Privacy-Preserving Adaptation of Large Language Models: An Evidence Framework

This review examines privacy-preserving adaptation of large language models. The organizing question is how adaptation utility should be balanced against memorization, inference attacks, and data governance. Ten related scholarly sources are synthesized through a decision-centered framework spanning problem definition, mechanism, measurement, evaluation, implementation, and governance. The review does not invent experiments, pooled estimates, or unreported quantitative results. It instead evaluates the strength and transferability of the available evidence, with particular attention to reporting a privacy mechanism without measuring the end-to-end exposure surface. The resulting framework links technical or empirical performance to explicit use conditions and identifies tests that should precede wider adoption in domain adaptation using sensitive organizational or personal data.

1. Agrawal, S., Sarkar, S., Aouedi, O., Yenduri, G., Piamrat, K., Alazab, M., Bhattacharya, S., Maddikunta, P. K. R., & Gadekallu, T. R. (2022). Federated Learning for intrusion detection system: Concepts, challenges and future directions. Computer Communications, 195, 346-361. https://doi.org/10.1016/j.comcom.2022.09.012 DOI
2. Bouacida, N., & Mohapatra, P. (2021). Vulnerabilities in Federated Learning. IEEE Access, 9, 63229-63249. https://doi.org/10.1109/access.2021.3075203 DOI
3. Ferrag, M. A., Ndhlovu, M., Tihanyi, N., Cordeiro, L. C., Debbah, M., Lestable, T., & Thandi, N. S. (2024). Revolutionizing Cyber Threat Detection With Large Language Models: A Privacy-Preserving BERT-Based Lightweight Model for IoT/IIoT Devices. IEEE Access, 12, 23733-23750. https://doi.org/10.1109/access.2024.3363469 DOI
4. Gosselin, R., Vieu, L., Loukil, F., & Benoit, A. (2022). Privacy and Security in Federated Learning: A Survey. Applied Sciences, 12(19), 9901. https://doi.org/10.3390/app12199901 DOI
5. Hallaji, E., Razavi-Far, R., Saif, M., Wang, B., & Yang, Q. (2024). Decentralized Federated Learning: A Survey on Security and Privacy. IEEE Transactions on Big Data, 10(2), 194-213. https://doi.org/10.1109/tbdata.2024.3362191 DOI
6. Liu, B., Ding, M., Shaham, S., Rahayu, W., Farokhi, F., & Lin, Z. (2021). When Machine Learning Meets Privacy. ACM Computing Surveys, 54(2), 1-36. https://doi.org/10.1145/3436755 DOI
7. Liu, X., Xie, L., Wang, Y., Zou, J., Xiong, J., Ying, Z., & Vasilakos, A. V. (2020). Privacy and Security Issues in Deep Learning: A Survey. IEEE Access, 9, 4566-4593. https://doi.org/10.1109/access.2020.3045078 DOI
8. Wang, Q., Yin, H., Chen, T., Yu, J., Zhou, A., & Zhang, X. (2021). Fast-adapting and privacy-preserving federated recommender system. The VLDB Journal, 31(5), 877-896. https://doi.org/10.1007/s00778-021-00700-6 DOI
9. Ye, M., Fang, X., Du, B., Yuen, P. C., & Tao, D. (2023). Heterogeneous Federated Learning: State-of-the-art and Research Challenges. ACM Computing Surveys, 56(3), 1-44. https://doi.org/10.1145/3625558 DOI
10. Yin, X., Zhu, Y., & Hu, J. (2021). A Comprehensive Survey of Privacy-preserving Federated Learning. ACM Computing Surveys, 54(6), 1-36. https://doi.org/10.1145/3460427 DOI