@inproceedings{chen-2025-pingan,
    title = "pingan-team at {S}em{E}val-2025 Task 2: {L}o{RA}-Augmented Qwen2.5 with {W}ikidata-Driven Entity Translation",
    author = "Chen, Diyang",
    editor = "Rosenthal, Sara  and
      Ros{\'a}, Aiala  and
      Ghosh, Debanjan  and
      Zampieri, Marcos",
    booktitle = "Proceedings of the 19th International Workshop on Semantic Evaluation (SemEval-2025)",
    month = jul,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/transition-to-people-yaml/2025.semeval-1.268/",
    pages = "2065--2070",
    ISBN = "979-8-89176-273-2",
    abstract = "This paper presents our solution for SemEval-2025 Task 2 on entity-aware machine translation. We propose a parameter-efficient adaptation framework using Low-Rank Adaptation (LoRA) to fine-tune the Qwen2.5-72B model, enabling effective knowledge transfer while preserving generalization capabilities. To address data scarcity and entity ambiguity, we design a Wiki-driven augmentation pipeline that leverages Wikidata{'}s multilingual entity mappings to generate synthetic training pairs. Our system achieves state-of-the-art performance across 10 languages, securing first place in the competition. Experimental results demonstrate significant improvements in both translation quality (COMET) and entity accuracy (M-ETA)."
}