Unsupervised Synthetic Image Attribution: Alignment and Disentanglement

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Liu, Zongfang, Chen, Guangyi, Sun, Boyang, Liu, Tongliang, Zhang, Kun
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911410742624256
author Liu, Zongfang
Chen, Guangyi
Sun, Boyang
Liu, Tongliang
Zhang, Kun
author_facet Liu, Zongfang
Chen, Guangyi
Sun, Boyang
Liu, Tongliang
Zhang, Kun
contents As the quality of synthetic images improves, identifying the underlying concepts of model-generated images is becoming increasingly crucial for copyright protection and ensuring model transparency. Existing methods achieve this attribution goal by training models using annotated pairs of synthetic images and their original training sources. However, obtaining such paired supervision is challenging, as it requires either well-designed synthetic concepts or precise annotations from millions of training sources. To eliminate the need for costly paired annotations, in this paper, we explore the possibility of unsupervised synthetic image attribution. We propose a simple yet effective unsupervised method called Alignment and Disentanglement. Specifically, we begin by performing basic concept alignment using contrastive self-supervised learning. Next, we enhance the model's attribution ability by promoting representation disentanglement with the Infomax loss. This approach is motivated by an interesting observation: contrastive self-supervised models, such as MoCo and DINO, inherently exhibit the ability to perform simple cross-domain alignment. By formulating this observation as a theoretical assumption on cross-covariance, we provide a theoretical explanation of how alignment and disentanglement can approximate the concept-matching process through a decomposition of the canonical correlation analysis objective. On the real-world benchmarks, AbC, we show that our unsupervised method surprisingly outperforms the supervised methods. As a starting point, we expect our intuitive insights and experimental findings to provide a fresh perspective on this challenging task.
format Preprint
id arxiv_https___arxiv_org_abs_2601_22663
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Unsupervised Synthetic Image Attribution: Alignment and Disentanglement
Liu, Zongfang
Chen, Guangyi
Sun, Boyang
Liu, Tongliang
Zhang, Kun
Computer Vision and Pattern Recognition
Artificial Intelligence
As the quality of synthetic images improves, identifying the underlying concepts of model-generated images is becoming increasingly crucial for copyright protection and ensuring model transparency. Existing methods achieve this attribution goal by training models using annotated pairs of synthetic images and their original training sources. However, obtaining such paired supervision is challenging, as it requires either well-designed synthetic concepts or precise annotations from millions of training sources. To eliminate the need for costly paired annotations, in this paper, we explore the possibility of unsupervised synthetic image attribution. We propose a simple yet effective unsupervised method called Alignment and Disentanglement. Specifically, we begin by performing basic concept alignment using contrastive self-supervised learning. Next, we enhance the model's attribution ability by promoting representation disentanglement with the Infomax loss. This approach is motivated by an interesting observation: contrastive self-supervised models, such as MoCo and DINO, inherently exhibit the ability to perform simple cross-domain alignment. By formulating this observation as a theoretical assumption on cross-covariance, we provide a theoretical explanation of how alignment and disentanglement can approximate the concept-matching process through a decomposition of the canonical correlation analysis objective. On the real-world benchmarks, AbC, we show that our unsupervised method surprisingly outperforms the supervised methods. As a starting point, we expect our intuitive insights and experimental findings to provide a fresh perspective on this challenging task.
title Unsupervised Synthetic Image Attribution: Alignment and Disentanglement
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2601.22663