mlsanityData to 14 Oct 2026

Backdoors Leave Structural Traces: FedMAST for Backdoor Detection and Containment in Federated Learning

arXiv 2609.27760 ↗First posted 4 Sep 2026cs.LG cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Federated learning enables distributed training without requiring clients to share their raw data. However, its reliance on the integrity of the client-submitted updates exposes the global model to stealthy backdoor poisoning. Existing defenses often rely on individual evidence sources, but stealth-constrained attacks can adapt to these signals. Such attacks can suppress anomaly signals they are optimized to evade, yet their poisoned updates still leave residual structural traces. We propose FedMAST, a Federated Multi-Axis Structural Tracing defense for backdoor detection in federated learning. FedMAST scores client updates using complementary structural, spectral, and historical evidence and then applies tiered filtering and round-level containment to limit adversarial influence. To capture traces that isolated signals may miss, FedMAST uses squeeze-pair coherence scoring to expose coupled feature distortions and signed spectral-drift tracking to reveal persistent directional changes over time. Across six backdoor attacks, FedMAST achieves lower attack success rate (ASR) than baseline defenses in all nine evaluated comparisons, averaging 1.51% ASR and 94.84% main-task accuracy (MTA) across the complete 200-round runs. Over the full 200-round method-aware CovertLayers run, FedMAST achieves 1.53% ASR and 92.26% MTA, compared with ASRs of 100.00%, 99.67%, 99.53%, and 32.84% for FedAvg, MultiKrum, AlignIns, and FLAME, respectively.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 21 of its 27 references that are in the map. First posted on arXiv 4 Sep 2026.

Builds on

21 of 27 references.

References · 21

  1. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,669
  2. 2019Measuring the Effects of Non-Identical Data Distribution for Federated Visual Classificationcited by 851
  3. 2018How To Backdoor Federated Learningcited by 765
  4. 2018Byzantine-Robust Distributed Learning: Towards Optimal Statistical Ratescited by 711
  5. 2018Analyzing Federated Learning through an Adversarial Lenscited by 351
  6. 2020FLTrust: Byzantine-robust Federated Learning via Trust Bootstrappingcited by 346
  7. 2018The Hidden Vulnerability of Distributed Learning in Byzantiumcited by 329
  8. 2020Attack of the Tails: Yes, You Really Can Backdoor Federated Learningcited by 309
  9. 2019A Little Is Enough: Circumventing Defenses For Distributed Learningcited by 288
  10. 2022FLDetector: Defending Federated Learning Against Model Poisoning Attacks via Detecting Malicious Clientscited by 125
  11. 2022Neurotoxin: Durable Backdoors in Federated Learningcited by 98
  12. 2021CRFL: Certifiably Robust Federated Learning against Backdoor Attackscited by 91
9 more
  1. 2022DeepSight: Mitigating Backdoor Attacks in Federated Learning Through Deep Model Inspectioncited by 81
  2. 2023FreqFed: A Frequency Analysis-Based Approach for Mitigating Poisoning Attacks in Federated Learningcited by 27
  3. 2023Multi-metrics adaptively identifies backdoors in Federated learningcited by 23
  4. 2023Backdoor Federated Learning by Poisoning Backdoor-Critical Layerscited by 18
  5. 2025Detecting Backdoor Attacks in Federated Learning via Direction Alignment Inspectioncited by 13
  6. 2025MARS: A Malignity-Aware Backdoor Defense in Federated Learningcited by 7
  7. 2024Adversarially Guided Stateful Defense Against Backdoor Attacks in Federated Deep Learningcited by 5
  8. 2025Heterogeneity-Oblivious Robust Federated Learningcited by 1
  9. 2026FedSurrogate: Backdoor Defense in Federated Learning via Layer Criticality and Surrogate Replacementcited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Enhancing the Effectiveness and Durability of Backdoor Attacks in Federated Learning through Maximizing Task Distinction
  2. 2023FedGrad: Mitigating Backdoor Attacks in Federated Learning Through Local Ultimate Gradients Inspection
  3. 2023On the Vulnerability of Backdoor Defenses for Federated Learning
  4. 2025TrojanDam: Detection-Free Backdoor Defense in Federated Learning through Proactive Model Robustification utilizing OOD Data
  5. 2025Runtime Backdoor Detection for Federated Learning via Representational Dissimilarity Analysis
  6. 2025BackFed: An Efficient & Standardized Benchmark Suite for Backdoor Attacks in Federated Learning
  7. 2024Concealing Backdoor Model Updates in Federated Learning by Trigger-Optimized Data Poisoning
  8. 2024Client-Side Patching against Backdoor Attacks in Federated Learning