[2604.04800] Forgetting to Witness: Efficient Federated Unlearning and Its Visible Evaluation

Computer Science > Machine Learning arXiv:2604.04800 (cs) [Submitted on 6 Apr 2026] Title:Forgetting to Witness: Efficient Federated Unlearning and Its Visible Evaluation Authors:Houzhe Wang, Xiaojie Zhu, Chi Chen View a PDF of the paper titled Forgetting to Witness: Efficient Federated Unlearning and Its Visible Evaluation, by Houzhe Wang and 2 other authors View PDF HTML (experimental) Abstract:With the increasing importance of data privacy and security, federated unlearning has emerged as a novel research field dedicated to ensuring that federated learning models no longer retain or leak relevant information once specific data has been deleted. In this paper, to the best of our knowledge, we propose the first complete pipeline for federated unlearning, which includes a federated unlearning approach and an evaluation framework. Our proposed federated unlearning approach ensures high efficiency and model accuracy without the need to store historical this http URL effectively leverages the knowledge distillation model alongside various optimization mechanisms. Moreover, we propose a framework named Skyeye to visualize the forgetting capacity of federated unlearning models. It utilizes the federated unlearning model as the classifier integrated into a Generative Adversarial Network (GAN). Afterward, both the classifier and discriminator guide the generator in generating samples. Throughout this process, the generator learns from the classifier's knowledge. The generator the...