@inproceedings{rohan-etal-2025-generative,
    title = "Generative Data Augmentation for Improving Semantic Classification",
    author = "Rohan, Shadman  and
      Akhter, Mahmud Elahi  and
      Moosa, Ibraheem Muhammad  and
      Mohammed, Nabeel  and
      Ali, Amin Ahsan  and
      Rahman, Akmmahbubur",
    editor = "Alam, Firoj  and
      Kar, Sudipta  and
      Chowdhury, Shammur Absar  and
      Hassan, Naeemul  and
      Prince, Enamul Hoque  and
      Tasnim, Mohiuddin  and
      Rony, Md Rashad Al Hasan  and
      Rahman, Md Tahmid Rahman",
    booktitle = "Proceedings of the Second Workshop on Bangla Language Processing (BLP-2025)",
    month = dec,
    year = "2025",
    address = "Mumbai, India",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-ijcnlp-aacl/2025.banglalp-1.28/",
    pages = "347--356",
    ISBN = "979-8-89176-314-2",
    abstract = "We study sentence-level generative data augmentation for Bangla semantic classification across four public datasets and three pretrained model families (BanglaBERT, XLM-Indic, mBERT). We evaluate two widely used, reproducible techniques{---}paraphrasing (mT5-based) and round-trip backtranslation (Bn{--}En{--}Bn){---}and analyze their impact under realistic class imbalance. Overall, augmentation often helps, but gains are tightly coupled to label quality: paraphrasing typically outperforms backtranslation and yields the most consistent improvements for the monolingual model, whereas multilingual encoders benefit less and can be more sensitive to noisy minority-class expansions. A key empirical observation is that the neutral class appears to be a major source of annotation noise, which degrades decision boundaries and can cap the benefits of augmentation even when positive/negative classes are clean and polarized. We provide practical guidance for Bangla sentiment pipelines: (i) use simple sentence-level augmentation to rebalance classes when labels are reliable; (ii) allocate additional curation and higher inter-annotator agreement targets to the neutral class. Our results indicate when augmentation helps and suggest that data quality{---}not model choice alone{---}can become the limiting factor."
}