@inproceedings{maharjan-etal-2026-sagarmatha,
    title = "{S}agarmatha at {S}em{E}val-2026 Task 9: Heterogeneous Ensembling and Hierarchical Task Conditioning for Multilingual Latent Distributional Divergence Modeling",
    author = "Maharjan, Sujal  and
      Shrestha, Astha  and
      Shrestha, Pratikshya",
    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-1.382/",
    pages = "3045--3051",
    ISBN = "979-8-89176-414-9",
    abstract = "The digital public square is increasingly fragmented by affective polarization, requiring computational systems capable of identifying discursive strategies such as dehumanization and vilification. This paper presents Sagarmatha, the system developed for SemEval-2026 Task 9. We propose a heterogeneous ensemble architecture that addresses the limitations of standard transformer fine-tuning across 22 languages. Our approach integrates mDeBERTa-v3, ReMBERT, LaBSE, mmBERT, and XLM-RoBERTa, through two primary architectural pillars: learnable weighted layer pooling and hierarchical task conditioning. While our final submission (a broad ensemble, R3) demonstrated high stability on the leaderboard, our primary architectural configuration (Weighted Polyglot, R1) yielded superior performance in complex multi-label tasks. The system ranked 1st globally in English and Hausa manifestation identification, and 1st in Telugu detection (2nd in categorization). All code and resources are available at https://github.com/SUJAL390/SagarmathaatSemevaltask9.git."
}