HEDGE: Heterogeneous Ensemble for Detection of AI-GEnerated Images in the Wild

TL;DR

HEDGE is a heterogeneous ensemble method that combines diverse detection routes, including data augmentation, multi-scale features, and backbone heterogeneity, to robustly detect AI-generated images in various real-world conditions.

Contribution

The paper introduces HEDGE, a novel ensemble framework that integrates multiple detection strategies to improve robustness against diverse generative models and distortions.

Findings

Achieves 4th place in NTIRE 2026 challenge.

Attains state-of-the-art performance on multiple benchmarks.

Demonstrates strong robustness across varied conditions.

Abstract

Robust detection of AI-generated images in the wild remains challenging due to the rapid evolution of generative models and varied real-world distortions. We argue that relying on a single training regime, resolution, or backbone is insufficient to handle all conditions, and that structured heterogeneity across these dimensions is essential for robust detection. To this end, we propose HEDGE, a Heterogeneous Ensemble for Detection of AI-GEnerated images, that introduces complementary detection routes along three axes: diverse training data with strong augmentation, multi-scale feature extraction, and backbone heterogeneity. Specifically, Route~A progressively constructs DINOv3-based detectors through staged data expansion and augmentation escalation, Route~B incorporates a higher-resolution branch for fine-grained forensic cues, and Route~C adds a MetaCLIP2-based branch for backbone diversity. All outputs are fused via logit-space weighted averaging, refined by a lightweight dual-gating mechanism that handles branch-level outliers and majority-dominated fusion errors. HEDGE achieves 4th place in the NTIRE 2026 Robust AI-Generated Image Detection in the Wild Challenge and attains state-of-the-art performance with strong robustness on multiple AIGC image detection benchmarks.