Defense Strategies Research Articles

Research on adversarial attack and defense of large language models

Abstract. Large language models (LLMs) have made excellent progress in text and image understanding and generation. However, with the wide range of applications of these models in various industries, the issue of their security, especially the defense against adversarial attacks, has become a focus of research. This study focuses on exploring the adversarial attacks faced by LLMs and their defense strategies, especially the design and optimization of defense mechanisms. Through literature review and case studies, this paper analyzes in detail the white-box and black-box attack patterns against LLMs, including model inversion, backdoor attacks, and token-based strategies. In response to these attacks, this paper proposes a series of defense strategies, including preventive measures such as data augmentation, adversarial training and model regularization, as well as real-time attack detection and response strategies such as anomaly detection and adversarial sample detection techniques. The core of this research is to improve the robustness and trustworthiness of LLMs, providing the necessary guarantees for their integration and sustainability in multiple industrial applications. In addition, this paper proposes future research directions, highlighting the importance of developing advanced defense systems, promoting interdisciplinary research and exploring new applications for LLMs. This research provides valuable insights into understanding and improving the security defense mechanisms of LLMs, which is essential for maintaining the security and user trust of these models.

Role of mucous cells in the defence system of the cnidarian Actinia equina (Linnaeus, 1758): The keystone of evolution

AbstractCnidarians are the first phylum of metazoans to exhibit a functional mucus layer. In evolution, mucous cells perform the classical secretory function of mucus and act as sentinels for external insults, particularly in mammals. However, this topic has not been much analysed in the lower metazoans. This study aims to investigate the role of mucous cells in Actinia equina (Anthozoa, Hexacorallia) (Linnaeus, 1758) by using confocal microscopy and bioinformatics. The results show colocalization of Toll‐like receptor 2 (TLR2) and Mucin2 (MUC2) on these cells, suggesting a crosstalk between defence response and mucus secretion in cnidarians. Furthermore, the presence of TLR2 on the surface of mucous cells suggests their involvement in host defence responses, as occurs in vertebrates. Moreover, through in silico analysis, we confirmed the presence of orthologous genes for the tested antibodies, providing additional evidence to support our results from an evolutionary perspective. These data suggest that mucous cells play a defensive role throughout evolution, helping us understand how these cells have developed and adapted across phylogenesis. This information also provides new insights into the defensive strategies of invertebrates and clarifies their evolutionary history.

The prophet on trial: Baptist dissenters before the Soviet courts, 1960s–1980s

ABSTRACT This article explores the courtroom behaviour of adherents of the oppositional dissenter wing within the Evangelical Baptist community in the Soviet Union, who faced trials for violating Soviet religious laws between the 1960s and 1980s. The analysis reveals that the accused believers developed a nuanced defence strategy encompassing three core elements: mobilizing support from their fellow believers in courtrooms, constructing communicative patterns that involved drawing prosecutors and judges into theologically driven dialogues about truth, and employing distinctive rhetoric aimed at questioning the legal foundations and legitimacy of Soviet religious policy. The accused Baptists adopted a marked confrontational stance towards state officials in charge of criminal prosecutions and trials, striving to overcome them in argument. Ultimately, the courtrooms became a platform for them to voice and promote their faith-based views on state politics and freedom.

When the "Matchmaker" Does Not Have Your Interest at Heart: Perceived Algorithmic Harms, Folk Theories, and Users' Counter-Strategies on Tinder

On online platforms, algorithms help us build and manage our relationships. However, their invisible interventions might also pose harm to these connections. Dating platforms offer a prime example where, despite extensive research on human-inflicted harm, the potential harm from the algorithms themselves, and user strategies for mitigating them, remains largely unexplored. In our analysis of 7,043 reviews and interviews with 30 Tinder users, we unveiled how users perceive algorithmic harm as damaging self-esteem, sabotaging potential relationships, encouraging antisocial behavior, and misrepresenting or marginalizing certain identities. We introduce a new algorithmic folk theory, the "conflict of interest" theory, perceived to perpetuate these harms. This theory encapsulates users' sense of a contradiction between the dating platform's promise of finding the perfect partner (leading to discontinued use of Tinder) and its commercial interest in retaining users to increase revenue. Users suspected various algorithmic processes pursuant to this theory, such as (a) throttling profile visibility, (b) manipulating users' matches, and (c) recommending large quantities of profiles that will not lead to matches. They also described various strategies in resistance or defense of these suspected algorithmic processes, such as engaging in counter-intuitive behaviour to disrupt the unfavorable algorithmic processes or leveraging location based filtering for match variety and safety. We conclude by discussing how the perceived algorithmic harms can inform the development of new algorithmic implementations that balance both user and company interests.

