FedFlex: Protecting Shared Features in Vertical Federated Learning via Differential Privacy

Summary

FedFlex is a vertical federated learning framework that protects sensitive information in shared features. It applies differential privacy in two steps: first adding task-agnostic noise for generic protection, then adaptively adjusting the scale and distribution of the shared features to recover model accuracy. The authors report that it resists prevalent privacy attacks and balances privacy and accuracy better than prior methods.

Related items

• MediumClaude Code’s ‘don’t ask again’ button has a secrets-spilling problemSimilar attack · CSO Online
• MediumCVE-2026-125386: langchain4j-aideepin insecure direct object reference in message endpointsSimilar attack · NVD/CVE Database
• InfoSNAPRAG: Cost-Efficient Knowledge Extraction Attack on RAG Systems via Adversarial PerturbationsSimilar attack · Arxiv (cs.CR + cs.CL + cs.LG)
• LowMGE: A Membership Guided Data Extraction Attack against Diffusion ModelsSimilar attack · Arxiv (cs.RO + cs.CV security)
• InfoGem5-Shark: Smaug-based Heterogeneous Deep Learning Accelerator Security and Robustness Evaluation and Characterization FrameworkSimilar attack · OpenAlex (peer-reviewed AI security)