Toward Generalizable Forgery Detection and Reasoning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Gao, Yueying, Chang, Dongliang, Yu, Bingyao, Qin, Haotian, Diao, Muxi, Chen, Lei, Liang, Kongming, Ma, Zhanyu
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914449492803584
author Gao, Yueying
Chang, Dongliang
Yu, Bingyao
Qin, Haotian
Diao, Muxi
Chen, Lei
Liang, Kongming
Ma, Zhanyu
author_facet Gao, Yueying
Chang, Dongliang
Yu, Bingyao
Qin, Haotian
Diao, Muxi
Chen, Lei
Liang, Kongming
Ma, Zhanyu
contents Accurate and interpretable detection of AI-generated images is essential for mitigating risks associated with AI misuse. However, the substantial domain gap among generative models makes it challenging to develop a generalizable forgery detection model. Moreover, since every pixel in an AI-generated image is synthesized, traditional saliency-based forgery explanation methods are not well suited for this task. To address these challenges, we formulate detection and explanation as a unified Forgery Detection and Reasoning task (FDR-Task), leveraging Multi-Modal Large Language Models (MLLMs) to provide accurate detection through reliable reasoning over forgery attributes. To facilitate this task, we introduce the Multi-Modal Forgery Reasoning dataset (MMFR-Dataset), a large-scale dataset containing 120K images across 10 generative models, with 378K reasoning annotations on forgery attributes, enabling comprehensive evaluation of the FDR-Task. Furthermore, we propose FakeReasoning, a forgery detection and reasoning framework with three key components: 1) a dual-branch visual encoder that integrates CLIP and DINO to capture both high-level semantics and low-level artifacts; 2) a Forgery-Aware Feature Fusion Module that leverages DINO's attention maps and cross-attention mechanisms to guide MLLMs toward forgery-related clues; 3) a Classification Probability Mapper that couples language modeling and forgery detection, enhancing overall performance. Experiments across multiple generative models demonstrate that FakeReasoning not only achieves robust generalization but also outperforms state-of-the-art methods on both detection and reasoning tasks. The code is available at: https://github.com/PRIS-CV/FakeReasoning.
format Preprint
id arxiv_https___arxiv_org_abs_2503_21210
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Toward Generalizable Forgery Detection and Reasoning
Gao, Yueying
Chang, Dongliang
Yu, Bingyao
Qin, Haotian
Diao, Muxi
Chen, Lei
Liang, Kongming
Ma, Zhanyu
Computer Vision and Pattern Recognition
Accurate and interpretable detection of AI-generated images is essential for mitigating risks associated with AI misuse. However, the substantial domain gap among generative models makes it challenging to develop a generalizable forgery detection model. Moreover, since every pixel in an AI-generated image is synthesized, traditional saliency-based forgery explanation methods are not well suited for this task. To address these challenges, we formulate detection and explanation as a unified Forgery Detection and Reasoning task (FDR-Task), leveraging Multi-Modal Large Language Models (MLLMs) to provide accurate detection through reliable reasoning over forgery attributes. To facilitate this task, we introduce the Multi-Modal Forgery Reasoning dataset (MMFR-Dataset), a large-scale dataset containing 120K images across 10 generative models, with 378K reasoning annotations on forgery attributes, enabling comprehensive evaluation of the FDR-Task. Furthermore, we propose FakeReasoning, a forgery detection and reasoning framework with three key components: 1) a dual-branch visual encoder that integrates CLIP and DINO to capture both high-level semantics and low-level artifacts; 2) a Forgery-Aware Feature Fusion Module that leverages DINO's attention maps and cross-attention mechanisms to guide MLLMs toward forgery-related clues; 3) a Classification Probability Mapper that couples language modeling and forgery detection, enhancing overall performance. Experiments across multiple generative models demonstrate that FakeReasoning not only achieves robust generalization but also outperforms state-of-the-art methods on both detection and reasoning tasks. The code is available at: https://github.com/PRIS-CV/FakeReasoning.
title Toward Generalizable Forgery Detection and Reasoning
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2503.21210