Automatic compilation of a Pre-Qin philosophy lexicon via large language models
Source abstract
Abstract
Pre-Qin philosophy is foundational to Chinese intellectual history, but its terminology features polysemy, contextuality, and school-specific variation, posing challenges to manual dictionary compilation. This study proposes an automatic lexicon construction method using large language models. A semantic framework is established, consisting of four core tasks: term identification, school classification, definition generation, and context translation. Corpora are sourced from Chinese Text Project and Guoxue Dashi, with structured term data based on the Encyclopedia of Chinese Philosophy. Training employs the LLaMA Factory with LoRA-based tuning on DeepSeek-R1-Distill-Qwen, Qwen3, and Llama3 models, using continued pretraining and few-shot fine-tuning. During inference, task-specific few-shot prompts guide the model in semantic generation for each task. Results show Qwen3 outperforms DeepSeek-R1-Distill-Qwen and Llama3 in semantic alignment and disambiguation. A structured term database and interactive Streamlit-based interface support an end-to-end workflow from corpus acquisition to dictionary display, highlighting the potential of large language models in automating lexicographic work on classical Chinese philosophy.
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Natural Language Processing Techniques · Digital Humanities and Scholarship · Topic Modeling
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (32)
Show source-provided references
1. D’Ambrosio, P. Boundary of the sky: environmentalism, Daoism, and the logic of increase. Asian Stud. 11, 41–67 (2023).10.4312/as.2023.11.2.41-67 ↗
2. Radice, T. Li (ritual) in early confucianism. Philos. Compass 12, e12463 (2017).10.1111/phc3.12463 ↗
3. Tan, J. & Bao, X. Reexamining the different paths to the dao of the Daodejing. Religions 13, 1216 (2022).10.3390/rel13121216 ↗
4. Aizawa, A. An information-theoretic perspective of TF-IDF measures. Inf. Process. Manag. 39, 45–65 (2003).10.1016/S0306-4573(02)00021-3 ↗
5. Mlambo, R., Skosana, N. & Matfunjwa, M. The extraction of terminology list using ParaConc for creating a quadrilingual dictionary. South. Afr. Linguist. Appl. Lang. Stud. 39, 82–91 (2021).
6. Kumar, P. Large language models (LLMs): survey, technical frameworks, and future challenges. Artif. Intell. Rev. 57, 260 (2024).10.1007/s10462-024-10888-y ↗
7. Zhang, Y. et al. Qwen3 embedding: advancing text embedding and reranking through foundation models. arXiv preprint. 2025. https://arxiv.org/abs/2506.05176.
8. Patil, R. & Gudivada, V. A review of current trends, techniques, and challenges in large language models (LLMs). Appl. Sci. 14, 2074 (2024).10.3390/app14052074 ↗
9. Zheng, J. & Xiao, X. A complex network approach to analyse pre-trained language models for ancient Chinese. R. Soc. Open Sci. 11, 240061 (2024).10.1098/rsos.240061 ↗
10. Wu, M. & Wang, D. Tracing the evolution of ancient Chinese military science through classical texts. Herit. Sci. 12, 431 (2024).10.1186/s40494-024-01536-3 ↗
11. Liu, C. et al. SikuGPT: a generative pre-trained model for intelligent information processing of ancient texts from the perspective of digital humanities. J. Comput. Cultural Herit. 17, 1–17 (2024).
