Logo image
Back
FakeBench: Probing Explainable Fake Image Detection via Large Multimodal Models
Journal article   Peer reviewed

FakeBench: Probing Explainable Fake Image Detection via Large Multimodal Models

Yixuan Li, Xuelin Liu, Xiaoyang Wang, Bu Sung Lee, Shiqi Wang, Anderson Rocha and Weisi Lin
IEEE transactions on information forensics and security, v 20, pp 8730-8745
2025
DOI: https://doi.org/10.1109/TIFS.2025.3597211

Abstract

Artificial intelligence benchmark Cognition explainability Faces fake image detection Forgery Generators Hair image forensics Large multimodal models Natural languages Skin Urban areas Visualization
The ability to distinguish whether an image is generated by artificial intelligence (AI) is a crucial ingredient in human intelligence, usually accompanied by a complex and dialectical forensic and reasoning process. However, current fake image detection models and databases focus on binary classification without understandable explanations for the general populace. This weakens the credibility of authenticity judgment and may conceal potential model biases. Meanwhile, large multimodal models (LMMs) have exhibited immense vision-language capabilities on various tasks, bringing the potential for explainable fake image detection. Therefore, we pioneer the probe of LMMs for explainable fake image detection by presenting a multimodal database encompassing descriptions of textual authenticity, the FakeBench. For construction, we first introduce a fine-grained taxonomy of generative visual forgery concerning human perception, based on which we collect forgery descriptions in human natural language with a human-in-the-loop strategy. FakeBench examines LMMs with four evaluation criteria: detection, reasoning, explanation and fine-grained forgery analysis, to obtain deeper insights into image authenticity-relevant capabilities. Experiments on various LMMs confirm their merits and demerits in different aspects of fake image detection tasks. This research presents a paradigm shift towards transparency for the fake image detection area and reveals the need for greater emphasis on forensic elements in visual-language research and AI risk control. FakeBench will be available at https://github.com/Yixuan423/FakeBench

Metrics

1 Record Views
3 citations in Web of Science

Details

Logo image