ResearchResearch paperEfficiency & Inference · Safety & Alignment · Data & Datasets1 source · Oct 13, 2026

SoK: Rethinking Membership Inference in Tabular Machine Learning

Membership Inference Attacks (MIAs) are currently the dominant approach to evaluating privacy in machine learning applications.

Existing evaluations focus predominantly on high-dimensional datasets, leaving the privacy behavior of small-to-medium datasets unexplored.
In this paper, we present a new taxonomy for MIAs and conduct a comprehensive study of the factors influencing vulnerability.
We evaluate four distinct state-of-the-art attacks across diverse tabular datasets and model architectures.
Most importantly, we show that records with unique signatures (single-outs) remain highly vulnerable, even when common defenses are applied.

[1]SoK: Rethinking Membership Inference in Tabular Machine Learning
arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 13, 02:44 PM
Membership Inference Attacks (MIAs) are currently the dominant approach to evaluating privacy in machine learning applications.
Existing evaluations focus predominantly on high-dimensional datasets, leaving the privacy behavior of small-to-medium datasets unexplored.

Extractive summary: sentences quoted from the sources.

Before this