aiqbench analysis
This arXiv preprint formalizes Split Federated Unlearning, the problem of removing a client's or class's contribution from a model trained with split federated learning, and proposes an initial method, SURE. The reported results are mixed: SURE approaches retraining-based baselines at modest extra cost, while existing unlearning methods adapted to this setting show clear forgetting-utility trade-offs. For ML readers, the practical signal is that unlearning in partitioned, federated systems is a distinct open problem, not a simple extension of centralized methods, and the results are early and need independent validation.
Split Federated Unlearning
arXiv:2610.36465v1 Announce Type: new Abstract: Split federated learning (SFL) combines partitioned model execution with federated aggregation to support collaborative learning across distributed Web and edge environments. However, removing learned contributions in SFL is challenging due to cross-partition coupling and historical influence propagation through dual-side aggregation. To the best of our knowledge, we provide the first systematic formulation of Split Federated Unlearning (SFU), encompassing client-wise and class-wise removal, counterfactual retraining, and partition-specific update constraints. We theoretically characterize historical influence propagation and mode-dependent unlearning feasibility, revealing structural limitations of restricted model updates. As an initial solution, we propose SURE, integrating dual-importance estimation, selective parameter perturbation, and importance-guided recovery. Extensive experiments with existing unlearning methods adapted to SFL reveal substantial forgetting-utility trade-offs across model architectures, data distributions, and update modes. SURE achieves competitive agreement with mode-matched retraining at modest additional cost, while further analyses highlight the impact of split position and update constraints on unlearning behavior. These findings establish SFU as a distinct research problem and provide a foundation for unlearning in partitioned federated systems.
Abstract:Split federated learning (SFL) combines partitioned model execution with federated aggregation to support collaborative learning across distributed Web and edge environments. However, removing learned contributions in SFL is challenging due to cross-partition coupling and historical influence propagation through dual-side aggregation. To the best of our knowledge, we provide the first systematic formulation of Split Federated Unlearning (SFU), encompassing client-wise and class-wise removal, counterfactual retraining, and partition-specific update constraints. We theoretically characterize historical influence propagation and mode-dependent unlearning feasibility, revealing structural limitations of restricted model updates. As an initial solution, we propose SURE, integrating dual-importance estimation, selective parameter perturbation, and importance-guided recovery. Extensive experiments with existing unlearning methods adapted to SFL reveal substantial forgetting-utility trade-offs across model architectures, data distributions, and update modes. SURE achieves competitive agreement with mode-matched retraining at modest additional cost, while further analyses highlight the impact of split position and update constraints on unlearning behavior. These findings establish SFU as a distinct research problem and provide a foundation for unlearning in partitioned federated systems.
Machine Learning (cs.LG); Cryptography and Security (cs.CR)
arXiv:2610.36465 [cs.LG]
(or arXiv:2610.36465v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2610.36465 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 265 words
Related coverage
Lethe: Principled Dual-Stream Update for Persistent Knowledge Erasure in Federated Unlearning
arXiv · Oct 6, 2026
\textsc{Lethe}: Principled Dual-Stream Update for Persistent Knowledge Erasure in Federated Unlearning
arXiv · Jun 5, 2026
Lethe: Adapter-Augmented Dual-Stream Update for Persistent Knowledge Erasure in Federated Unlearning
arXiv · Jun 3, 2026
pFedUL: Layer-Aware Federated Unlearning for Personalized Federated Learning
arXiv · Jun 16, 2026