Recent advances in AI-generated image models significantly impact the digital art domain, raising concerns about the misuse of AI-generated imagery. To address the limitations of existing detection approaches, which lack pixel-level localization and robustness to diverse generation methods, we introduce CLIP-SegFusion, a multitask learning framework for synthetic image detection. Built upon the CLIP model for feature extraction, CLIP-SegFusion jointly performs image-level classification and pixel-level segmentation by integrating a segmentation module and an attention-guided feature fusion mechanism. This enables sophisticated authenticity verification and explicit localization of the manipulated regions. In our extensive experiments, CLIP-SegFusion demonstrates strong performance, achieving an average F1 score of \(92.63\%\) on the in-distribution (ID) dataset and \(83.66\%\) on the out-of-distribution (OOD) datasets, while outperforming existing state-of-the-art approaches on OOD benchmarks by a margin of \(5.80\%\). The experimental results further demonstrate that multi-scale feature modeling and the joint optimization of classification and segmentation objectives significantly enhance the robustness and generalization capability of the model in complex artistic image scenarios. Our code is available at: https://github.com/dxlabskku/CLIP-SegFusion.

Previous Click-Through Rate (CTR) prediction models rely on enumerating high-order feature combinations up to a fixed order, limiting expressiveness and scalability. Recent studies explored arbitrary-order interaction modeling through two major paradigms: log-based and graph-based methods. However, both paradigms suffer from inherent weaknesses: log-based methods lack stability, and graph-based methods lack generalizability, as both attempt to model overly diverse combinations of features, many of which may be noisy or redundant. This observation provokes a central question: What if only a small set of core interactions is sufficient? To explore this, we progressively mask feature interactions and find that removing up to 90% of them results in negligible performance degradation. This suggests that most interactions are unnecessary. Motivated by this finding, we propose SCV: Selectively Crossing Vectors, a CTR prediction framework that reformulates feature interaction learning as a sparse edge selection task over a globally shared feature-interaction graph. By modeling feature interactions over a globally learned graph and dynamically fusing expert outputs in an instance-aware manner, the SCV effectively leverages global consistency and local adaptability. We further introduce a label-biased self-distillation objective to mitigate the effects of noisy supervision and stabilize training. Experiments on public CTR benchmarks show that SCV achieves state-of-the-art performance while reducing computational cost by up to 66%, validating the effectiveness of globally sparse yet locally adaptive interaction modeling. All codes are available at: https://github.com/bw-99/scv.

Photo restoration technology enables preserving visual memories in photographs. However, physical prints are vulnerable to various forms of deterioration, ranging from physical damage to loss of image quality, etc. While restoration by human experts can improve the quality of outcomes, it often comes at a high price in terms of cost and time for restoration. In this work, we present the AI-based photo restoration framework composed of multiple stages, where each stage is tailored to enhance and restore specific types of photo damage, accelerating and automating the photo restoration process. By integrating these techniques into a unified architecture, our framework aims to offer a one-stop solution for restoring old and deteriorated photographs. Furthermore, we present a novel old photo restoration dataset because we lack a publicly available dataset for our evaluation.