图书情报工作

图书情报工作 >

2026 , Vol. 70 >Issue 3: 3 - 11

DOI: https://doi.org/10.13266/j.issn.0252-3116.2026.03.001

专家视点

大语言模型范式下的计算人文研究

  • 黄水清 ,
  • 齐月 ,
  • 赵雪 ,
  • 许乾坤 ,
  • 吴梦成
展开
  • 1 南京农业大学信息管理学院 南京 210095;
    2 人文与社会计算江苏省高校哲学社会科学重点研究基地 南京 210095
黄水清，教授，博士，博士生导师，E-mail：sqhuang@njau.edu.cn；齐月，博士研究生；赵雪，博士研究生；许乾坤，博士研究生；吴梦成，博士研究生。

收稿日期: 2025-06-20

  修回日期: 2025-08-27

  网络出版日期: 2026-02-09

基金资助

本文系国家社会科学基金重大项目“先秦诸子典籍知识库建设及词典编纂”（项目编号：22&ZD262）、江苏省社会科学基金一般项目“《二十四史》人物知识图谱构建与应用研究”（项目编号：24TQB004）和南京农业大学中央高校基本科研业务费项目“古籍人物知识组织及挖掘研究”（项目编号：SKYD2025039）研究成果之一。

收起

Large Language Models Paradigm-Based Computational Humanities Research

  • Huang Shuiqing ,
  • Qi Yue ,
  • Zhao Xue ,
  • Xu Qiankun ,
  • Wu Mengcheng
Expand
  • 1 School of Information Management, Nanjing Agricultural University, Nanjing 210095;
    2 Research Center for Humanities and Social Computing, Nanjing Agricultural University, Nanjing 210095

Received date: 2025-06-20

  Revised date: 2025-08-27

  Online published: 2026-02-09

Supported by

This work is supported by the National Social Science Fund of China key project titled “Construction of Knowledge Base and Dictionary Compilation for Pre-Qin Philosophical Texts”(Grant No. 22&ZD262); the Jiangsu Provincial Social Science Foundation project titled “Research on the Construction and Application of Knowledge Graphs for Figures in the ‘Twenty-Four Histories’” (Grant No. 24TQB004); and project of the Fundamental Research Funds for the Central Universities of Nanjing Agricultural University, titled “Knowledge Organization and Data Mining of Characters in Classical Chinese Literature” (Grant No. SKYD2025039).

Fold

摘要

[目的/意义] 大语言模型的发展正深刻影响着计算人文这一交叉领域的研究，不仅带来了计算技术的突破，更推动了研究范式的转变，探析大语言模型对于计算人文方法论体系的影响具有重要意义。[方法/过程] 对大语言模型范式下的计算人文研究展开深入探讨，梳理大语言模型前沿技术特别是DeepSeek的技术框架，阐明大语言模型的自然语言理解与生成两种关键能力，结合大语言模型在计算人文领域的应用现状，反思当前研究存在的问题与挑战，展望大语言模型范式下计算人文研究的发展前景。[结果/结论] 大语言模型为计算人文开辟了全新的技术路径，拓宽了研究视角与方法，正在逐步实现从计算工具到计算思维的转变。但同时也应正视其能力的局限，在融合前沿技术方法与传统人文问题时仍需有所把握。

关键词： 计算人文; 数字人文; 大语言模型; 大语言模型

本文引用格式

黄水清 , 齐月 , 赵雪 , 许乾坤 , 吴梦成 . 大语言模型范式下的计算人文研究[J]. 图书情报工作, 2026 , 70(3) : 3 -11 . DOI: 10.13266/j.issn.0252-3116.2026.03.001

Abstract

[Purpose/Significance] The development of large language models (LLMs) is profoundly impacting research in computational humanities, an interdisciplinary field, not only bringing breakthroughs in computational technologies but also driving a paradigm shift in research. Analyzing the influence of LLMs on the methodological framework of computational humanities is of great significance. [Method/Process] This article conducts an in-depth exploration of computational humanities research under the LLMs paradigm, reviews large language model technologies, particularly DeepSeek's technical framework, and clarifies the two key capabilities of LLMs' natural language understanding and generation. Based on the current applications of LLMs in the field, this paper reflects on existing challenges and limitations while envisioning future prospects for research under the LLMs paradigm. [Result/Conclusion] LLMs have opened new technical pathways for computational humanities, broadening research perspectives and methodologies, and facilitating a shift from computational tools to computational thinking. However, it is equally crucial to acknowledge their limitations and maintain a balanced approach when integrating cutting-edge methods with traditional humanities inquiries.

