Deepfake technologies pose a growing threat to the Internet of Things (IoT), enabling identity spoofing and the spread of misinformation. Although numerous face forgery detectors have been developed to counter these risks, their realworld deployment is often limited by inherent dataset biases. Existing domain adaptation techniques offer potential remedies, but typically rely on access to raw source data and employ data-dependent alignment strategies under a transductive learning paradigm, raising substantial privacy concerns for source domain individuals. This study revisits the problem from the perspective of black-box domain adaptation and introduces a detection framework that leverages only the predictions from the source model. The method is grounded in a local relation-guided AUC optimization strategy, which leverages the robustness of AUC-based objectives in noisy environments while addressing the limitations of conventional AUC optimization, particularly its vulnerability to confirmation bias and reliance on a fixed decision threshold. To this end, two key components are introduced. First, a nearest-neighbor calibration mechanism is presented, where the local relation feature (LRF), a parameter-free representation, captures differences between real and fake images without favoring specific forgery types, thereby reducing bias inherited from the source model. Second, a GMM-based adaptive thresholding scheme is employed to address asymmetric predictions by dynamically determining the optimal decision boundary for real and fake images. Experiments across multiple datasets show that our approach achieves superior generalization compared to state-of-the-art methods. Moreover, the framework is compatible with a broad range of source and target model configurations to enhance detection performance.

Deepfake technology poses a significant threat to the Internet of Things by enabling identity spoofing and the dissemination of misinformation. Although numerous face forgery detectors have been developed to counter the risks of facial deepfakes, detecting forgeries across varying quality levels, particularly under extreme degradations, remains a critical challenge. Current face forgery detection methods largely rely on identifying low-level artifacts, which are highly susceptible to distortions introduced by image degradation. Recognizing that restoration can recover critical forensic cues from severely degraded forgeries, this study proposes a quality-agnostic deepfake detection framework that leverages a blind face restoration model to enhance robustness. The framework incorporates an auxiliary restoration branch alongside the original detection pathway. While the original branch operates directly on the degraded input, the restoration branch performs detection on facial images restored by a blind face restoration model. In addition, a Saliency-Guided Restoration objective is introduced to enhance alignment between the restoration and detection tasks. A Restoration Similarity-Aware Fusion mechanism is further designed to adaptively integrate predictions from both branches based on input quality. To assess the robustness of our approach under extreme degradation, we establish a specialized, real-world-inspired benchmark that simulates diverse degradation scenarios. Comprehensive experiments on both the proposed and existing benchmarks demonstrate that our method consistently achieves superior robustness in various degradation scenarios.

Deep neural networks are highly susceptible to adversarial attacks. Although adversarial training is an effective defense, its efficacy in long-tailed settings remains limited. Current methods are constrained in long-tailed scenarios by early representation bias, imbalanced optimization, and insufficient class-aware adaptivity. To address these challenges, we propose Decoupled-Curriculum Adversarial Robustness Learning, a two-stage framework. The Initial Representation and Balancing stage employs inter-class margin adjustment via LDAM with a Deferred Re-weighting schedule to build balanced, transferable representations. The Adaptive Correctness-Aware Robustness Learning stage adapts the per-class PGD perturbation budget using class-wise correctness signals and optimizes a balanced loss that combines mean, medium-class protection, and tail-class reinforcement components, aligning robust learning across head, medium, and tail classes. Experiments on standard long-tailed benchmarks validate the effectiveness of our approach, yielding consistent improvements in both natural and robust performance.

The rapid progress of generative adversarial networks (GANs) and diffusion models (DMs) has enabled photorealistic image synthesis, raising critical concerns about image authenticity verification and source attribution. Existing hierarchical detection frameworks remain constrained by error propagation, weak cross-layer information sharing, and limited discriminative power for highly similar architectures. To address these challenges, we propose a multi-feature adaptive hierarchical framework for fine-grained attribution of AI-generated images. The framework first unifies complementary representations from frequency, noise, color, and semantic domains through a multi-feature extraction module, providing richer cues beyond conventional designs. We then introduces adaptive mechanisms—including dynamic decisionmaking, confidence-weighted fusion, and dual-path classification—that mitigate cascading errors and enable flexible dependency control across layers. Extensive experiments on a large-scale dataset validate the effectiveness of our approach, showing clear improvements over conventional methods and state-of-the-art detectors, particularly in distinguishing confusable GAN variants such as ProGAN, StyleGAN, and StyleGAN2. These results demonstrate the robustness and scalability of the proposed design for reliable AIGC detection and model attribution.