UniGenDet: A Unified Generative-Discriminative Framework for Co-Evolutionary Image Generation and Generated Image Detection
2026-04-23 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors developed UniGenDet, a new system that combines making images and detecting fake images into one process. They created special ways for these two tasks to share information and help each other get better. Their method helps make more realistic images while also improving how well fake images can be detected. Tests showed their approach works very well compared to others.
image generationimage detectiongenerative networksdiscriminative frameworksself-attentionmultimodal learningfine-tuningadversarial informationgenerative alignmentauthenticity identification
Authors
Yanran Zhang, Wenzhao Zheng, Yifei Li, Bingyao Yu, Yu Zheng, Lei Chen, Jiwen Lu, Jie Zhou
Abstract
In recent years, significant progress has been made in both image generation and generated image detection. Despite their rapid, yet largely independent, development, these two fields have evolved distinct architectural paradigms: the former predominantly relies on generative networks, while the latter favors discriminative frameworks. A recent trend in both domains is the use of adversarial information to enhance performance, revealing potential for synergy. However, the significant architectural divergence between them presents considerable challenges. Departing from previous approaches, we propose UniGenDet: a Unified generative-discriminative framework for co-evolutionary image Generation and generated image Detection. To bridge the task gap, we design a symbiotic multimodal self-attention mechanism and a unified fine-tuning algorithm. This synergy allows the generation task to improve the interpretability of authenticity identification, while authenticity criteria guide the creation of higher-fidelity images. Furthermore, we introduce a detector-informed generative alignment mechanism to facilitate seamless information exchange. Extensive experiments on multiple datasets demonstrate that our method achieves state-of-the-art performance. Code: \href{https://github.com/Zhangyr2022/UniGenDet}{https://github.com/Zhangyr2022/UniGenDet}.