arXiv：cs.CL· Xuemei Tang, Chengxi Yan, Jinghang Gu, Chu-Ren Huang·· 3 小时前AI 评分31
CHisAgent：面向中国古代文化体系事件分类体系构建的多智能体框架
CHisAgent: A Multi-Agent Framework for Event Taxonomy Construction in Ancient Chinese Cultural Systems
AI 导读
CHisAgent 是一个用于中国古代历史文化分类体系构建的多智能体 LLM 框架，将构建过程拆分为自下而上的 Inducer、自上而下的 Expander 和证据引导的 Enricher 三个阶段。该框架基于《二十四史》构建了覆盖政治、军事、外交与社会生活的大规模领域事件分类体系，无参考与有参考评估均显示其结构连贯性和覆盖度提升。分析还表明该分类体系支持跨文化对齐。
正文
Abstract:Despite strong performance on many tasks, large language models (LLMs) show limited ability in historical and cultural reasoning, particularly in non-English contexts such as Chinese history. Taxonomic structures offer an effective mechanism to organize historical knowledge and improve understanding. However, manual taxonomy construction is costly and difficult to scale. Therefore, we propose \textbf{CHisAgent}, a multi-agent LLM framework for historical taxonomy construction in ancient Chinese contexts. CHisAgent decomposes taxonomy construction into three role-specialized stages: a bottom-up \textit{Inducer} that derives an initial hierarchy from raw historical corpora, a top-down \textit{Expander} that introduces missing intermediate concepts using LLM world knowledge, and an evidence-guided \textit{Enricher} that integrates external structured historical resources to ensure faithfulness. Using the \textit{Twenty-Four Histories}, we construct a large-scale, domain-aware event taxonomy covering politics, military, diplomacy, and social life in ancient China. Extensive reference-free and reference-based evaluations demonstrate improved structural coherence and coverage, while further analysis shows that the resulting taxonomy supports cross-cultural alignment.
| Comments: | EMNLP 2026 findings |
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2601.05520 [cs.CL] |
| (or arXiv:2601.05520v4 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2601.05520 arXiv-issued DOI via DataCite |
Submission history
From: Xuemei Tang [view email]
[v1]
Fri, 9 Jan 2026 04:28:45 UTC (2,649 KB)
[v2]
Wed, 2 Sep 2026 09:10:37 UTC (2,957 KB)
[v3]
Thu, 3 Sep 2026 05:02:41 UTC (2,957 KB)
[v4]
Wed, 7 Oct 2026 05:12:49 UTC (2,957 KB)
来源：arXiv：cs.CL · arxiv.org