Backdoor attacks inject hidden behaviors into deep neural networks, causing them to behave normally on benign inputs but produce attacker-specified results under trigger conditions, which poses significant threats to model integrity. However, transfer learning, as a prevalent paradigm for training neural networks, often compromises backdoors during the fine-tuning process. To address this issue, we propose Robust Backdoor Attacks against Fine-tuning-based Transfer Learning via Self-Attention Distillation (RBA-SAD), a robust backdoor attack scheme that manipulates model’s ability to recognize the trigger via layer-wise and top-down self-attention distillation, effectively withstanding the impact of fine-tuning. Additionally, imperceptible noise is employed as a trigger to evade defenses based on input purification. To enhance the effectiveness of self-attention distillation, we optimize trigger generation to achieve multi-layer feature collisions, ensuring that the representations of poisoned samples at each corresponding layer are similar to those of clean samples. Extensive experiments on three benchmark transfer-learning tasks show that RBA-SAD achieves over 97% attack success rate while remaining stealthy against seven state-of-the-art backdoor defenses. Moreover, it generalizes well to additional settings, including Vision Transformers, ImageNet-scale transfers, and parameter-efficient fine-tuning with Low-Rank Adaptation.
Behavioral profiling is widely used to characterize patterns in user activity, typically through clustering techniques that group similar behaviors. However, existing research often overlooks the role of feature engineering and assumes the availability of extensive historical data, assumptions that rarely hold in real-world settings. This study presents a case study on behavioral profiling using Microsoft Entra ID sign-in logs from a real-world cloud environment, and provides a systematic analysis of key design choices in profiling. In particular, we investigate how domain knowledge in feature engineering and the choice of data scope affect the quality and interpretability of behavioral profiles. Profiles are evaluated using both quantitative metrics and expert validation. Our results demonstrate that domain-informed features improve clustering quality and profiling accuracy. Moreover, profiles constructed from all available data remain distinctive across user groups, even under limited data availability, highlighting the robustness of the approach in realistic scenarios. Expert evaluation further shows that interpretability depends on the consistency and abstraction level of features. Moreover, the resulting profiles expose behavioral patterns that extend beyond experts’ typical user-centered analyses, providing a more comprehensive view of authentication behavior. Overall, our findings demonstrate that data-driven profiling, supported by expert validation, enables the construction of interpretable behavioral representations in cloud authentication environments.
With the rapid growth of network traffic, data has become increasingly complex and voluminous, posing significant challenges for accurate anomaly detection. Traditional deep learning approaches often fail to capture the rich interdependencies and heterogeneous nature of traffic features, limiting their effectiveness in identifying subtle or evolving abnormal patterns. To address these challenges, this paper proposes GS-HF (GraphSAGE with heterogeneous features), an anomaly detection framework. The method integrates both statistical and image-like features extracted from raw traffic data to capture multi-dimensional characteristics, and then constructs a graph based on the fused heterogeneous features to better represent relationships among traffic flows. An improved GraphSAGE model is applied for detection, incorporating Focal Loss to handle class imbalance in real-world anomaly datasets. Extensive experiments on multiple network traffic datasets demonstrate that GS-HF achieves superior detection performance compared to existing methods, highlighting its robustness and effectiveness in handling diverse and complex traffic patterns.
As privacy violations and data misuse on social media become more prevalent, understanding the drivers of effective privacy protection is critical. This study addresses the privacy paradox, the contradiction wherein users express high privacy concerns but fail to adopt protective behaviors. It examines how awareness of online identity disclosure influences two key protective actions: controlled use and identity masking, with a specific focus on the moderating role of online privacy concerns. Drawing on a quota-based sample of 1000 Pakistani social media users and employing structural equation modeling, the research finds that while greater awareness generally promotes proactive behaviors; however, this relationship is significantly moderated by privacy concerns. At lower concern levels, awareness strongly predicts protection; however, as concerns intensify, the positive effect of awareness significantly weakens, suggesting disengagement or privacy-fatigue. These insights clarify the conditional role of concerns and highlight the need for tailored privacy education and platform design.
With the rise and advancements of technology, digital security is facing unprecedented challenges. Traditional security measures, such as strong credentials and multi-factor authentication (MFA), are no longer sufficient to protect digital assets from attacks and unauthorized access or to effectively verify user authenticity. Zero trust architecture (ZTA) has emerged as a solution in response to these issues. Leading organizations, including Google, Forrester, Palo Alto, and CISCO, have initiated strategies to design their own deployable ZTAs to offer the perks of ZT to other organizations. It operates on various principles such as removing implicit trust, continuously authenticating users, and de-perimeterization. It provides a top-notch defense against various major cyberattacks on organizations. This study offers a comprehensive and multi-dimensional synthesis of ZTAs. It also examines the principles and implementation strategies of various ZTAs, comparing them using multiple parameters to highlight their strengths and weaknesses. In addition to the synthesis of ZTAs, the work demonstrates its practical value with sector-specific examples to make it more relevant for real-world applications. Through a detailed analysis, the study identified potential research gaps and areas in existing research that require further investigation. These insights can guide future developments in digital security. The findings aim to help organizations select and implement the most effective ZT strategies to enhance their cybersecurity posture.