12. Xu, Q. et al. Automatic recognition of cross-language classic entities based on large language models. npj Herit. Sci. 13, 1–12 (2025).10.1038/s40494-025-01624-y ↗
13. Wei, X. The use of large language models for translating Buddhist texts from classical Chinese to modern English: an analysis and evaluation with ChatGPT 4, ERNIE Bot 4, and Gemini Advanced. Religions 15, 1559 (2024).10.3390/rel15121559 ↗
14. Jiahuan, C. et al. TongGu: Mastering Classical Chinese Understanding with Knowledge-Grounded Large Language Models. In Findings of the Association for Computational Linguistics: EMNLP 2024, 4196–4210, (Miami, Florida, USA. Association for Computational Linguistics, 2024).10.18653/v1/2024.findings-emnlp.243 ↗
15. Hua, R. et al. Lingdan: enhancing encoding of traditional Chinese medicine knowledge for clinical reasoning tasks with large language models. J. Am. Med. Inform. Assoc. 31, 2019–2029 (2024).10.1093/jamia/ocae087 ↗
16. Lew, R. ChatGPT as a COBUILD lexicographer. Humanities Soc. Sci. Commun. 10, 1–10 (2023).
17. Merx, R., Vylomova, E. & Kurniawan, K. Generating bilingual example sentences with large language models as lexicography assistants. In Proc. 22nd Annual Workshop of the Australasian Language Technology Association, 64–74 (Canberra, Australia. Association for Computational Linguistics, 2024).
18. Lew, R. Dictionaries and lexicography in the AI era. Humanities Soc. Sci. Commun. 11, 1–8 (2024).
19. Ptasznik, B., Wolfer, S. & Lew, R. A learners’ dictionary versus ChatGPT in receptive and productive lexical tasks. Int. J. Lexicogr. 37, 322–336 (2024).10.1093/ijl/ecae011 ↗
20. Lew, R., Ptasznik, B. & Wolfer, S. The effectiveness of ChatGPT as a lexical tool for English, compared with a bilingual dictionary and a monolingual learner’s dictionary. Humanities Soc. Sci. Commun. 11, 1–10 (2024).
21. De Schryver, G. M. Generative AI and lexicography: the current state of the art using ChatGPT. Int. J. Lexicogr. 36, 355–387 (2023).10.1093/ijl/ecad021 ↗
22. Hu, B. et al. Research on the evolution of “Ren” and “Li” in SikuQuanshu Confucian classics. Front. Psychol. 12, 603344 (2021).10.3389/fpsyg.2021.603344 ↗
23. Wong, P. H. Dao, harmony and personhood: towards a Confucian ethics of technology. Philos. Technol. 25, 67–86 (2012).10.1007/s13347-011-0021-z ↗
24. Chen, X. et al. A fine-grained self-adapting prompt learning approach for few-shot learning with pre-trained language models. Knowl.-Based Syst. 299, 111968 (2024).10.1016/j.knosys.2024.111968 ↗
25. Tianyu, G., Adam, F. & Danqi, C. Making Pre-trained Language Models Better Few-shot Learners. In Proc. 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing, Vol. 1, 3816–3830, (Association for Computational Linguistics, 2021).10.18653/v1/2021.acl-long.295 ↗
26. Yuan, J. et al. Prompt-based learning for few-shot class-incremental learning. Alex. Eng. J. 120, 287–295 (2025).10.1016/j.aej.2025.02.008 ↗
27. Huang, C. et al. Lorahub: Efficient cross-task generalization via dynamic lora composition. arXiv preprint 2023. https://arxiv.org/abs/2307.13269.
28. Yaowei, Z. et al. LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models. In Proc, 62nd Annual Meeting of the Association for Computational Linguistics (Vol. 3: System Demonstrations) 400–410 (Bangkok, Thailand. Association for Computational Linguistics, 2024).10.18653/v1/2024.acl-demos.38 ↗
29. Zhang, B. et al. When scaling meets LLM finetuning: the effect of data, model and finetuning method. arXiv preprint. 2024. https://arxiv.org/abs/2402.17193.
30. Yang, A. et al. Qwen3 technical report. arXiv preprint. 2025. https://arxiv.org/abs/2505.09388.
31. DeepSeek-AI et al. DeepSeek-R1: incentivizing reasoning capability in LLMs via reinforcement learning. arXiv preprint. 2025. https://arxiv.org/abs/2501.12948.
32. Grattafiori, A. et al. The Llama 3 herd of models. arXiv preprint. 2024. https://arxiv.org/abs/2407.21783.