mlsanityData to 14 Oct 2026

On the Inference (In-)Security of Vertical Federated Learning: Efficient Auditing against Inference Tampering Attack

arXiv 2507.02376 ↗First posted 3 Jul 2025cs.SE cs.AI cs.DCcited by 0

Not on the map. Why?

Abstract (arXiv)

Vertical Federated Learning (VFL) is an emerging distributed learning paradigm for cross-silo collaboration without accessing participants' data. However, existing VFL work lacks a mechanism to audit the inference correctness of the data party. The malicious data party can modify the local data and model to mislead the joint inference results. To exploit this vulnerability, we design a novel Vertical Federated Inference Tampering (VeFIT) attack, allowing the data party to covertly tamper with the local inference and mislead results on the task party's final prediction. VeFIT can decrease the task party's inference accuracy by an average of 34.49%. Existing defense mechanisms can not effectively detect this attack, and the detection performance is near random guessing. To mitigate the attack, we further design a Vertical Federated Inference Auditing (VeFIA) framework. VeFIA helps the task party to audit whether the data party's inferences are executed as expected during large-scale online inference. VeFIA does not leak the data party's privacy nor introduce additional latency. The core design is that the task party can use the inference results from a framework with Trusted Execution Environments (TEE) and the coordinator to validate the correctness of the data party's computation results. VeFIA guarantees that, as long as the proportion of inferences attacked by VeFIT exceeds 5.4%, the task party can detect the malicious behavior of the data party with a probability of 99.99%, without any additional online overhead. VeFIA's random sampling validation of VeFIA achieves 100% positive predictive value, negative predictive value, and true positive rate in detecting VeFIT. We further validate VeFIA's effectiveness in terms of privacy protection and scalability on real-world datasets. To the best of our knowledge, this is the first paper discussing the inference auditing problem towards VFL.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 30 of its 90 references that are in the map. First posted on arXiv 3 Jul 2025.

Builds on

30 of 90 references.

References · 30

  1. 2018Representation Learning with Contrastive Predictive Codingcited by 7,777
  2. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,669
  3. 2018MINE: Mutual Information Neural Estimationcited by 683
  4. 2018Split learning for health: Distributed deep learning without sharing raw patient datacited by 395
  5. 2020CLUB: A Contrastive Log-ratio Upper Bound of Mutual Informationcited by 235
  6. 2020Serving DNNs like Clockwork: Performance Predictability from the Bottom Upcited by 120
  7. 2020Feature Inference Attack on Model Predictions in Vertical Federated Learningcited by 106
  8. 2020Unleashing the Tiger: Inference Attacks on Split Learningcited by 105
  9. 2018Slalom: Fast, Verifiable and Private Execution of Neural Networks in Trusted Hardwarecited by 102
  10. 2018MLCapsule: Guarded Offline Deployment of Machine Learning as a Servicecited by 31
  11. 2021FedV: Privacy-Preserving Federated Learning over Vertically Partitioned Datacited by 31
  12. 2023A Survey on Vertical Federated Learning: From a Layered Perspectivecited by 30
18 more
  1. 2022CrowdGuard: Federated Backdoor Detection in Federated Learningcited by 27
  2. 2023BayBFed: Bayesian Backdoor Defense for Federated Learningcited by 23
  3. 2023No Privacy Left Outside: On the (In-)Security of TEE-Shielded DNN Partition for On-Device MLcited by 21
  4. 2023Vertical Federated Knowledge Transfer via Representation Distillation for Healthcare Collaboration Networkscited by 19
  5. 2019Privacy-Preserving Inference in Machine Learning Services Using Trusted Execution Environmentscited by 17
  6. 2024BackdoorIndicator: Leveraging OOD Data for Proactive Backdoor Detection in Federated Learningcited by 17
  7. 2020A Vertical Federated Learning Method for Interpretable Scorecard and Its Application in Credit Scoringcited by 16
  8. 2023FLShield: A Validation Based Federated Learning Framework to Defend Against Poisoning Attackscited by 15
  9. 2024Inf2Guard: An Information-Theoretic Framework for Learning Privacy-Preserving Representations against Inference Attackscited by 13
  10. 2023martFL: Enabling Utility-Driven Data Marketplace with a Robust and Verifiable Federated Learning Architecturecited by 11
  11. 2022ADI: Adversarial Dominating Inputs in Vertical Federated Learning Systemscited by 9
  12. 2022Securing Federated Sensitive Topic Classification against Poisoning Attackscited by 6
  13. 2022Dordis: Efficient Federated Learning with Dropout-Resilient Differential Privacycited by 6
  14. 2021Preventing Machine Learning Poisoning Attacks Using Authentication and Provenancecited by 5
  15. 2022OLIVE: Oblivious Federated Learning on Trusted Execution Environment against the risk of sparsificationcited by 4
  16. 2024VFLIP: A Backdoor Defense for Vertical Federated Learning via Identification and Purificationcited by 4
  17. 2018Interface-Based Side Channel Attack Against Intel SGXcited by 2
  18. 2024A Learning-Based Attack Framework to Break SOTA Poisoning Defenses in Federated Learningcited by 2

Built on by

Not cited yet.