Recent Advances in Federated Learning Driven Large Language Models: A Survey on Architecture, Performance, and Security

TL;DR

This survey reviews recent progress in federated learning for large language models, focusing on architecture, performance, security, and the emerging field of machine unlearning to enhance privacy and adaptability.

Contribution

It provides a comprehensive overview of recent methods for federated LLMs, especially on unlearning techniques, and evaluates their efficiency, privacy, and utility trade-offs.

Findings

Unlearning methods vary in efficiency and privacy guarantees.

Federated LLMs face challenges in balancing performance and security.

Case studies demonstrate practical effectiveness of unlearning strategies.

Abstract

Federated Learning (FL) offers a promising paradigm for training Large Language Models (LLMs) in a decentralized manner while preserving data privacy and minimizing communication overhead. This survey examines recent advancements in FL-driven LLMs, with a particular emphasis on architectural designs, performance optimization, and security concerns, including the emerging area of machine unlearning. In this context, machine unlearning refers to the systematic removal of specific data contributions from trained models to comply with privacy regulations such as the Right to be Forgotten. We review a range of strategies enabling unlearning in federated LLMs, including perturbation-based methods, model decomposition, and incremental retraining, while evaluating their trade-offs in terms of efficiency, privacy guarantees, and model utility. Through selected case studies and empirical evaluations, we analyze how these methods perform in practical FL scenarios. This survey identifies critical research directions toward developing secure, adaptable, and high-performing federated LLM systems for real-world deployment.