Key words： computational humanities; digital humanities; LLM; DeepSeek

参考文献

[1] 黄水清, 刘浏, 王东波. 计算人文学科的内涵、体系及机遇[J]. 图书与情报, 2023(1): 1-11, 153, 145. (HUANG S Q, LIU L, WANG D B. The connotation, system and opportunity of computational humanities[J]. Library and information, 2023(1): 1-11, 153, 145.)
[2] XU J, GUO Z, HE J, et al. Qwen2.5-Omni technical report[EB/OL]. [2025-10-06]. https://arxiv.org/abs/2503.20215.
[3] WEI J, BOSMA M, ZHAO V Y, et al. Finetuned language models are zero-shot learners[EB/OL]. [2025-10-06]. https://arxiv.org/abs/2109.01652.
[4] OUYANG L, WU J, JIANG X, et al. Training language models to follow instructions with human feedback[J]. Advances in neural information processing systems, 2022, 35: 27730-27744.
[5] WEIDINGER L, MELLOR J, RAUH M, et al. Ethical and social risks of harm from language models[EB/OL]. [2025-12-08]. https://arxiv.org/abs/2112.04359.
[6] WEI J, TAY Y, BOMMASANI R, et al. Emergent abilities of large language models[EB/OL]. [2025-10-06]. https://arxiv.org/abs/2206.07682.
[7] DAI D, DENG C, ZHAO C, et al. Deepseekmoe: towards ultimate expert specialization in mixture-of-experts language models[EB/OL]. [2025-10-06]. https://arxiv.org/abs/2401.06066.
[8] LIU A, FENG B, WANG B, et al. Deepseek-v2: A strong, economical, and efficient mixture-of-experts language model[EB/OL]. [2025-09-19]. https://arxiv.org/abs/2405.04434.
[9] LIU A, FENG B, XUE B, et al. Deepseek-v3 technical report[EB/OL]. [2025-09-18]. https://arxiv.org/abs/2412.19437.
[10] SHAO Z, WANG P, ZHU Q, et al. Deepseekmath: Pushing the limits of mathematical reasoning in open language models[EB/OL]. [2025-09-27]. https://arxiv.org/abs/2402.03300.
[11] SCHULMAN J, WOLSKI F, DHARIWAL P, et al. Proximal policy optimization algorithms[EB/OL]. [2025-08-28]. https://arxiv.org/abs/1707.06347.
[12] DING X, CHEN H, ZHANG X, et al. Re-parameterizing your optimizers rather than architectures[EB/OL]. [2025-09-09]. https://arxiv.org/abs/2205.15242.
[13] GUO D, YANG D, ZHANG H, et al. Deepseek-r1: Incentivizing reasoning capability in LLMs via reinforcement learning[EB/OL]. [2025-09-22]. https://arxiv.org/abs/2501.12948.
[14] 夏天, 于凯, 余芊蓉, 等. 基于生成式大语言模型的古文自动断句与标点研究[J]. 图书情报工作, 2025, 69(5): 59-70. (XIA T, YU K, YU Q R, et al. Automatic sentence segmentation and automatic punctuation for ancient Chinese texts based on generative large language model[J]. Library and information service, 2025, 69(5): 59-70.)
[15] 王希羽, 王东波. 基于大语言模型的跨语言典籍自动分词研究[J]. 图书馆杂志, 2025, 44(9): 104-115. (WANG X Y, WANG D B. Research on automatic word segmentation of cross-language classics based on large language model[J]. Library Journal, 2025, 44(9): 104-115.)
[16] TANGHERLINI T R, CHEN R. Travels with BERT: Surfacing the intertextuality in Hans Christian Andersen’s travel writing and fairy tales through the network lens of large language model-based topic modeling[J]. Orbis litterarum, 2024, 79(6): 519-562.
[17] GOKCEOGLU G, CAVUSOGLU D, AKBAS E, et al. A multi-level multi-label text classification dataset of 19th century Ottoman and Russian literary and critical texts[EB/OL]. [2025-07-21]. https://arxiv.org/abs/2407.15136.
[18] JADHAV S, SHANBHAG A, THAKURDESAI A, et al. On limitations of LLM as Annotator for low resource languages[EB/OL]. [2025-09-01]. https://arxiv.org/abs/2411.17637.
[19] 刘畅, 张琪, 王东波, 等. 基于大语言模型技术的古籍限定域关系抽取及应用研究[J]. 情报学报, 2025, 44(2): 200-219. (LIU C, ZHANG Q, WANG D B, et al. Research on the extraction and application of ancient books’ restricted domain relation based on large language model technology[J]. Journal of the China Society for Scientific and Technical Information, 2025, 44(2): 200-219.)
