@inproceedings{turk-etal-2025-clac,
    title = "{CL}a{C} at {DISRPT} 2025: Hierarchical Adapters for Cross-Framework Multi-lingual Discourse Relation Classification",
    author = "Turk, Nawar  and
      Comitogianni, Daniele  and
      Kosseim, Leila",
    editor = "Braud, Chlo{\'e}  and
      Liu, Yang Janet  and
      Muller, Philippe  and
      Zeldes, Amir  and
      Li, Chuyuan",
    booktitle = "Proceedings of the 4th Shared Task on Discourse Relation Parsing and Treebanking (DISRPT 2025)",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-emnlp/2025.disrpt-1.3/",
    pages = "36--47",
    ISBN = "979-8-89176-344-9",
    abstract = "We present our submission to Task 3 (Discourse Relation Classification) of the DISRPT 2025 shared task. Task 3 introduces a unified set of 17 discourse relation labels across 39 corpora in 16 languages and six discourse frameworks, posing significant multilingual and cross{-}formalism challenges. We first benchmark the task by fine{-}tuning multilingual BERT{-}based models (mBERT, XLM{-}RoBERTa{-}Base, and XLM{-}RoBERTa{-}Large) with two argument{-}ordering strategies and progressive unfreezing ratios to establish strong baselines. We then evaluate prompt{-}based large language models (namely Claude Opus 4.0) in zero{-}shot and few{-}shot settings to understand how LLMs respond to the newly proposed unified labels. Finally, we introduce HiDAC, a Hierarchical Dual{-}Adapter Contrastive learning model. Results show that while larger transformer models achieve higher accuracy, the improvements are modest, and that unfreezing the top 75{\%} of encoder layers yields performance comparable to full fine{-}tuning while training far fewer parameters. Prompt{-}based models lag significantly behind fine{-}tuned transformers, and HiDAC achieves the highest overall accuracy (67.5{\%}) while remaining more parameter{-}efficient than full fine{-}tuning."
}