Rethinking VLMs for Image Forgery Detection and Localization

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Guo, Shaofeng, Cui, Jiequan, Hong, Richang
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914390991699968
author Guo, Shaofeng
Cui, Jiequan
Hong, Richang
author_facet Guo, Shaofeng
Cui, Jiequan
Hong, Richang
contents With the rapid rise of Artificial Intelligence Generated Content (AIGC), image manipulation has become increasingly accessible, posing significant challenges for image forgery detection and localization (IFDL). In this paper, we study how to fully leverage vision-language models (VLMs) to assist the IFDL task. In particular, we observe that priors from VLMs hardly benefit the detection and localization performance and even have negative effects due to their inherent biases toward semantic plausibility rather than authenticity. Additionally, the location masks explicitly encode the forgery concepts, which can serve as extra priors for VLMs to ease their training optimization, thus enhancing the interpretability of detection and localization results. Building on these findings, we propose a new IFDL pipeline named IFDL-VLM. To demonstrate the effectiveness of our method, we conduct experiments on 9 popular benchmarks and assess the model performance under both in-domain and cross-dataset generalization settings. The experimental results show that we consistently achieve new state-of-the-art performance in detection, localization, and interpretability.Code is available at: https://github.com/sha0fengGuo/IFDL-VLM.
format Preprint
id arxiv_https___arxiv_org_abs_2603_12930
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Rethinking VLMs for Image Forgery Detection and Localization
Guo, Shaofeng
Cui, Jiequan
Hong, Richang
Computer Vision and Pattern Recognition
Machine Learning
68T45
I.4.8; I.4.9; I.2.10; K.6.5
With the rapid rise of Artificial Intelligence Generated Content (AIGC), image manipulation has become increasingly accessible, posing significant challenges for image forgery detection and localization (IFDL). In this paper, we study how to fully leverage vision-language models (VLMs) to assist the IFDL task. In particular, we observe that priors from VLMs hardly benefit the detection and localization performance and even have negative effects due to their inherent biases toward semantic plausibility rather than authenticity. Additionally, the location masks explicitly encode the forgery concepts, which can serve as extra priors for VLMs to ease their training optimization, thus enhancing the interpretability of detection and localization results. Building on these findings, we propose a new IFDL pipeline named IFDL-VLM. To demonstrate the effectiveness of our method, we conduct experiments on 9 popular benchmarks and assess the model performance under both in-domain and cross-dataset generalization settings. The experimental results show that we consistently achieve new state-of-the-art performance in detection, localization, and interpretability.Code is available at: https://github.com/sha0fengGuo/IFDL-VLM.
title Rethinking VLMs for Image Forgery Detection and Localization
topic Computer Vision and Pattern Recognition
Machine Learning
68T45
I.4.8; I.4.9; I.2.10; K.6.5
url https://arxiv.org/abs/2603.12930