@inproceedings{boscher-etal-2025-enhancing,
    title = "Enhancing Low-Resource Text Classification with {LLM}-Generated Corpora : A Case Study on Olfactory Reference Extraction",
    author = {Boscher, C{\'e}dric  and
      Bruderer, Shannon  and
      Largeron, Christine  and
      Eglin, V{\'e}ronique  and
      Egyed-Zsigmond, El{\"o}d},
    editor = "Inui, Kentaro  and
      Sakti, Sakriani  and
      Wang, Haofen  and
      Wong, Derek F.  and
      Bhattacharyya, Pushpak  and
      Banerjee, Biplab  and
      Ekbal, Asif  and
      Chakraborty, Tanmoy  and
      Singh, Dhirendra Pratap",
    booktitle = "Proceedings of the 14th International Joint Conference on Natural Language Processing and the 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics",
    month = dec,
    year = "2025",
    address = "Mumbai, India",
    publisher = "The Asian Federation of Natural Language Processing and The Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-ijcnlp-aacl/2025.ijcnlp-long.161/",
    pages = "3004--3027",
    ISBN = "979-8-89176-298-5",
    abstract = "Extracting sensory information from text, particularly olfactory references, is challenging due to limited annotated datasets and the implicit, subjective nature of sensory experiences. This study investigates whether GPT-4o-generated data can complement or replace human annotations. We evaluate human- and LLM-labeled corpora on two tasks: coarse-grained detection of olfactory content and fine-grained sensory term extraction. Despite lexical variation, generated texts align well with real data in semantic and sensorimotor embedding spaces. Models trained on synthetic data perform strongly, especially in low-resource settings. Human annotations offer better recall by capturing implicit and diverse aspects of sensoriality, while GPT-4o annotations show higher precision through clearer pattern alignment. Data augmentation experiments confirm the utility of synthetic data, though trade-offs remain between label consistency and lexical diversity. These findings support using synthetic data to enhance sensory information mining when annotated data is limited."
}