@inproceedings{gaido-etal-2025-findings,
    title = "Findings of the {WMT} 2025 Shared Task on Model Compression: Early Insights on Compressing {LLM}s for Machine Translation",
    author = "Gaido, Marco  and
      Grundkiewicz, Roman  and
      Gowda, Thamme  and
      Negri, Matteo",
    editor = "Haddow, Barry  and
      Kocmi, Tom  and
      Koehn, Philipp  and
      Monz, Christof",
    booktitle = "Proceedings of the Tenth Conference on Machine Translation",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-emnlp/2025.wmt-1.25/",
    pages = "484--494",
    ISBN = "979-8-89176-341-8",
    abstract = "We present the results of the first edition of the Model Compression shared task, organized as part of the 10th Conference on Machine Translation (WMT25). The task challenged participants to compress Large Language Models (LLMs) toward enabling practical deployment in resource-constrained scenarios, while minimizing loss in translation performance. In this edition, participants could choose to compete in either a constrained track, which required compressing a specific model (Aya Expanse 8B) evaluated on a limited set of language pairs (Czech{\textrightarrow}German, Japanese{\textrightarrow}Chinese, and English{\textrightarrow}Arabic), or an unconstrained track, which placed no restrictions on the model and allowed submissions for any of the 15 language directions covered by the General MT task (GenMT). We received 12 submissions from three teams, all in the constrained track. They proposed different compression solutions and covered various language combinations: all targeted Czech{\textrightarrow}German, while one covered all the language pairs. Evaluation was conducted separately for each language, measuring translation quality using COMET and MetricX, model size, and inference speed on an Nvidia A100 GPU."
}