@inproceedings{hu-etal-2026-taxpraben,
    title = "{T}ax{P}ra{B}en: A Scalable Benchmark for Structured Evaluation of {LLM}s in {C}hinese Real-World Tax Practice",
    author = "Hu, Gang  and
      Chen, Yating  and
      Ding, Haiyan  and
      Gao, Wang  and
      Jiajia, Huang  and
      Peng, Min  and
      Xie, Qianqian  and
      Yue, Kun",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.acl-long.1765/",
    pages = "38061--38104",
    ISBN = "979-8-89176-390-6",
    abstract = "While Large Language Models (LLMs) excel in various general domains, they exhibit notable gaps in the highly specialized, knowledge-intensive, and legally regulated Chinese tax domain. Consequently, while tax-related benchmarks are gaining attention, many focus on isolated NLP tasks, neglecting real-world practical capabilities. To address this issue, we introduce TaxPraBen, the first dedicated benchmark for Chinese taxation practice. It combines 10 traditional application tasks, along with 3 pioneering real-world scenarios: tax risk prevention, tax inspection analysis, and tax strategy planning, sourced from 14 datasets totaling 7.3K instances. TaxPraBen features a scalable structured evaluation paradigm designed through process of ``structured parsing{---}field alignment extraction{---}numerical and textual matching'', enabling end-to-end tax practice assessment while being extensible to other domains. We evaluate 19 LLMs based on Bloom{'}s taxonomy. The results indicate significant performance disparities: all closed-source large-parameter LLMs excel, and Chinese LLMs like Qwen2.5 generally exceed multilingual LLMs, while the YaYi2 LLM, fine-tuned with some tax data, shows only limited improvement. TaxPraBen[{\ensuremath{<}}https://anonymous.4open.science/r/TaxPraBen/{\ensuremath{>}}] serves as a vital resource for advancing evaluations of LLMs in practical applications."
}