@inproceedings{wu-etal-2026-cyut,
    title = "{CYUT} at {S}em{E}val-2026 Task 3: Multi-Task Dimensional Aspect Sentiment Regression with Polar Multi-Zone Labeling in {VA} Space",
    author = "Wu, Shih-Hung  and
      Chen, Xian-Yan  and
      Jian, Yi-Min",
    editor = "Kochmar, Ekaterina  and
      Ghosh, Debanjan  and
      North, Kai  and
      Komachi, Mamoru",
    booktitle = "Proceedings of the 20th {I}nternational {W}orkshop on {S}emantic {E}valuation (2026)",
    month = jul,
    year = "2026",
    address = "San Diego, California, USA",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl-workshops/2026.semeval-1.23/",
    pages = "153--159",
    ISBN = "979-8-89176-414-9",
    abstract = "This paper describes CYUT{'}s system for SemEval-2026 Task{\textasciitilde}3 Track{\textasciitilde}B, a multilingual aspect-based dimensional sentiment regression task. We formulate the task as continuous Valence{--}Arousal (VA) prediction and adopt a multi-task learning (MTL) framework with auxiliary tasks automatically derived from gold VA annotations, including polarity, intensity, and quadrant classification. However, these coarse-grained labels may still suffer from regional imbalance in the VA space, leaving some regions with insufficient auxiliary supervision. To address this issue, we extend the system with Polar Multi-Zone Labeling (PMZL) and use its seven-zone variant, PMZL-7. PMZL-7 partitions the VA plane into one core neutral region and six non-central zones based on the directional distribution of non-central samples. This design reduces the risk of auxiliary-label imbalance while supplementing directional information that conventional auxiliary tasks cannot directly capture. We evaluate XLM-R and two generative pretrained models. Results show that PMZL-7 is strongly model-dependent: it provides more stable improvements for Qwen2 and Ministral, while its effect on XLM-R is less consistent. On the official test set, our system achieves the best performance on the NigerianPidgin subset among all participating systems."
}