# Robust Federated Inference Source: [Robust Federated Inference]() Collection index: [llms.txt]() Data updated: 2026-04-16T05:00:11.996Z Authors: Akash Dhasade, Sadegh Farhadkhani, Rachid Guerraoui, Nirupam Gupta, Maxime Jacovella, Anne-Marie Kermarrec, Rafael Pinot Published: 2025-09-30T22:08:55.000Z ## Abstract Federated inference, in the form of one-shot federated learning, edge ensembles, or federated ensembles, has emerged as an attractive solution to combine predictions from multiple models. This paradigm enables each model to remain local and proprietary while a central server queries them and aggregates predictions. Yet, the robustness of federated inference has been largely neglected, leaving them vulnerable to even simple attacks. To address this critical gap, we formalize the problem of robust federated inference and provide the first robustness analysis of this class of methods. Our analysis of averaging-based aggregators shows that the error of the aggregator is small either when the dissimilarity between honest responses is small or the margin between the two most probable classes is large. Moving beyond linear averaging, we show that problem of robust federated inference with non-linear aggregators can be cast as an adversarial machine learning problem. We then introduce an advanced technique using the DeepSet aggregation model, proposing a novel composition of adversarial training and test-time robust aggregation to robustify non-linear aggregators. Our composition yields significant improvements, surpassing existing robust aggregation methods by 4.7 - 22.2% in accuracy points across diverse benchmarks. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Image Classification on CIFAR-10 (Adversarial Robustness Metrics)](): Image Classification; CIFAR-10; Logit Flipping Accuracy: 67.6. - [Text Classification on AG-News (Adversarial Robustness Metrics)](): Text Classification; AG-News; Logit Flipping: 85.7. - [Image Classification on CIFAR-100 alpha=0.3 (test)](): Image Classification; CIFAR-100 alpha=0.3 (test); Logit Flipping: 75.7. - [Image Classification on CIFAR-100 (Specific Robustness Attacks)](): Image Classification; CIFAR-100; Logit Flipping Robustness: 78. ## Original sources and implementation [Original paper]() [PDF]()