RealStats: A Rigorous Real-Only Statistical Framework for Fake Image Detection

TL;DR

RealStats introduces a statistically rigorous, training-free framework for fake image detection that combines multiple detectors and provides interpretable probability scores to improve robustness against evolving generative models.

Contribution

It presents a novel, formal statistical framework that aggregates multiple detection methods without training, enhancing robustness and interpretability in fake image detection.

Findings

Effective in diverse detection scenarios

Provides interpretable probability scores

Robust against distributional shifts

Abstract

As generative models continue to evolve, detecting AI-generated images remains a critical challenge. While effective detection methods exist, they often lack formal interpretability and may rely on implicit assumptions about fake content, potentially limiting robustness to distributional shifts. In this work, we introduce a rigorous, statistically grounded framework for fake image detection that focuses on producing a probability score interpretable with respect to the real-image population. Our method leverages the strengths of multiple existing detectors by combining training-free statistics. We compute p-values over a range of test statistics and aggregate them using classical statistical ensembling to assess alignment with the unified real-image distribution. This framework is generic, flexible, and training-free, making it well-suited for robust fake image detection across diverse and evolving settings.