Remus Brad

Organizations face escalating cyber risk, expanding attack surfaces, increasingly automated adversaries, and constrained security resources. Organizations are looking for practical mechanisms to improve security resilience by transforming threat modeling from a periodic design activity into a continuous, evidence-driven decision process. This paper offers a snapshot of the literature review of the threat modeling for composable architectures, shows why automation is difficult in this context, and proposes an automation framework to allocate scarce resources according to risk exposure to composable architecture components, where applications, services, identities, data flows, and autonomous agents are assembled and reconfigured across distributed environments. Composable architecture shows in an amplified way the gap between the static and dynamic security approaches, and our proposal helps to define the attributes needed for setting automation boundaries at the service level for risk prioritization based on threat modeling for security remediation actions. The conceptual framework integrates Zero Trust principles, control-effectiveness measurement, and human-in-the-loop governance and examines how automation with artificial intelligence changes the threat landscape by introducing risks that are difficult to measure and fast-changing. The results show that automation should be controlled with defined boundaries explained through measurable attributes for transparent decisions. It also proposes that AI should not replace expert judgement; rather, it should augment security teams by improving information quality, revealing hidden dependencies, supporting adaptive prioritization under uncertainty, and enabling resilience-oriented investment decisions for composable, distributed, and increasingly autonomous systems.

This paper provides a comprehensive analysis of several neural network architectures designed for the detection of generative image inpainting. The central challenge addressed is the increasing sophistication of inpainting technologies, driven by generative AI, which renders traditional forensic methods based on simple statistical or visual artifacts obsolete. This paper investigates a feature-centric hypothesis: that visually seamless inpainting manipulations introduce consistent, subtle, and detectable artifacts in specific high-frequency and statistical domains. The primary domain of interest is the feature space defined by the Dual-Tree Complex Wavelet Transform (DTCWT). The core of this report details the methodology and evaluation of six novel deep learning architectures, each custom-designed to operate on these wavelet-domain features. These architectures explore diverse design philosophies, including dense feature aggregation (Custom UNet++), explicit global context integration (UNet++ with Global Average), recurrent spatial modeling (ConvLSTM), unsupervised generative anomaly detection (Variational Autoencoder), and channel-specific processing (StackDeepAll). A rigorous experimental evaluation was conducted on a custom inpainting detection dataset. The quantitative results demonstrate the clear superiority of the StackDeepAll architecture. This model, which processes each of the 12 DTCWT channels through independent, parallel UNet-like encoders before fusion, achieved a state-of-the-art Intersection over Union (IoU) of 0.8126 in validation. This result significantly surpassed all other proposed models, including a stable Variational Autoencoder (0.3711 IoU) and overfitting-prone ConvLSTM models (0.4514 IoU). The findings establish that for high-dimensional, engineered features like complex wavelets, a channel-specific processing paradigm that avoids premature feature-blending is a powerful and highly effective design strategy. This feature-centric approach is validated as a potent alternative to contemporary end-to-end spatial-domain models.

In the current digital era, the spread of fake news presents serious difficulties. This study offers a thorough analysis of recent developments in false news automatic detection techniques, from traditional methods to the most recent developed models like large language models. The review identifies four perspectives on automatic detection of fake news that are oriented towards knowledge, style, propagation, and source of the misinformation. This paper describes how automatic detection methods use data science techniques such as deep learning, large language models, and traditional machine learning. In addition to discussing the shortcomings of existing approaches, such as the absence of datasets, this paper emphasizes the multidimensional function of large language models in creating and identifying fake news while underlining the necessity for textual, visual, and audio common analysis, multidisciplinary collaboration, and greater model transparency.

In today’s era of technology, where information is readily available anytime and from anywhere, safeguarding our privacy and sensitive data is more important than ever. The thermal sensors embedded within a CPU are primarily utilized for monitoring and regulating the processor’s temperature during operation. However, they can serve as valuable components in increasing the security of a system as well, by enabling the detection of anomalies through temperature monitoring. This study presents three distinct methods demonstrating that anomalies in CPU heat dissipation can be effectively detected using the thermal sensors of a CPU, even under conditions of significant environmental use. First, it evaluates the Hot-n-Cold anomaly detection technique across various noisy environments, demonstrating that the presence of additional lines of code inserted into a Linux command can be identified through thermal analysis. Second, it detects the CryptoTrooper ransomware attack by fingerprinting the associated cryptographic processes in terms of temperature. Finally, it detects unauthorized system login attempts by capturing and analyzing their distinctive thermal signatures. This study demonstrates that various detection mechanisms can be implemented using thermal sensors to enhance system security. It also motivates the need for further research in this relatively underexplored area with the goal of developing more effective methods of protecting data.

