All Issue

2025 Vol.38, Issue 4 Preview Page

Research Paper

Design and Evaluation of StructCPT: A Retrieval-Augmented Generation System Specialized for the Korean Architectural Structural Engineering Domain

StructCPT: 한국 건축구조공학 도메인 특화 검색-증강 생성 시스템의 설계와 평가

Seok-Jae Heo1
,
Won-Jun Choi1
,
Sang-Hyun Lee2*
허 석재1
,
최 원준1
,
이 상현2*
1Research Professor, Seismic Retrofitting and Remodeling Research Center, Dankook University, Yongin, 16890, Korea
2Professor, Department of Architectural Engineering, Dankook University, Yongin, 16890, Korea
1단국대학교 리모델링연구소 연구교수
2단국대학교 건축학부 교수
*Corresponding Author
31 August 2025. pp. 257-267
PDF XML Citation
Abstract

Recent advancements in the field of architectural structural engineering have increasingly integrated artificial intelligence (AI) technologies that employ large language models (LLMs). However, these technologies face notable limitations, such as failing to incorporate region-specific structural design standards (e.g., KDS) or producing inaccurate and hallucinatory information. Retrieval-augmented generation (RAG) has emerged as a viable approach to overcoming these challenges. This study proposes StructCPT, an RAG system tailored for the Korean architectural structural engineering domain, and evaluates its performance. StructCPT is a domain-adapted retriever that extracts pertinent information from a Korean-language structural engineering knowledge base in real-time. It leverages maximum similarity retrieval (MAXIM), a contrastive learning-based embedding technique designed to enhance semantic similarity between queries and documents. Experimental results show that StructCPT consistently and significantly outperforms conventional retrieval methods, including BM25, Contriever, and SPECTER, across key quantitative metrics (Recall@10, MRR, and Precision@10). Notably, substantial improvements were observed in handling specialized structural engineering terminology and syntactically complex queries, resulting in enhanced retrieval accuracy and recall. Furthermore, StructCPT demonstrated practical scalability in terms of retrieval speed and memory usage, confirming its applicability to real-world scenarios. This study represents the first implementation of an RAG system specifically customized for structural engineering. It also outlines future developments, including the integration of multimodal data (e.g., drawings, images) and the incorporation of knowledge graphs, thereby laying a foundational basis for safe and trustworthy AI-driven decision support in structural engineering.

Keywords
large language Model (LLM)
retrieval-augmented generation (RAG)
domain-specific artificial intelligence
StructCPT
structural engineering knowledge base

최근 건축 구조공학 분야에서도 대규모 언어 모델(LLM)을 활용한 인공지능(AI) 기술 도입이 증가하고 있지만, 한국형 구조설계 기준과 같은 지역 특화 규정을 반영하지 못하거나 잘못된 정보를 제공하는 환각 현상 등 여러 한계를 보인다. 이를 극복할 수 있는 유망한 기술로서 검색-증강 생성(RAG)이 제시되고 있으며, 본 논문에서는 한국 건축 구조공학 도메인에 특화된 RAG 시스템인 StructCPT를 개발하여 그 성능을 평가하였다. StructCPT는 한국어 기반 구조공학 지식베이스에서 질의에 적합한 정보를 실시간으로 추출하는 도메인 특화 검색기이며, 대조학습 기반의 MAXIM(Maximum Similarity Retrieval) 임베딩 기법을 이용하여 질의와 문서 간 최대 의미적 유사도를 학습한다. 실험 결과 StructCPT는 BM25, Contriever, SPECTER와 같은 기존 범용 검색 기법들 대비 정량적 평가 지표에서 일관되고 유의미한 성능 향상을 보여주었다. 특히 구조공학 전문 용어 처리와 복합적 질의에 대한 검색 정확도 및 재현율에서 월등히 높은 성과를 나타냈으며, 실제 구조공학 문제 적용에서도 높은 정확도를 달성하였다. 또한 검색 속도와 메모리 사용 측면에서도 실무 적용에 적합한 효율성을 입증하였다. 본 연구는 구조공학 분야에 특화된 최초의 RAG 시스템 개발 사례로서, 향후 도면･이미지 등 멀티모달 정보와 지식그래프 통합을 통한 추가 발전 방향을 제시하며, 안전하고 신뢰할 수 있는 AI 기반 구조공학 의사결정 지원의 기초를 마련하였다.

키워드
대규모 언어 모델
검색 증강 생성
도메인 특화인공지능
StructCPT
구조공학지식베이스
References
1

Abacha, A.B., Yim, W.W., Adams, G., Snider, N., Yetisgen Yildiz, M. (2023) Overview of the MEDIQA Chat 2023 Shared Tasks on the Summarization & Generation of Doctor Patient Conversations, Proceedings of the 5th Clinical Natural Language Processing Workshop, p.503513.

2

Abootorabi, M.M., Zobeiri, A., Dehghani, M., Mohammadkhani, M., Mohammadi, B., Ghahroodi, O., Soleymani B,M., Asgari, E. (2025) Ask in Any Modality: A Comprehensive Survey on Multimodal Retrieval Augmented Generation, arXiv preprint arXiv:2502.08826.

10.18653/v1/2025.findings-acl.861
3

Borgeaud, S., Mensch, A., Hoffmann, J., Cai, T., Rutherford, E., Millican, K., van den Driessche, G., Lespiau, J.-B., Damoc, B., Clark, A., de Las Casas, D., Guy, A., Menick, J., Ring, R., Hennigan, T., Huang, S., Maggiore, L., Jones, C., Cassirer, A., Brock, A., Paganini, M., Irving, G., Vinyals, O., Osindero, S., Simonyan, K., Rae, J.W., Elsen, E., Sifre, L. (2022) Improving Language Models by Retrieving from Trillions of Tokens, Int. Conf. Machine Learning, p.22062240.