Analysis of War Optimization Algorithm in a Multi-Loop Power System Based on Directional Overcurrent Relays

In electrical power systems, ensuring a reliable, precise, and efficient relay strategy is crucial for safe and trustworthy operation, especially in multi-loop distribution systems. Overcurrent relays (OCRs) have emerged as effective solutions for these challenges. This study focuses on optimizing the coordination of OCRs to minimize the overall operational time of main relays, thereby reducing power outages. The optimization problem is addressed by adjusting the time multiplier setting (TMS) using the War Strategy Optimization (WSO) algorithm, which efficiently solves this constrained problem. This algorithm mimics ancient warfare strategies of attack and defense to solve complex optimization problems efficiently. The results show that WSO provides superior performance in minimizing total operating time and achieving global optimum solutions with reduced computational effort, outperforming traditional optimization methods (i.e., SM, HPSO, GA, RTO, and JAYA). The proposed algorithm shows a net time gains of 7.77 s, 2.57 s, and 0.8484 s when compared to GA, RTO, and JAYA respectively. This robust protection coordination ensures better reliability and efficiency in multi-loop power systems.

A Comprehensive Survey on Cyber-Physical System Security in the Internet of Things (IoT): Addressing and Solutions

Background: Cyber-Physical Systems (CPSs) integrate computing, control, and communication technologies, bridging cyberspace and the physical world to enhance critical infrastructure and safety-critical systems. Existing surveys often address CPS security from a single perspective, necessitating a more comprehensive approach. Methods: This paper presents a thorough review of CPS security from three perspectives: the physical domain, the cyber domain, and the cyber-physical domain. We examine attacks on physical components like sensors, cyber-attacks targeting CPSs, and integrated cyber-physical attacks. For each domain, we analyse corresponding detection and defence mechanisms. Results: Our review reveals that CPSs face significant security threats across all domains. In the physical domain, attacks on sensors can disrupt system operations, but various defences are available. In the cyber domain, CPSs are vulnerable to malware, hacking, and denial-of-service attacks, with several detection and defence strategies in place. The cyber-physical domain highlights complex threats where cyber-attacks cause physical damage, requiring comprehensive security approaches. Conclusion: By examining CPS security from multiple perspectives, this review provides a holistic understanding of current threats and defence mechanisms. It identifies future research directions to enhance CPS security, aiming to better protect critical infrastructure against evolving cyber threats.

Watch Out for the Inherent Vulnerabilities in Developing Multi-tenant Cloud-FPGA: Communication Protocols

As FPGAs are being deployed in the cloud infrastructure for acceleration, the technology of multi-tenant FPGA has emerged as a topic of interest. This development has drawn considerable attention to its security issues. While previous research primarily focused on the security of applications, there has been limited exploration of the vulnerabilities inherent in FPGA IPs. In our work, we examine the vulnerabilities of two widely used data transmission protocols in modern FPGAs: the Advanced eXtensible Interface (AXI) and Peripheral Component Interconnect Express (PCIe). Our experiments, conducted with commercial FPGA development kits, launched fault injection attacks through the shared power distribution network (PDN). Through non-invasive electromagnetic (EM) trace measurement, we characterize the voltage fluctuation across various attack patterns. Subsequently, we simulate real-world data transfers using two crafted datasets with different statistical characteristics. The experimental results demonstrate the unique security vulnerabilities of the current AXI and PCIe protocols in the context of a multi-tenant cloud-FPGA. In response to such vulnerability, we further propose two defense strategies: InChAXI that utilizes integrity checking for AXI-based data, and FCPCIe that employs frequency scaling for PCIe-based data. The performance evaluation demonstrates that our proposed defenses can significantly reduce the fault injections on the AXI-based data transmission by 705 times with small overheads – 0.5% in hardware footprint and 7.9% in latency, respectively. On the other hand, FCPCIe effectively prevents the fault injection attack during the PCIe-based data transmission by reducing the user clock frequency, while incurring a 10.13% overhead on data throughput.

