GenShield: Unified Detection and Artifact Correction for AI-Generated Images

Diffusion-based image synthesis has made AI-generated images (AIGI) increasingly photorealistic, raising urgent concerns about authenticity in applications such as misinformation detection, digital forensics, and content moderation. Despite the substantial advances in AIGI detection, how to correct detected AI-generated images with visible artifacts and restore realistic appearance remains largely underexplored. Moreover, few existing work has established the connection between AIGI detection and artifact correction. To fill this gap, we propose GenShield, a unified autoregressive framework that jointly performs explainable AIGI detection and controllable artifact correction in a closed loop from diagnosis to restoration, revealing a mutually reinforcing relationship between these two tasks. We further introduce a Visual Chain-of-Thought based curriculum learning strategy that enables self-explained, multi-step "diagnose-then-repair" correction with an explicit stopping criterion. A high-quality dataset with large-scale "artifact-restored" pairs is also constructed alongside a unified evaluation pipeline. Extensive experiments on our correction benchmark and mainstream AIGI detection benchmarks demonstrate state-of-the-art performance and strong generalization of our method.

AI-generated images are becoming increasingly realistic, which makes it harder to judge whether a picture is authentic. This matters for news verification, digital forensics, online content moderation, and public trust. Existing methods mainly focus on detecting whether an image is AI-generated, but they rarely explain the visible problems in the image or try to fix them. We introduce GenShield, a system that links detection with repair. It identifies suspicious visual clues, explains what may look unnatural, and then uses this information to improve the image step by step until no further repair is needed. We also build a large dataset of flawed and repaired images and a unified evaluation process. Experiments show that GenShield can both detect AI-generated images accurately and reduce visible artifacts across different types of images.