[20] 周正达, 王昊, 汪琳, 等. ChatKG：一种基于大语言模型和提示工程的非遗知识图谱构建框架——以中国非遗陶瓷制作工艺为例[J/OL]. [2025-10-06]. https://link.cnki.net/urlid/31.1108.g2.20250224.1136.008. (ZHOU Z D, WANG H, WANG L, et al. ChatKG: a framework for constructing intangible cultural heritage knowledge graphs based on large language model and prompt engineering: a case study of Chinese intangible cultural heritage ceramics craft[J/OL]. [2025-10-06]. https://link.cnki.net/urlid/31.1108.g2.20250224.1136.008.)
[21] KONG L X, CHU Y G, MA Z, et al. Mix-Lingual relation extraction: dataset and a training approach[J]. Journal of computer science and technology, 2025, 40(1): 42-59.
[22] 叶文豪, 胡蝶, 王东波, 等. 基于古籍大模型的无监督互文自动发现研究[J]. 图书情报工作, 2024, 68(23): 41-51. (YE W H, HU D, WANG D B, et al. Research on unsupervised automatic intertextual discovery based on large models of ancient books[J]. Library and information service, 2024, 68(23): 41-51.)
[23] 张强, 高颖, 任豆豆, 等. 融合DeepSeek-R1和RAG技术的先秦文化元典智能问答研究[J]. 现代情报, 2026, 46(1): 173-186. (ZHANG Q, GAO Y, REN D D, et al. Research on intelligent question answering for pre-qin cultural classics by integrating deepSeek-r1 and RAG technologies[J]. Journal of Modern Information, 2026, 46(1): 173-186.)
[24] 赵衍, 张慧, 杨祎辰. 大语言模型在文本翻译中的质量比较研究——以《繁花》翻译为例[J]. 外语电化教学, 2024(4): 60-66, 109. (ZHAO Y, ZHANG H, YANG Y C. Comparative study on the translation quality of large language models: taking the translation of “Fan Hua” as an example[J]. Technology enhanced foreign language education, 2024(4): 60-66, 109.)
[25] 吴娜, 刘畅, 刘江峰, 等. AIGC驱动古籍自动摘要研究：从自然语言理解到生成[J]. 图书馆论坛, 2024, 44(9): 111-123. (WU N, LIU C, LIU J F, et al. AIGC-Driven Research on automatic summarization of ancient classics: from natural language understanding to natural language generation[J]. Library tribune, 2024, 44(9): 111-123.)
[26] 苑中旭, 李理, 何凡, 等. 融合思维链与知识图谱的中医问答模型[J]. 计算机工程与应用, 2025, 61(4): 158-166. (YUAN Z X, LI L, HE F, et al. Traditional Chinese medicine question answering model based on chain-of-thought and knowledge graph[J]. Computer engineering and applications, 2025, 61(4): 158-166.)
[27] 杨海峰. 用户行为在信息检索中的研究现状及发展动态评述[J]. 图书情报知识, 2015(6): 79-87. (YANG H F. Review on the research status and development trend of users behavior in information retrieval[J]. Documentation, information & knowledge, 2015(6): 79-87.)
[28] 刘浏, 齐月, 刘雏菲, 等. 计算人文下的古籍引书研究及全文本知识库的构建[J]. 情报学报, 2023, 42(12): 1498-1512. (LIU L, QI Y, LIU C F, et al. Research on ancient book citations from the perspective of computational humanities and the construction of full-text knowledge base[J]. Journal of the China Society for Scientific and Technical Information, 2023, 42(12): 1498-1512.)
[29] 魏向清, 刘润泽, 王东波, 等. 人文社科汉英术语知识库构建探索：理念与方法[M]. 南京: 南京大学出版社, 2022. (WEI X Q, LIU R Z, WANG D Z, et al. Exploration on the construction of a Chinese-English terminology knowledge base for humanities and social sciences: concepts and methods[M]. Nanjing: Nanjing University Press, 2022.)
[30] 刘雪扬. 文言词汇知识库在古文教学中的建构与应用[J]. 汉字文化, 2025(8): 28-30. (LIU X Y. Construction and application of Classical Chinese vocabulary knowledge base in ancient text teaching[J]. Sinogram culture, 2025(8): 28-30.)
[31] 张文亮, 丁淼, 魏来. 满文古籍知识库研究与构建[J]. 图书情报工作, 2025, 69(5): 126-139. (ZHANG W L, DING M, WEI L. Research and construction of the knowledge base of Manchu ancient books[J]. Library and information service, 2025, 69(5): 126-139.)
[32] 李佳斌, 魏庭新, 曲维光, 等. 大语言模型下古诗笺注知识库的构建与应用[J]. 图书馆论坛, 2025, 45(3): 99-109. (LI J B, WEI T X, QU W G, et al. Construction and application of a knowledge base for ancient poetry annotation with large language models[J]. Library tribune, 2025, 45(3): 99-109.)
[33] MCCARTY W. A New Companion to Digital Humanitie[M]. SCHREIBMAN S, SIEMENS R, UNSWORTH J. Chichester: John Wiley & Sons, 2015: 67-83.
[34] 黄水清, 刘浏, 王东波. 计算人文教育的回顾和探讨[J]. 情报理论与实践, 2024, 47(7): 1-9. (HUANG S Q, LIU L, WANG D B. Review and discussion of computational humanities education[J]. Information studies: theory & application, 2024, 47(7): 1-9.)
Options
文章导航

/

〈 〉