4

Cohan, A., Feldman, S., Beltagy, I., Downey, D., Weld, D.S. (2020). Specter: Document Level Representation Learning using CitationInformed Transformers, arXiv preprint arXiv: 2004.07180.

10.18653/v1/2020.acl-main.207
5

deFiteroDomínguez, D., GarcíaCabot, A., GarcíaLópez, E. (2024) Automated Multiple Choice Question Generation in Spanish using Neural Language Models, Neural Comput. & Appl., 36(29), p.1822318235.

10.1007/s00521-024-10076-7
6

Feng, Y., Zhu, W., Lu, X. (2024). Quake Former: A Uniform Approach to Earthquake Ground Motion Prediction Using Masked Transformers, arXiv preprint arXiv:2412.00815.

7

Han, H., Wang, Y., Shomer, H., Guo, K., Ding, J., Lei, Y., Halappanavar, M., Rossi, R.A., Mukherjee, S., Tang, X., He, Q., Hua, Z., Long, B., Zhao, T., Shah, N., Javari, A., Xia, Y., Tang, J. (2025), Retrieval Augmented Generation with Graphs (Graph RAG), arXiv preprint arXiv:2501.00309.

10.48550/arXiv.2501.00309
8

Izacard, G., Caron, M., Hosseini, L., Riedel, S., Bojanowski, P., Joulin, A., Grave, E. (2021) Unsupervised Dense Information Retrieval with Contrastive Learning, arXiv preprint arXiv: 2112.09118.

10.48550/arXiv.2112.09118
9

Ji, Z., Lee, N., Frieske, R., Yu, T., Su, D., Xu, Y., Ishii, E., Bang, Y., Chen, D., Dai, W., Chan, H.S., Madotto, A., Fung, P. (2023) Survey of Hallucination in Natural Language Generation, ACM Comput. Surv., 55(12), p.138.

10.1145/3571730
10

Karpukhin, V., Oguz, B., Min, S., Lewis, P., Wu, L., Edunov, S., Chen, D., Yih, W.-t. (2020). Dense Passage Retrieval for Open Domain Question Answering, EMNLP, 1, p.67696781.

10.18653/v1/2020.emnlp-main.550
11

Lewis, P., Perez, E., Piktus, A., Petroni, F., Karpukhin, V., Goyal, N., Küttler, H., Lewis, M., Yih, W.-t., Rocktäschel, T., Riedel, S., Kiela, D. (2020) Retrieval Augmented Generation for Knowledge Intensive NLP Tasks, Adv. Neural Inf. Proc. Syst., 33, p.94599474.

10.48550/arXiv.2005.11401
12

Meng, S., Zhou, Y., Zheng, Q., Liao, B., Chang, M., Zhang, T., Djerrad, A. (2024) SeisGPT: A PhysicsInformed Data Driven Large Model for RealTime Seismic Response Prediction, arXiv preprint arXiv:2410.20186.

10.48550/arXiv.2410.20186
13

Robertson, S., Zaragoza, H. (2009) The Probabilistic Relevance Framework: BM25 and Beyond, Found. & Trends® Inf. Retr., 3(4), p.333389.

10.1561/1500000019
14

Tapeh, A.T.G., Naser, M.Z. (2023) Artificial Intelligence, Machine Learning, and Deep Learning in Structural Engineering: A Scientometrics Review of Trends and Best Practices, Arch. Comput. Methods Eng., 30(1), p.115159.

10.1007/s11831-022-09793-w
15

Wei, J., Wang, X., Schuurmans, D., Bosma, M., Ichter, B., Xia, F., Chi, E.H., Le, Q.V., Zhou, D. (2022) Chain of Thought Prompting Elicits Reasoning in Large Language Models, Adv. Neural Inf. Proc. Syst., 35, p.2482424837.

10.48550/arXiv.2201.11903
16

Xia, Y., Zhou, J., Shi, Z., Chen, J., Huang, H. (2025) Improving Retrieval Augmented Language Model with Self Reasoning, Proceedings of the AAAI Conference on Artificial Intelligence, 39(24), p.2553425542.

10.1609/aaai.v39i24.34743
17

Xiong, L., Xiong, C., Li, Y., Tang, K.-F., Liu, J., Bennett, P., Ahmed, J., Overwijk, A. (2020) Approximate Nearest Neighbor Negative Contrastive Learning for Dense Text Retrieval, arXiv preprint arXiv:2007.00808.

10.48550/arXiv.2007.00808
18

Zhang, T., Patil, S.G., Jain, N., Shen, S., Zaharia, M., Stoica, I., Gonzalez, J.E. (2024). RAFT: Adapting Language Model to Domain Specific RAG, First Conference on Language Modeling.

Information
  • Publisher :Computational Structural Engineering Institute of Korea
  • Publisher(Ko) :한국전산구조공학회
  • Journal Title :Journal of the Computational Structural Engineering Institute of Korea
  • Journal Title(Ko) :한국전산구조공학회 논문집
  • Volume : 38
  • No :4
  • Pages :257-267
  • Received Date : 2025-04-21
  • Revised Date : 2025-05-12
  • Accepted Date : 2025-05-20
  • DOI :https://doi.org/10.7734/COSEIK.2025.38.4.257
Journal Informaiton Journal of the Computational Structural Engineering Institute of Korea Journal of the Computational Structural Engineering Institute of Korea
Online Submission
  • NRF
  • KOFST
  • crossref crossmark
  • crossref cited-by
  • crosscheck
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close