Scaling Up AI-Generated Image Detection with Generator-Aware Prototypes

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Qin, Ziheng, Ji, Yuheng, Tao, Renshuai, Tian, Yuxuan, Liu, Yuyang, Wang, Yipu, Zheng, Xiaolong
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911586073968640
author Qin, Ziheng
Ji, Yuheng
Tao, Renshuai
Tian, Yuxuan
Liu, Yuyang
Wang, Yipu
Zheng, Xiaolong
author_facet Qin, Ziheng
Ji, Yuheng
Tao, Renshuai
Tian, Yuxuan
Liu, Yuyang
Wang, Yipu
Zheng, Xiaolong
contents The pursuit of a universal AI-generated image (AIGI) detector often relies on aggregating data from numerous generators to improve generalization. However, this paper identifies a paradoxical phenomenon we term the Benefit then Conflict dilemma, where detector performance stagnates and eventually degrades as source diversity expands. Our systematic analysis, diagnoses this failure by identifying two core issues: severe data-level heterogeneity, which causes the feature distributions of real and synthetic images to increasingly overlap, and a critical model-level bottleneck from fixed, pretrained encoders that cannot adapt to the rising complexity. To address these challenges, we propose Generator-Aware Prototype Learning (GAPL), a framework that constrain representation with a structured learning paradigm. GAPL learns a compact set of canonical forgery prototypes to create a unified, low-variance feature space, effectively countering data heterogeneity.To resolve the model bottleneck, it employs a two-stage training scheme with Low-Rank Adaptation, enhancing its discriminative power while preserving valuable pretrained knowledge. This approach establishes a more robust and generalizable decision boundary. Through extensive experiments, we demonstrate that GAPL achieves state-of-the-art performance, showing superior detection accuracy across a wide variety of GAN and diffusion-based generators. Code is available at https://github.com/UltraCapture/GAPL
format Preprint
id arxiv_https___arxiv_org_abs_2512_12982
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Scaling Up AI-Generated Image Detection with Generator-Aware Prototypes
Qin, Ziheng
Ji, Yuheng
Tao, Renshuai
Tian, Yuxuan
Liu, Yuyang
Wang, Yipu
Zheng, Xiaolong
Computer Vision and Pattern Recognition
The pursuit of a universal AI-generated image (AIGI) detector often relies on aggregating data from numerous generators to improve generalization. However, this paper identifies a paradoxical phenomenon we term the Benefit then Conflict dilemma, where detector performance stagnates and eventually degrades as source diversity expands. Our systematic analysis, diagnoses this failure by identifying two core issues: severe data-level heterogeneity, which causes the feature distributions of real and synthetic images to increasingly overlap, and a critical model-level bottleneck from fixed, pretrained encoders that cannot adapt to the rising complexity. To address these challenges, we propose Generator-Aware Prototype Learning (GAPL), a framework that constrain representation with a structured learning paradigm. GAPL learns a compact set of canonical forgery prototypes to create a unified, low-variance feature space, effectively countering data heterogeneity.To resolve the model bottleneck, it employs a two-stage training scheme with Low-Rank Adaptation, enhancing its discriminative power while preserving valuable pretrained knowledge. This approach establishes a more robust and generalizable decision boundary. Through extensive experiments, we demonstrate that GAPL achieves state-of-the-art performance, showing superior detection accuracy across a wide variety of GAN and diffusion-based generators. Code is available at https://github.com/UltraCapture/GAPL
title Scaling Up AI-Generated Image Detection with Generator-Aware Prototypes
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.12982