ABSTRACT As a fundamental task in restoring continuous visual signals, image inpainting plays a critical role in autonomous driving perception, medical imaging, video editing and digital heritage preservation. Driven by deep learning and large‐scale generative models, the field has transitioned from low‐level texture synthesis to high‐level semantic generation, yielding major breakthroughs in structural fidelity and visual realism. Centring on the generative paradigm as the architectural trajectory, this survey systematically categorizes the 30‐year evolution of image inpainting into three distinct technological generations: traditional prior‐driven synthesis, deep learning data‐driven reconstruction and modern foundation model‐driven generation. Despite this progress, highly competitive methods still struggle with large‐scale missing regions, global consistency in complex scenes, fine‐grained micro‐details and alignment with human visual perception. To address these gaps, we critically evaluate the technical paradigms and main bottlenecks within each of these evolutionary stages. We categorize and compare mainstream breakthroughs across high‐resolution restoration, text‐guided synthesis and complex scene generation. Furthermore, we compile standard benchmarks, evaluation metrics and quantitative performance comparisons of representative algorithms. Finally, we dissect open challenges—focusing on cross‐scene generalization and evaluation metric alignment—and outline future trajectories, particularly the integration of inpainting with text‐guided foundation models, providing a definitive reference for future theoretical and engineering advancements.

To defend against adversarial structural attacks on graphs, we analyze attacks through the lens of mutual information and discover the “pairwise effect". This effect reveals that structural attacks effectively degrade the performance of victim GNNs when these GNNs receive the modified structure paired with the given node attributes as training input. Therefore, we propose a novel defense strategy that renders structural attacks ineffective by disrupting the pairing of modified structures and node attributes during the training of victim GNNs, which we call “disrupting the pairwise effect". To implement this idea, we propose two simple yet effective training strategies: Structural Fine-Tuning (SF) and Progressive Structural Training (PST), which disrupt the pairwise effect through node attributes pre-training followed by structure fine-tuning and progressive structure training, respectively. Compared to existing robust GNNs, our strategies avoid time-consuming techniques, thereby improving the robustness of GNNs while enhancing training speed. Additionally, these strategies can be easily applied to a wide range of commonly used GNNs, including robust GNN variants, making them highly adaptable to different models and applications. We provide theoretical analysis of the proposed training strategies and conduct extensive experiments on various datasets to demonstrate their effectiveness. Datasets and codes of this paper are available at https://github.com/Xing-Ai1003/Revisiting-Adversarial-Robustness-of-GNNs.

Dynamic searchable encryption (DSE) with forward and backward privacy reduces leakages in early-stage schemes. Security enhancement comes with a price - maintaining updatable keyword-wise state information. State information, if stored locally, incurs significant client-side storage overhead for keyword-rich datasets, potentially hindering real-world deployments.

Video-based services have become popular. Clients often outsource their videos to the cloud to relieve local maintenance. However, privacy has become a major concern, since many videos contain sensitive information. Although retrieving (unencrypted) videos has been extensively investigated, retrieving encrypted multimedia has received relatively rare attention, at best in a limitation of image-based similarity searches. We initiate the study of scalable encrypted video search, enabling clients to query videos similar to an image search. Our modular framework leverages intrinsic attributes of videos, such as semantics and visuals, to effectively capture their contents. We propose a two-step approach whereby lightweight searchable encryption techniques are used for pre-screening, followed by an interactive approach for fine-grained search. Furthermore, we present three instantiations, including one centralized-writer instantiation and two distributed-writer instantiations, to effectively cater to varying needs and scenarios: (1) The centralized one employs forward and backward private searchable encryption [CCS 2017] over deep hashing [CVPR 2020]. (2) Motivated by distributed computing, the multi-writer instantiations building atop HSE [Usenix Security 2022] allows searching the relevant videos contributed by multiple intuitions collaboratively. Our experimental results illustrate their practical performance over multiple real-world datasets, whether in a centralized setting or distributed setting.

Softmax and sigmoid, composing exponential functions (ex) and division (1/x), are activation functions often required in training. Secure computation on non-linear, unbounded 1/x and ex is already challenging, let alone their composition. Prior works aim to compute softmax by its exact formula via iteration (CrypTen, NeurIPS ’21) or with ASM approximation (Falcon, PoPETS ’21). They fall short in efficiency and/or accuracy. For sigmoid, existing solutions such as ABY2.0 (Usenix Security ’21) compute it via piecewise functions, incurring logarithmic communication rounds.

Federated learning (FL) typically faces data heterogeneity, i.e., distribution shifting among clients. Sharing clients' information has shown great potentiality in mitigating data heterogeneity, yet incurs a dilemma in preserving privacy and promoting model performance. To alleviate the dilemma, we raise a fundamental question: Is it possible to share partial features in the data to tackle data heterogeneity? In this work, we give an affirmative answer to this question by proposing a novel approach called **Fed**erated **Fe**ature **d**istillation (FedFed). Specifically, FedFed partitions data into performance-sensitive features (i.e., greatly contributing to model performance) and performance-robust features (i.e., limitedly contributing to model performance). The performance-sensitive features are globally shared to mitigate data heterogeneity, while the performance-robust features are kept locally. FedFed enables clients to train models over local and shared data. Comprehensive experiments demonstrate the efficacy of FedFed in promoting model performance.

Video-based services have become popular. Clients often outsource their videos to the cloud to relieve local maintenance. However, privacy has emerged as a major concern since many videos contain sensitive information. While retrieving (unencrypted) videos has been widely studied, encrypted multimedia retrieval receives rare attention, at best in a limited form of similarity searches on images.