Immune defense adaptation of Strauchbufo raddei population in heavy metal polluted area: Insights from developmental and environmental perspectives

The adjustment of immune defense mechanisms is a crucial aspect of biological adaptation to stressful environments. Amphibians, with their unique metamorphic process, experience distinct life stages and exhibit diverse immune defense components. While previous studies have focused on specific immune changes during particular life stages under stress, this research addresses a critical gap by exploring the adaptive immune defense strategies of Strauchbufo raddei in heavy metal-polluted environments. We conducted laboratory experiments, exposing offspring from both polluted and unpolluted areas to control and heavy metal treatments, while continuously monitoring changes in immune components during key metamorphic stages. Notably, we examined the role of the skin microbiome, a crucial but often overlooked barrier against pathogens. The results indicated that individuals from polluted areas exhibited some tolerance to heavy metal exposure, though overall immune function remained diminished. During metamorphosis, when immune defenses are most vulnerable, the skin microbiome rapidly enriched beneficial bacteria, preventing pathogenic colonization and playing a pivotal role in maintaining immune defense in contaminated environments. Moreover, our research highlights energy allocation strategies involving corticosterone and body fat content, enabling populations to maintain development despite immune compromise. The immune adaptations observed may be fixed through genetic assimilation, suggesting a rapid evolutionary response to environmental stress. However, this reduces phenotypic plasticity, making populations more vulnerable to future environmental changes. This study provides key insights into the survival strategies of amphibian populations in heavy metal-contaminated areas, laying the foundation for future research on molecular and evolutionary adaptations.

Betasatellite: A Hijacker of Plant Innate Immunity

Betasatellites associated with begomoviruses significantly exacerbate viral infections in economically important crops, prompting the evolution of various plant defense mechanisms, including RNA interference, phytohormone signalling and cellular modifications using autophagy and ubiquitination. Approximately 13.5-kDa βC1 protein has two key functions: It boosts viral replication and weakens plant defenses. By interfering with signal transduction pathways and RNA silencing, it makes the plant more vulnerable to infection. Additionally, βC1 impairs chloroplast function, complicating the plant's defense strategies. In contrast, the nuclear shuttle protein (βV1) protein, though less understood, appears to facilitate viral replication and interacts with the helper virus replication enhancer protein (REn), leading to altered localization within the cell. As only 40% of betasatellites encode the βV1 protein, further investigation into its molecular interactions and functions could provide crucial insights for developing effective antiviral strategies against geminivirus infections. This review highlights the complex interplay between betasatellites and host defenses, offering avenues for future research in crop protection.

Enhancing Cybersecurity Resilience through Advanced Red Teaming Exercises and MITRE ATT&CK Framework Integration: A Paradigm Shift in Cybersecurity Assessment

As cybersecurity threats evolve at an alarming pace, conventional defense strategies are becoming progressively less effective. Addressing this urgent challenge, our study introduces a transformative approach to red teaming exercises by integrating the MITRE ATT&CK framework. This novel integration utilizes real-world attacker tactics and behaviors to create highly realistic scenarios that rigorously test defenses and reveal previously unidentified vulnerabilities. Our comprehensive evaluation confirms a significant enhancement in the realism and effectiveness of red teaming, leading to improved vulnerability identification and the generation of actionable insights for proactive remediation. This study uniquely contributes by providing a structured, data-driven methodology that not only aligns current defenses with emergent threat tactics but also promotes continuous improvement in cybersecurity resilience. Such advancements are crucial for organizations aiming to stay ahead of attackers and maintain robust cyber defenses in dynamic threat environments. The significance of this study lies in providing a structured approach to bolster cybersecurity resilience against evolving threats and offering a model for continuous defense improvement. Future research should explore the impact of such integration on long-term cybersecurity strategies.

