@inproceedings{luong-etal-2025-vinumfcr,
    title = "{V}i{N}um{FCR}: A Novel {V}ietnamese Benchmark for Numerical Reasoning Fact Checking on Social Media News",
    author = "Luong, Nhi Ngoc Phuong  and
      Le, Anh Thi Lan  and
      Huynh, Tin Van  and
      Nguyen, Kiet Van  and
      Nguyen, Ngan",
    editor = "Flek, Lucie  and
      Narayan, Shashi  and
      Phương, L{\^e} Hồng  and
      Pei, Jiahuan",
    booktitle = "Proceedings of the 18th International Natural Language Generation Conference",
    month = oct,
    year = "2025",
    address = "Hanoi, Vietnam",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/author-page-you-zhang-rochester/2025.inlg-main.9/",
    pages = "134--147",
    abstract = "In the digital era, the internet provides rapid and convenient access to vast amounts of information. However, much of this information remains unverified, particularly with the increasing prevalence of falsified numerical data, leading to public confusion and negative societal impacts. To address this issue, we developed ViNumFCR, a first dataset dedicated to fact-checking numerical information in Vietnamese. Comprising over 10,000 samples collected and constructed from online newspaper across 12 different topics. We assessed the performance of various fact-checking models, including Pretrained Language Models and Large Language Models, alongside retrieval techniques for gathering supporting evidence. Experimental results demonstrate that the XLM-R{\_}Large model achieved the highest accuracy of 90.05{\%} on the fact-checking task, while the combined SBERT + BM25 model attained a precision of over 97{\%} on the evidence retrieval task. Additionally, we conducted an in-depth analysis of the linguistic features of the dataset to understand the factors influencing the performance models. The ViNumFCR dataset is publicly available to support further research."
}