zknews

trending3

tag: membership-inference ✕

01▲
When T2I Synthetic Data Backfires: Amplified Privacy Risks in Real-Synthetic Mix Training arxiv.org
Turns out T2I synthetic data can make membership inference on the real half easier, and this paper quantifies the mess.
arxiv machine-learning membership-inference privacy security synthetic-data
2 pts/hbrown/3 days ago/discuss
02▲
From Multiplicity to Vulnerability: Privacy Amplification Risk from One-Dataset-Multiple-Model Exposure arxiv.org
Privacy loss accumulates across multiple models trained on one dataset, and PRIME boosts membership inference by aggregating them.
arxiv black-box-attacks cryptanalysis machine-learning-security membership-inference privacy
10 pts/veramarsh/12 days ago/4 comments
03▲
Privacy Vulnerabilities of Attention Layers in Tabular Foundation Models and Protection of High-Risk Queries arxiv.org
Studies membership leakage in tabular foundation models via attention, then adds an inference-time k-anonymity style defense. Lovely.
arxiv attention machine-learning membership-inference privacy tabular-data transformers
0 pts/ivan_stderr/21 days ago/3 comments