Integrated Defense: Combining Aspects of Social Media, Cyber and Technology in the Context of Asymmetric Warfare

Integrated defense is a core component of Indonesia's defense and security strategy, designed to protect the country's sovereignty, territorial integrity, and national safety from both physical and non-physical threats, especially cyber threats. With the rapid technological advancements and the rise of social media, cyber threats have become a significant concern as they impact various aspects of national security. Indonesia's large population of internet and social media users heightens the urgency for strategic attention to cyber defense. The government has introduced various policies and regulations, such as those by the Ministry of Defense, to strengthen cyber security and integrate it into national defense mechanisms. This study uses a qualitative descriptive approach, analyzing literature and legal frameworks to highlight the significance of integrated defense in addressing modern, asymmetric threats. In asymmetric warfare, non-military means, including cyber-attacks, are used to weaken a nation's ideological, political, economic, and social systems. The research emphasizes the need for a comprehensive and adaptive strategy, involving collaboration between the government, private sector, and civil society, to counter these complex threats. Strengthening cyber defense through education, training, and public awareness is crucial for safeguarding Indonesia's national security in the digital age.

Candida albicans infection model in Drosophila melanogaster suggests a strain-specific virulent factor boosting a stormy innate immune response

IntorductionPathogens drive the evolution of host defence strategies, with both innate and adaptive immune systems playing key roles. Priming enhances the innate immune system’s readiness by functionally reprogramming immune cells after initial exposure to stimuli, like β-glucans. In this sense, Drosophila melanogaster is a valuable model to evaluate the role of innate immunity to control infections.ObjectivesIn this study we aimed to set light on the immune priming effect of oral treatment with heat-killed M. manresensis and two different heat-killed C. albicans isolates upon systemic infection by C. albicans in the D. melanogaster model.MethodsA clinical and a control ATCC 90028 Candida albicans strain were used. Flies were primed through oral administration of heat-killed C. albicans (hkCa), both clinical and control, and hk-Mycolicibacterium manresensis. After priming, flies were systemically infected with both C. albicans isolates. Host survival, pathogen load, and immune response in response to treatment and infection were evaluated.ResultsBoth treatments showed a significant capacity to enhance the expression of antimicrobial peptides, in particular Diptericin, and Drosomycin in males. This response had a marked sexual dimorphism due to the difference in Upd3, Nox, and Duox expression. Surprisingly, even when priming was able to avoid the growth of both C. albicans strains, survival was not improved in the case of the clinical isolate, causing an unexpected mortality rate in hours, regardless of the host’s sex. Gene expression analysis 24 hours post-infection showed an exacerbated increase in Diptericin, Drosomycin and Upd3 expression upon infection with the clinical strain.ConclusionData herein suggests the presence of a strain-specific component in C. albicans as the booster of a “stormy” innate immune response, which must be further investigated, and position D. melanogaster as a useful model for evaluating virulent factors related to the modulation of the innate immunity.

Antimicrobial resistance: Biofilms, small colony variants, and intracellular bacteria

Soft tissue and bone infections continue to be a serious complication in orthopedic and trauma surgery. Both can lead to a high burden for the patients and the healthcare system. Musculoskeletal infections can be induced by intraoperative contamination, bacterial contamination of open wounds or hematogenous bacterial spread. During the recent decades, advances were achieved in the understanding of pathogenesis and antibiotic resistance. Despite some progress in the diagnosis and advancing of therapeutic concepts, groundbreaking successful improvement of treatment concepts is still missing. Current therapy concepts are based on the two pillars consisting of surgical debridement with joint or bone reconstruction as well as prolonged antibiotic therapy. An improved understanding of both host and pathogen-related factors leading to treatment failure is essential in musculoskeletal infections. Therefore, this review aims to give an overview of pathogen-related pathophysiology in musculoskeletal infections. It describes defense strategies of pathogens such as (1) biofilm, its development, characteristics, and treatment options. In addition, (2) characteristics of small colony variants and (3) intracellular bacteria are highlighted. Lastly (4) an outlook for potential and promising future therapeutic strategies is provided.

Involvement of E3 Ubiquitin Ligases in Viral Infections of the Human Host.

