hybrid",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-emnlp/2022.deeplo-1.4/",
    doi = "10.18653/v1/2022.deeplo-1.4",
    pages = "30--37",
    abstract = "Training a tagger for Named Entity Recognition (NER) requires a substantial amount of labeled data in the task domain. Manual labeling is a tedious and complicated task. Semisupervised learning methods can reduce the quantity of labeled data necessary to train a model. However, these methods require large quantities of unlabeled data, which remains an issue in many cases.We address this problem by generating unlabeled data. Large language models have proven to be powerful tools for text generation. We use their generative capacity to produce new sentences and variations of the sentences of our available data. This generation method, combined with a semi-supervised method, is evaluated on CoNLL and I2B2. We prepare both of these corpora to simulate a low resource setting. We obtain significant improvements for semisupervised learning with synthetic data against supervised learning on natural data."
}