@inproceedings{tan-etal-2025-lionguard,
    title = "{L}ion{G}uard 2: Building Lightweight, Data-Efficient {\&} Localised Multilingual Content Moderators",
    author = "Tan, Leanne  and
      Chua, Gabriel  and
      Ge, Ziyu  and
      Lee, Roy Ka-Wei",
    editor = {Habernal, Ivan  and
      Schulam, Peter  and
      Tiedemann, J{\"o}rg},
    booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing: System Demonstrations",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-emnlp/2025.emnlp-demos.20/",
    pages = "264--285",
    ISBN = "979-8-89176-334-0",
    abstract = "Modern moderation systems increasingly support multiple languages, but often fail to address localisation and low-resource variants{---}creating safety gaps in real-world deployments. Small models offer a potential alternative to large LLMs, yet still demand considerable data and compute. We present LionGuard 2, a lightweight, multilingual moderation classifier tailored to the Singapore context, supporting English, Chinese, Malay, and partial Tamil. Built on pre-trained OpenAI embeddings and a multi-head ordinal classifier, LionGuard 2 outperforms several commercial and open-source systems across 17 benchmarks, including both Singapore-specific and public English datasets. The system is actively deployed within the Singapore Government, demonstrating practical efficacy at scale. Our findings show that high-quality local data and robust multilingual embeddings can achieve strong moderation performance, without fine-tuning large models. We release our model weights and part of our training data to support future work on LLM safety."
}