aiqbench analysis
This research introduces PRIVEE, a defense mechanism designed to protect Vertical Federated Learning systems from feature inference attacks by obfuscating confidence scores without sacrificing predictive accuracy. By transforming shared representations rather than exposing raw data, the method significantly increases reconstruction error for adversaries while maintaining model utility across organizations with disjoint features. This work addresses critical privacy vulnerabilities in collaborative AI training pipelines where traditional methods fail to prevent input reconstruction via prediction probabilities.
PRIVEE: Privacy-Preserving Vertical Federated Learning Against Feature Inference Attacks
arXiv:2512.12840v2 Announce Type: replace Abstract: Vertical Federated Learning (VFL) enables collaborative model training across organizations that share common user samples but hold disjoint feature spaces. Despite its potential, VFL is susceptible to feature inference attacks, in which adversarial parties exploit shared confidence scores (prediction probabilities) during inference to reconstruct private input features of other participants. To counter this threat, we propose PRIVEE (PRIvacy-preserving Vertical fEderated lEarning), a novel defense mechanism named after the French word priv\'ee, meaning "private." PRIVEE obfuscates confidence scores while preserving critical properties such as relative ranking and inter-score distances. Rather than exposing raw scores, PRIVEE only shares transformed representations, mitigating risk of reconstruction attacks without degrading model prediction accuracy. Extensive experiments show that PRIVEE achieves up to a 30 times increase in reconstruction error (MSE) against feature inference attacks, compared to the strongest competing defense, while preserving full predictive performance against advanced feature inference attacks.
Abstract:Vertical Federated Learning (VFL) enables collaborative model training across organizations that share common user samples but hold disjoint feature spaces. Despite its potential, VFL is susceptible to feature inference attacks, in which adversarial parties exploit shared confidence scores (prediction probabilities) during inference to reconstruct private input features of other participants. To counter this threat, we propose PRIVEE (PRIvacy-preserving Vertical fEderated lEarning), a novel defense mechanism named after the French word privée, meaning "private." PRIVEE obfuscates confidence scores while preserving critical properties such as relative ranking and inter-score distances. Rather than exposing raw scores, PRIVEE only shares transformed representations, mitigating risk of reconstruction attacks without degrading model prediction accuracy. Extensive experiments show that PRIVEE achieves up to a 30 times increase in reconstruction error (MSE) against feature inference attacks, compared to the strongest competing defense, while preserving full predictive performance against advanced feature inference attacks.
Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
arXiv:2512.12840 [cs.LG]
(or arXiv:2512.12840v2 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2512.12840 Focus to learn more arXiv-issued DOI via DataCite
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 · 239 words
Related coverage
A Bayesian Vertical Federated Learning Framework for Multivariate Reduced-Rank High-Dimensional Regression
arXiv · Sep 22, 2026
CausShield: Sample Reconstruction-Resilient Vertical FL via Causal Representation Learning
arXiv · Jun 9, 2026
FedPS: Federated Preprocessing for structured data via aggregated Statistics
arXiv · Sep 5, 2026
Active-Passive Federated Learning for Vertically Partitioned Multi-view Data
arXiv · Jun 10, 2026