Publication
Abstract
Large Language Models (LLMs) have demonstrated remarkable capabilities in natural language processing tasks, yet their inherent vulnerabilities to data leakage pose significant security and privacy risks. This paper presents a comprehensive analysis of assessment methods and protection strategies for addressing data leakage risks in LLMs. A systematic evaluation framework is proposed, incorporating multi-dimensional risk assessment models and quantitative metrics for vulnerability detection. The research examines various protection mechanisms across different stages of the LLM lifecycle, from data pre-processing to post-deployment monitoring. Through extensive analysis of protection techniques, the study reveals that integrated defense strategies combining gradient protection, query filtering, and output sanitization achieve optimal security outcomes, with risk reduction rates exceeding 95%. The implementation of these protection mechanisms demonstrates varying effectiveness across different operational scenarios, with performance impacts ranging from 8% to 18%. The research contributes to the field by establishing standardized evaluation criteria and proposing enhanced protection strategies that balance security requirements with system performance. The findings provide valuable insights for developing robust security frameworks in LLM deployments, while identifying critical areas for future research in adaptive defense mechanisms and scalable protection solutions.
Keywords
Large Language Models , Data Leakage Protection , Security Assessment , Privacy-Preserving Machine Learning .
Metadata
Disciplines: Artificial Intelligence Technology
Subjects: Natural Language Processing
Cite This Article
APA Style
Xiao, X., Zhang, Y., Xu, J., Ren, W. & Zhang, J. (2025). Assessment methods and protection strategies for data leakage risks in large language models. Journal of Industrial Engineering and Applied Science, 3(2), 6-15. https://doi.org/10.70393/6a69656173.323736
Acknowledgments
The authors thank the editor and anonymous reviewers for their helpful comments and valuable suggestions.
FUNDING
INSTITUTIONAL REVIEW BOARD STATEMENT
DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.
INFORMED CONSENT STATEMENT
CONFLICT OF INTEREST
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
AUTHOR CONTRIBUTIONS
References
1.
Das, B. C., Amini, M. H., & Wu, Y. (2024). Security and privacy challenges of large language models: A survey. ACM Computing Surveys.
2.
Balloccu, S., Schmidtová, P., Lango, M., & Dušek, O. (2024). Leak, cheat, repeat: Data contamination and evaluation malpractices in closed-source llms. arXiv preprint arXiv:2402.03927.
3.
Mathis, M., Blom, J. F., Nemecek, T., Bravin, E., Jeanneret, P., Daniel, O., & de Baan, L. (2022). Comparison of exemplary crop protection strategies in Swiss apple production: Multi-criteria assessment of pesticide use, ecotoxicological risks, environmental and economic impacts. Sustainable Production and Consumption, 31, 512-528.
4.
Kim, S., Yun, S., Lee, H., Gubri, M., Yoon, S., & Oh, S. J. (2024). Propile: Probing privacy leakage in large language models. Advances in Neural Information Processing Systems, 36.
5.
Stucki, A. O., Barton-Maclaren, T. S., Bhuller, Y., Henriquez, J. E., Henry, T. R., Hirn, C., ... & Clippinger, A. J. (2022). Use of new approach methodologies (NAMs) to meet regulatory requirements for the assessment of industrial chemicals and pesticides for effects on human health. Frontiers in Toxicology, 4, 964553.
6.
Yao, Y., Duan, J., Xu, K., Cai, Y., Sun, Z., & Zhang, Y. (2024). A survey on large language model (llm) security and privacy: The good, the bad, and the ugly. High-Confidence Computing, 100211.
7.
Zhang, X., Xu, H., Ba, Z., Wang, Z., Hong, Y., Liu, J., ... & Ren, K. (2024). Privacyasst: Safeguarding user privacy in tool-using large language model agents. IEEE Transactions on Dependable and Secure Computing.
8.
Nahar, J., Hossain, M. S., Rahman, M. M., & Hossain, M. A. (2024). Advanced Predictive Analytics For Comprehensive Risk Assessment In Financial Markets: Strategic Applications And Sector-Wide Implications. Global Mainstream Journal of Business, Economics, Development & Project Management, 3(4), 39-53.
9.
Shen, Q., Zhang, Y., & Xi, Y. (2024). Deep Learning-Based Investment Risk Assessment Model for Distributed Photovoltaic Projects. Journal of Advanced Computing Systems, 4(3), 31-46.