Viral infections are one of the principal causes of global primary health crises, with increased rate of infection and mortality demonstrated by the newer progeny of viruses. Viral invasion of the host involves utilization of various cellular machinery. Ubiquitination is one of a few central regulatory systems used by viruses for establishment of the infections in the host. Members of the ubiquitination system are involved in carrying out proteasomal degradation or functional modification of proteins in numerous cellular processes. E3 ubiquitin ligases play a major role in this system through recognition and recruitment of protein substrates and catalyzing the transfer of ubiquitin to these substrates. The versatility of ubiquitin ligases frequently makes them useful tools for the viruses, for either utilizing or degrading other cellular machineries, for carrying out their multiplication or inactivating the defensive strategies of the host. Therefore, these ligases are important targets for aiming at major pathways causing viral protein degradation or functional modification of the infection process. In this review, we have discussed the role and mechanism of different types of ubiquitin ligases in the context of infections of mainly human viruses, highlighting the viral proteins directly interacting with the ligases. Knowledge about these direct interactions is central in understanding the ubiquitin-dependent processes. This comprehensive account may also be beneficial for pharmaceutical exploration of E3 ligase-based broad-spectrum antiviral treatment.

In Defense of Anti-Haecceitism: Chance and Counterfactuals

AbstractBoris Kment formulates two challenges to anti-haecceitism: one based on the concept of chance, and the other involving counterfactuals. In this paper I refute the chance-based argument by appropriately reformulating its premises in a non-individualistic language. Regarding the argument from counterfactuals I consider three strategies of defense. The strategy that is ultimately accepted as most promising is a revamped approach based on the distinction between possible worlds and possibilities. I counter Kment’s claim that the introduction of fine-grained possibilities within possible worlds is haecceitism in disguise. My conclusion is that anti-haecceitism survives the criticism, and is an overall more attractive position than its competitor.

Evaluating and enhancing the robustness of vision transformers against adversarial attacks in medical imaging.

Deep neural networks (DNNs) have demonstrated exceptional performance in medical image analysis. However, recent studies have uncovered significant vulnerabilities in DNN models, particularly their susceptibility to adversarial attacks that manipulate these models into making inaccurate predictions. Vision Transformers (ViTs), despite their advanced capabilities in medical imaging tasks, have not been thoroughly evaluated for their robustness against such attacks in this domain. This study addresses this research gap by conducting an extensive analysis of various adversarial attacks on ViTs specifically within medical imaging contexts. We explore adversarial training as a potential defense mechanism and assess the resilience of ViT models against state-of-the-art adversarial attacks and defense strategies using publicly available benchmark medical image datasets. Our findings reveal that ViTs are vulnerable to adversarial attacks even with minimal perturbations, although adversarial training significantly enhances their robustness, achieving over 80% classification accuracy. Additionally, we perform a comparative analysis with state-of-the-art convolutional neural network models, highlighting the unique strengths and weaknesses of ViTs in handling adversarial threats. This research advances the understanding of ViTs robustness in medical imaging and provides insights into their practical deployment in real-world scenarios.

Guilty or innocent? Could the exotic lizard Hemidactylus mabouia (Squamata, Gekkonidae) be responsible for the displacement of the native species Notomabuya frenata (Squamata, Mabuyidae) in anthropic environments?

Hemidactylus mabouia is a gecko species that has successfully colonized southern North, Central, and South America. In Paraguay, there is a commonly held belief that this species has displaced the native skink Notomabuya frenata in urban areas. In this study, we compared several ecological features of N. frenata and H. mabouia to better understand the degree of their niche overlap. A comparative analysis was conducted based on a literature survey and the examination of museum specimens from Asunción (Paraguay). Furthermore, differences in microhabitat use between H. mabouia and N. frenata were investigated through a review of observational records available on iNaturalist. Our findings indicate that the two species share only a few biological traits, such as an insectivore diet, an ability to adapt to living close to human dwellings, and the use of mimesis as a primary defensive strategy. Both lizards are generalist feeders, although there are notable differences in their frequency of prey consumption. Regarding habitat use, the majority of N. frenata specimens were documented on grass or other natural substrates, as well as on trees and shrubs. In the case of H. mabouia, the majority of records originate from smooth and brick walls, including metal and glass surfaces. Consequently, there is no evidence to suggest that dietary overlap could induce competition between these species. Therefore, it can be posited that the displacement of the native lizard N. frenata is a consequence of human activities that have altered its preferred microhabitat while expanding potential habitat for H. mabouia.

