Virginia Military Institute

As cyber threats evolve, conventional block ciphers face growing vulnerabilities due to their static configurations and exposure to adaptive attack strategies. This is particularly problematic in Internet of Things and lightweight environments, where cryptographic mechanisms must balance efficiency with resilience. To address this challenge, we introduce a three-layer adaptive encryption framework that combines moving target defense (MTD) with chaotic maps to enhance cryptosystem security. The framework dynamically reconfigures substitution boxes, permutation tables, and key-dependent transformations across layers, expanding the attacker's search space and reducing susceptibility to cryptanalysis. We demonstrate the proposed framework using the Data Encryption Standard (DES) due to its simple structure and long-standing role as a benchmark for lightweight ciphers, despite its discontinuation. Security analysis confirms that the approach achieves high entropy, strong avalanche effect, and resistance to brute-force, differential, and side-channel attacks. Experimental results show that the proposed framework significantly improves efficiency compared to earlier dynamic schemes and legacy 3DES, while maintaining real-time suitability for IoT and dynamic applications.

ABSTRACT The ChatGPT application is widely utilised across diverse research domains; however, its adoption within educational libraries remains limited. This study investigates the integration of ChatGPT into online academic libraries and introduces a ChatGPT‐based framework designed to enhance user experience and service delivery. The framework was evaluated using two empirical metrics: the user satisfaction score (USS) and the resource engagement rate (RER). The evaluation results indicate that ChatGPT prompts can generate realistic scenarios and facilitate active role‐playing, thereby improving the quality and variety of library services. Empirical data show that the proposed approach achieved high USS values, ranging from 4.4 to 4.8 out of 5, while RER increased to 79.5% between January and May 2023. These findings suggest that ChatGPT has considerable potential to strengthen educational library services. Specifically, its integration supports personalised assistance, timely responses and access to subject‐specific expertise. Moreover, AI‐driven Chatbots can curate digital resources, design tailored learning pathways and promote self‐directed learning. Beyond individual support, ChatGPT contributes to collaborative learning environments and ensures users’ access to current and diverse information resources.

The identification of individual molecular species in the interstellar medium (ISM) via radio astronomical observations relies on precise laboratory measurement of the species' rotational spectrum. Following the detections of benzonitrile and ethynylbenzene in the ISM, we perform a search for the disubstituted benzene derivatives that feature both a nitrile and an ethynyl substituent: the ortho (o), meta (m), and para (p) isomers of ethynylbenzonitrile (EBN). Having previously been measured in the millimeter-wave regime, we extend the rotational spectroscopy of the isomeric family down to the centimeter-wave regime where hyperfine splitting due to the electric quadrupole moment of the 14N atom can be resolved. Transitions of the singly substituted 13C and 15N isotopologues of the o- and p-EBN isomers are observed in their respective spectra, and the rotational constants of these isotopologues are also determined. Experimental ground state structural parameters are determined for o- and p-EBN with the rotational constants of the observed isotopologues, and their geometries are compared to that of benzene to determine how the substituents distort the aromatic ring. With the hyperfine splitting characterized, the rotational spectra are compared to the narrow emission lines observed toward the cold molecular cloud TMC-1. No individual transitions are identified in the observational data, and upper limits on the column densities of the three isomers are derived from the observations. Rate coefficients and product branching ratios are theoretically estimated for the formation of the EBN isomers from both benzonitrile and ethynylbenzene via CCH and CN addition, respectively. Based on these formation rates, and the observed column densities of other substituted aromatic species, the column densities of each of the EBN isomers are estimated to be less than an order of magnitude below their computed upper limits.

While it is known that chronic unpredictable stress and negative events adversely affect neurobiological outcomes, much less is known regarding the neurobiological impact of positive emotions such as chronic anticipation of appetitive events. From a translational perspective, an enhanced understanding of the impact of extended exposure to positive emotions may provide novel insights into effective non-pharmacological, behavior-based approaches to enhance mental resilience. Here, we investigate a novel rodent model of chronic Unpredictable Positive Event Response (UPER) training in male and female Long Evans rats to examine behavioral, neural, and endocrine effects of enhanced anticipation of positive events. Rats were exposed to either 3 weeks of daily, randomly administered, cued positive events (UPER training) or exposure to the same positive events administered at the same time (i.e., in a predictable manner) each day to control for anticipation (Enriched Control Training; ENR). Following UPER and ENR training, rats were assessed for cognitive bias, exploratory behaviors, and persistence in a Cognitive Bias Assessment paradigm, Novelty-Suppressed Feeding Task, and an Unattainable Puzzle Reward Task, respectively. In the Cognitive Bias Assessment, a trend for UPER-trained males to respond with an optimistic bias was observed. A main effect of training was observed in the Unattainable Puzzle Reward Task, with UPER-trained rats exhibiting reduced latency to interact with the novel object. A sex-dependent latency to consume a food reward in a Novelty-Suppressed Feeding Task was also seen. Focusing on fecal corticosterone metabolite (FCM) levels following anticipation-enhanced versus anticipation-minimized training, UPER-trained rats exhibited a trend for lower levels than ENR-trained rats. No c-fos activation differences were observed between the groups. Overall, these preliminary findings suggest that anticipation for positive events may have sex-specific effects on emotional responses to uncertain events. Accordingly, further research may determine relevance of this model in preclinical models of psychiatric diseases.