10.
Chen, J., Zhang, Y., & Wang, S. (2024). Deep Reinforcement Learning-Based Optimization for IC Layout Design Rule Verification. Journal of Advanced Computing Systems, 4(3), 16-30.
11.
Ju, C. (2023). A Machine Learning Approach to Supply Chain Vulnerability Early Warning System: Evidence from US Semiconductor Industry. Journal of Advanced Computing Systems, 3(11), 21-35.
12.
Çakmakçı, R., Salık, M. A., & Çakmakçı, S. (2023). Assessment and principles of environmentally sustainable food and agriculture systems. Agriculture, 13(5), 1073.
13.
Ju, C., & Ma, X. (2024). Real-time Cross-border Payment Fraud Detection Using Temporal Graph Neural Networks: A Deep Learning Approach. International Journal of Computer and Information System (IJCIS), 5(1), 103-114.
14.
Chen, H., Shen, Z., Wang, Y. and Xu, J., 2024. Threat Detection Driven by Artificial Intelligence: Enhancing Cybersecurity with Machine Learning Algorithms.
15.
Liang, X., & Chen, H. (2019, July). A SDN-Based Hierarchical Authentication Mechanism for IPv6 Address. In 2019 IEEE International Conference on Intelligence and Security Informatics (ISI) (pp. 225-225). IEEE.
16.
Pasino, A., De Angeli, S., Battista, U., Ottonello, D., & Clematis, A. (2021). A review of single and multi-hazard risk assessment approaches for critical infrastructures protection. International Journal of Safety and Security Engineering, 11(4), 305-318.
17.
Liang, X., & Chen, H. (2019, August). HDSO: A High-Performance Dynamic Service Orchestration Algorithm in Hybrid NFV Networks. In 2019 IEEE 21st International Conference on High Performance Computing and Communications; IEEE 17th International Conference on Smart City; IEEE 5th International Conference on Data Science and Systems (HPCC/SmartCity/DSS) (pp. 782-787). IEEE.
18.
Chen, H., & Bian, J. (2019, February). Streaming media live broadcast system based on MSE. In Journal of Physics: Conference Series (Vol. 1168, No. 3, p. 032071). IOP Publishing.
19.
Ke, Z., Zhou, S., Zhou, Y., Chang, C. H., & Zhang, R. (2025). Detection of AI Deepfake and Fraud in Online Payments Using GAN-Based Models. arXiv preprint arXiv:2501.07033.
20.
Yu, Q., Ke, Z., Xiong, G., Cheng, Y., & Guo, X. (2025). Identifying Money Laundering Risks in Digital Asset Transactions Based on AI Algorithms.
21.
Xu, J., Chen, H., Xiao, X., Zhao, M., Liu, B. (2025). Gesture Object Detection and Recognition Based on YOLOv11.Applied and Computational Engineering,133,81-89.
22.
Khan, S., Naushad, M., Lima, E. C., Zhang, S., Shaheen, S. M., & Rinklebe, J. (2021). Global soil pollution by toxic elements: Current status and future perspectives on the risk assessment and remediation strategies–A review. Journal of Hazardous Materials, 417, 126039.
23.
Ke, Z., Xu, J., Zhang, Z., Cheng, Y., & Wu, W. (2024). A Consolidated Volatility Prediction with Back Propagation Neural Network and Genetic Algorithm. arXiv preprint arXiv:2412.07223.
24.
Weidinger, L., Mellor, J., Rauh, M., Griffin, C., Uesato, J., Huang, P. S., ... & Gabriel, I. (2021). Ethical and social risks of harm from language models. arXiv preprint arXiv:2112.04359.
25.
Lukas, N., Salem, A., Sim, R., Tople, S., Wutschitz, L., & Zanella-Béguelin, S. (2023, May). Analyzing leakage of personally identifiable information in language models. In 2023 IEEE Symposium on Security and Privacy (SP) (pp. 346-363). IEEE.
26.
Huang, J., Shao, H., & Chang, K. C. C. (2022). Are large pre-trained language models leaking your personal information?. arXiv preprint arXiv:2205.12628.
PUBLISHER'S NOTE
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Copyright © 2025 The Author(s). Published by Southern United Academy of Sciences.
This work is licensed under a Creative
Commons Attribution 4.0 International License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
