![[2602.19918] RobPI: Robust Private Inference against Malicious Client](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.19918] RobPI: Robust Private Inference against Malicious Client
Summary
The paper presents RobPI, a robust private inference protocol designed to counteract malicious client attacks, demonstrating significant improvements in security and efficiency over existing methods.
Why It Matters
As machine learning applications proliferate, ensuring data privacy against malicious actors becomes critical. This research addresses vulnerabilities in current private inference protocols, offering a solution that enhances security and reliability in sensitive data handling.
Key Takeaways
- RobPI significantly reduces the success rate of malicious client attacks by approximately 91.9%.
- The protocol requires over 10 times more queries to compromise compared to existing methods.
- RobPI integrates cryptographic techniques to enhance security without sacrificing efficiency.
Computer Science > Cryptography and Security arXiv:2602.19918 (cs) [Submitted on 23 Feb 2026] Title:RobPI: Robust Private Inference against Malicious Client Authors:Jiaqi Xue, Mengxin Zheng, Qian Lou View a PDF of the paper titled RobPI: Robust Private Inference against Malicious Client, by Jiaqi Xue and 2 other authors View PDF HTML (experimental) Abstract:The increased deployment of machine learning inference in various applications has sparked privacy concerns. In response, private inference (PI) protocols have been created to allow parties to perform inference without revealing their sensitive data. Despite recent advances in the efficiency of PI, most current methods assume a semi-honest threat model where the data owner is honest and adheres to the protocol. However, in reality, data owners can have different motivations and act in unpredictable ways, making this assumption unrealistic. To demonstrate how a malicious client can compromise the semi-honest model, we first designed an inference manipulation attack against a range of state-of-the-art private inference protocols. This attack allows a malicious client to modify the model output with 3x to 8x fewer queries than current black-box attacks. Motivated by the attacks, we proposed and implemented RobPI, a robust and resilient private inference protocol that withstands malicious clients. RobPI integrates a distinctive cryptographic protocol that bolsters security by weaving encryption-compatible noise into the log...