Employing machine learning algorithms in the medical field has proven successful for some time now. Mostly computer vision techniques have been applied to medical images, while medical sound data has been somewhat overlooked. By using electronic stethoscopes, it is now possible to process both heartbeats and lung sounds. While some products are available for detecting anomalies in heartbeats, addressing lung-related anomalies presents a more intricate challenge. Applying a deep learning approach is hindered by insufficient data. Although some datasets do exist, the size and diversity of the data are too small for comprehensive analysis. This paper introduces a novel technique for detecting anomalies in lung sounds: first by combining two datasets, second by automatically segmenting each sound into respiratory cycles, and third by employing GFCCs as sound features.

As technology continues to evolve, it provides significant benefits, such as seamless access to information, while also introducing critical security challenges, including privacy vulnerabilities and the rapid spread of misinformation. One major class of security threats involves software-based attacks, divided in side-channel attacks, which exploit unintended information leakage, and covert-channel attacks, which establish unauthorized communication pathways to bypass security policies. This survey provides a comprehensive review of the most cited attacks of these types in the past decade, along with their associated countermeasures. We classify the attacks and countermeasures based on the targeted system component and fundamental attack mechanisms. In addition, we conduct a comparative analysis of the attacks, considering key aspects such as attack methodology, targeted algorithms, victim system environment, reliability, performance impact, and mitigation strategies. Furthermore, we examine defense mechanisms, assessing their effectiveness, computational overhead, and applicability across various attack scenarios. Finally, we discuss emerging trends in attack techniques and countermeasures, highlighting open challenges and future research directions to secure modern computing systems.

The rapid advancement of image inpainting tools, especially those aimed at removing artifacts, has made digital image manipulation alarmingly accessible. This paper proposes several innovative ideas for detecting inpainting forgeries based on a low-level noise analysis by combining Dual-Tree Complex Wavelet Transform (DT-CWT) for feature extraction with convolutional neural networks (CNN) for forged area detection and localization, and lastly by employing an innovative combination of texture segmentation with noise variance estimations. The DT-CWT offers significant advantages due to its shift-invariance, enhancing its robustness against subtle manipulations during the inpainting process. Furthermore, its directional selectivity allows for the detection of subtle artifacts introduced by inpainting within specific frequency bands and orientations. Various neural network architectures were evaluated and proposed. Lastly, we propose a fusion detection module that combines texture analysis with noise variance estimation to give the forged area. Also, to address the limitations of existing inpainting datasets, particularly their lack of clear separation between inpainted regions and removed objects—which can inadvertently favor detection—we introduced a new dataset named the Real Inpainting Detection Dataset. Our approach was benchmarked against state-of-the-art methods and demonstrated superior performance over all cited alternatives.

This study investigates the effectiveness of a proposed version of Meta’s LLaMA 3 model in detecting fake claims across bilingual (English and Romanian) datasets, focusing on a multi-class approach beyond traditional binary classifications in order to better mimic real-world scenarios. The research employs a proposed version of the LLaMA 3 model, optimized for identifying nuanced categories such as “Mostly True” and “Mostly False”, and compares its performance against leading large language models (LLMs) including Open AI’s ChatGPT versions, Google’s Gemini, and similar LLaMA models. The analysis reveals that the proposed LLaMA 3 model consistently outperforms its base version and older LLaMA models, particularly in the Romanian dataset, achieving the highest accuracy of 39% and demonstrating superior capabilities in identifying nuanced claims, over all the compared large language models. However, the model’s performance across both languages highlights some challenges, with generally low accuracy and difficulties in handling ambiguous categories by all the LLMs. The study also underscores the impact of language and cultural context on model reliability, noting that even state-of-the-art models like ChatGPT 4.o and Gemini exhibit inconsistencies when applied to Romanian text and more than a binary true/false approach.