Retrieval-Augmented Generation (RAG) for bibliographic analysis of complex texts as the case study of “Semiramis” by Alexey S. Khomyakov

The author examines application of the retrieval-augmented generation (RAG) technology to bibliographic processing of large unstructured philosophical texts as exemplified by Alexey S. Khomyakov’s “Semiramis”. She describes the principles of RAG system operation based on text semantic analysis and vector indexing that enables to extract relevant fragments promptly. The author also focuses on the challenges of processing “Semiramis”, hard-to-analyze threevolume uncompleted work without distinct topical structure. The paper comprises RAG and vector indexing description, as well as the findings of system testing on complex issues that demand intellectual analysis of the historiosophic content. The cited examples of the system responses demonstrate its ability to extract and classify arguments and ideas scattered within the text. The author concludes on the prospectivity of the method not only for Khomyakov’s legacy studies but also for other large-scale sources (diaries, memoirs, archival documents, etc.), hard to describe with traditional bibliographic methods. The paper makes the practical contribution to developing digital hermeneutics and demonstrates AI potential for the humanities research.

1. Stepanov V. K. Iskusstvenny`i` intellekt v deiatel`nosti bibliotek: teoreticheskie podhody` i prakticheskie resheniia (k itogam nauchno-prakticheskoi` konferentcii «Primenenie iskusstvennogo intellekta v bibliotechno-informatcionnoi` deiatel`nosti») // Nauchny`e i tekhnicheskie biblioteki. 2024. № 11. S. 15–30. https://doi.org/10.33186/1027-3689-3632024-11-15-30.

2. Neshcheret M. Iu. Nei`roseti v biblioteke: novoe v bibliograficheskom obsluzhivanii // Nauchny`e i tekhnicheskie biblioteki. 2024. № 1. S. 105–128. https://doi.org/10.33186/1027-3689-2024-1-105-128; Bermejo S. Enhancing Annotated Bibliography Generation with LLM Ensembles // arXiv. 2024. arXiv:2412.20864v1 [cs.CL]. URL: https://arxiv.org/abs/2412.20864 (data obrashcheniia: 24.11.2025). https://doi.org/10.48550/arXiv.2412.20864.

3. A. S. Homiakov. Polnoe sobranie sochinenii` i pisem v 12 tomakh. T. 5. Sankt-Peterburg : Rostok, 2024. S. 503.

4. Achiam J., Adler S., Agarwal S., Ahmad L. et al. (2023). GPT-4 Technical Report. arXiv preprint arXiv:2303.08774.

5. Lewis P., Perez E., Piktus A., Petroni F., Karpukhin V., Goyal N., Küttler H., Lewis M., Yih W., Rocktäschel T., Riedel S. and Kiela D. (2020). Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks. arXiv preprint arXiv:2005.11401.

6. Zuo S. et al. Context-Aware Query Rewriting for Improving Users’ Search Experience in E-Commerce // Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Industry Track). 2023. P. 538–547.

7. Aliev R. Improving E-commerce Search with Category-Aligned Retrieval // arXiv preprint arXiv:2510.21711. 2025.

8. Taylor J. A. Survey of AI tools in Library Tech: Accelerating into and Swerving around the Curve // Technical Services Quarterly. 2025. Vol. 42, No. 2. DOI: 10.1080/07317131.2025.2467574.

9. Breeding M. 2025 Library Systems Report: Companies see platform upgrades, new leadership, and AI enhancements // American Libraries Magazine. 2025. May. URL: https://americanlibrariesmagazine.org/2025/05/01/2025-library-systems-report/

10. Harper C. The Librarian Skillset and the Needs of Artificial Intelligence // Journal of Library Administration. 2025. Vol. 65, No. 4. DOI: 10.1080/01639374.2025.2539787.

11. Pinecone. Ofitcial`naia dokumentatciia. URL: https://docs.pinecone.io (data obrashcheniia: 28.11.2025).

12. Chroma. Ofitcial`naia dokumentatciia. URL: https://docs.trychroma.com (data obrashcheniia: 28.11.2025).

13. Milvus. Ofitcial`naia dokumentatciia. URL: https://milvus.io/docs (data obrashcheniia: 28.11.2025).

14. FAISS. Ofitcial`naia dokumentatciia. URL: https://faiss.ai/index.html (data obrashcheniia: 28.11.2025).

15. Platform V Vector DB (SberTekh). Ofitcial`naia dokumentatciia. URL: https://sbertech.ru/documents/platform-v-vector-db/1.0.0 (data obrashcheniia: 28.11.2025).

16. YDB Vector Search (Яндекс). Ofitcial`naia dokumentatciia. URL: https://ydb.tech/docs/en/dev/vector-indexes (data obrashcheniia: 28.11.2025).

17. Emel`ianov V. V. «Shumerskii` vopros» F. Vai`sbaha i sovremenny`e granitcy` shumerskogo voprosa // Indoevropei`skoe iazy`koznanie i classicheskaia filologiia. Vy`p. 29 (1). Sankt-Peterburg : Institut lingvisticheskikh issledovanii` RAN, 2025. S. 640–668, osob. s. 660–663.