The Great Powers Competition and Increasing Entropy in the Local Media Across Africa and Asia

Abstract The current United States National Defense Strategy prioritizes strategic competition with Russia and China, whereas for the previous two decades, the priority was counter-terrorism. This research aims to measure the extent to which competition among great powers is intensifying and identify where such intensification is taking place. To achieve this, we utilize a large, open-source dataset called GDELT 2.0, which provides an archive of global news articles. We calculate the entropy for local press coverage of great powers (e.g., the United States, China, Russia, France, and India) across 96 African and Asian nations to measure static competition levels. Additionally, we identify African and Asian nations where competition levels have changed significantly between 2015 and 2020. Our findings show that seven countries are experiencing increased competition among great powers, whereas seven other countries experience increased domination by one great power. When we aggregate the data geographically, we find that Southeast Asia is the only subregion where there is a significant decrease in competition. Our results align with the primary concern of the National Defense Strategy: that American influence is decreasing. Specifically, we have identified increased competition across Africa and increased domination across Asia. These trends are due to a relatively higher reporting of Chinese in-country public and private entities between 2015 and 2020.

Allelopathic Interactions Between the Green-Tide-Forming Ulva prolifera and the Golden-Tide-Forming Sargassum horneri Under Controlled Laboratory Conditions

Harmful algal blooms (HABs) represent a significant global marine ecological disaster. In the Yellow Sea, green and golden tides often occur simultaneously or sequentially, suggesting that interspecific competition involves not only spatial and resource competition but also allelopathy. This study investigated the allelopathic interactions between Ulva prolifera and Sargassum horneri using physiological and biochemical parameters, including relative growth rate (RGR), cell ultrastructure, chlorophyll fluorescence, enzyme activity, and metabolomics analysis. The results showed that S. horneri filtrate significantly inhibited U. prolifera growth, while U. prolifera filtrate had no significant effect on S. horneri. Both algal filtrates caused cellular damage and affected photosynthesis, enzyme activities, and metabolism. However, their allelopathic responses differed: U. prolifera may rely on internal compensatory mechanisms, while S. horneri may depend on defense strategies. These findings provide insights into the dynamics of green and golden tides and support the scientific control of HABs through allelopathy.

Building cybersecurity resilience: integrating defense and recovery investment strategies in an expected resilience framework

PurposeTo remain competitive in an unpredictable environment where the complexity and frequency of cybercrime are rapidly increasing, a cyber resiliency strategy is vital for business continuity. However, one of the barriers to improving cyber resilience is that security defense and accident recovery do not combine efficaciously, as embodied by emphasizing cyber security defense strategies, leaving firms ill-prepared to respond to attacks. The present study thus develops an expected resilience framework to assess cyber resilience, analyze cyber security defense and recovery investment strategies and balance security investment allocation strategies.Design/methodology/approachBased on the expected utility theory, this paper presents an expected resilience framework, including an expected investment resilience model and an expected profit resilience model that directly addresses the optimal joint investment decisions between defense and recovery. The effects of linear and nonlinear recovery functions, risk interdependence and cyber insurance on defense and recovery investment are also analyzed.FindingsAccording to the findings, increasing the defense investment coefficient reduces defense and recovery investment while increasing the expected resilience. The nonlinear recovery function requires a smaller defense investment and overall security investment than the linear one, reflecting the former’s advantages in lowering cybersecurity costs. Moreover, risk interdependence has positive externalities for boosting defense and recovery investment, meaning that the expected profit resilience model can reduce free-riding behavior in security investments. Insurance creates moral hazard for firms by lowering defensive investment, yet after purchasing insurance, expanded coverage and cost-effectiveness incentivize firms to increase defense and recovery spending, respectively.Originality/valueThe paper is innovative in its methodology as it offers an expected cyber resilience framework for integrating defense and recovery investment and their effects on security investment allocation, which is crucial for building cybersecurity resilience but receives little attention in cybersecurity economics. It also provides theoretical advances for cyber resilience assessment and optimum investment allocation in other fields, such as cyber-physical systems, power and water infrastructure – moving from a resilience triangle metric to an expected utility theory-based method.