The thermal conductivity of battery components has become an important factor for predicting the temperature distribution within battery cells during their operation. The importance of this increases with a growing cell size and is especially important for prismatic hard case cells. However, its determination is still in an early stage, and different measurement approaches provide largely varying results. Herein, the thermal conductivity of electrode coatings is investigated by measuring the through‐plane thermal conductivity of electrode stacks. For this, the laser flash analysis (LFA) and guarded hot plate (GHP) method are used as common measuring approaches for battery components. A comparison of the different methods for different active materials, porosities, and electrode thicknesses is conducted and confirmed the strong variations between the two methods seen in the literature. Both methods have their challenges and limitations, where the most important is that LFA is best for thin and low‐porous electrodes and that the GHP is best for thick and high‐porous electrodes. The study discusses these and other effects around in greater detail.

This article aims to investigate the optimal control of a class of fractional stochastic integrodifferential equations (SIDEs) driven by a Rosenblatt process and Poisson jumps in a Hilbert space. Fractional-order derivatives offer the advantage of more accurately modelling real-world scenarios compared to integer-order derivatives, particularly in problems involving memory or hereditary characteristics. Despite the widespread belief that the Rosenblatt process is insufficient and cannot be substituted by fractional Brownian motion (fBm), we show that the Rosenblatt process does, in fact, converge and can be used to real-world scenarios. We provide a detailed discussion of the Rosenblatt process, including its properties, and an explanation of how it can be approximated by the Wiener process. We study a novel set of sufficient conditions for the existence and uniqueness of mild solutions. Then the optimal control is demonstrated through the application of Balder’s theorem. Finally, we validate the theoretical findings with a practical example and a numerical simulation.

IntroductionThe advent of large language models and their applications have gained significant attention due to their strengths in natural language processing.MethodsIn this study, ChatGPT and DeepSeek are utilized as AI models to assist in diagnosis based on the responses generated to clinical questions. Furthermore, ChatGPT, Claude, and DeepSeek are used to analyze images to assess their potential diagnostic capabilities, applying the various sensitivity analyses described. We employ prompt engineering techniques and evaluate their abilities to generate high quality responses. We propose several prompts and use them to answer important information on conjunctivitis.ResultsOur findings show that DeepSeek excels in offering precise and comprehensive information on specific topics related to conjunctivitis. DeepSeek provides detailed explanations and in depth medical insights. In contrast, the ChatGPT model provides generalized public information on the infection, which makes it more suitable for broader and less technical discussions. In this study, DeepSeek achieved a better performance with a 7% hallucination rate compared to ChatGPT's 13%. Claude demonstrated perfect 100% accuracy in binary classification, significantly outperforming ChatGPT's 62.5% accuracy.DiscussionDeepSeek showed limited performance in understanding images dataset on conjunctivitis. This comparative analysis serves as an insightful reference for scholars and health professionals applying these models in varying medical contexts.

Efficient modeling of the complex dynamics between flow physics and chemical kinetics during flame propagation is crucial for understanding combustion mechanisms in real-world applications such as propulsion, energy conversion, and emission control. A two-dimensional, high-order, compact finite difference computational fluid dynamics (CFD) code in cylindrical coordinates has been developed to investigate the interaction between flow and chemistry in a laminar non-premixed flame. The numerical framework (implemented in Fortran) employs a fifth-order scheme for convective terms and a fourth-order scheme for viscous terms to enhance the resolution of steep gradients characteristic of reacting flows. Conversely, a fourth-order Runge–Kutta (RK4) time integration scheme improves solution stability and reduces the convergence error by approximately 20% compared to the simpler Euler-forward (EF) method. To accurately capture the intricate coupling between flow and chemistry, the steady flamelet model (SFM) with detailed kinetics is incorporated into the CFD framework, enabling a reduced yet robust representation of chemical kinetics. The model is validated against benchmark experimental and numerical datasets, demonstrating strong agreement and predictive capability, with 60% and 15% lower Root Mean Squared Error (RMSE) for temperature and H2O mole fraction, compared to the species-transport-based numerical results. A systematic parametric study was conducted using a canonical co-flow laminar methane–air diffusion flame configuration. The effects of varying Reynolds number, fuel-to-air velocity ratio, and combustor geometry are analyzed in detail. The study reveals that increasing the combustor length leads to more anchored and spatially confined flame structures. Additionally, the study demonstrates that increasing the inner diameter enhances radial diffusion, resulting in more dispersed flame fronts. These findings underscore the role of combustor geometry in optimizing designs for both efficiency and emissions control. Furthermore, the proposed higher-order scheme advances reactive flow modeling by offering a validated and computationally efficient tool for high-resolution simulations of chemically reacting flows.