Pioneering advancements in artificial intelligence, especially in genAI, have enabled significant possibilities for content creation, but also led to widespread misinformation and false content. The growing sophistication and realism of deepfakes is raising concerns about privacy invasion, identity theft, and has societal, business impacts, including reputational damage and financial loss. Many deepfake detectors have been developed to tackle this problem. Nevertheless, as for every AI model, the deepfake detectors face the wrath of lack of considerable generalization to unseen scenarios and cross-domain deepfakes. Besides, adversarial robustness is another critical challenge, as detectors drastically underperform to the slightest imperceptible change. Most state-of-the-art detectors are trained on static datasets and lack the ability to adapt to emerging deepfake attack trends. These three crucial challenges though hold paramount importance for reliability in practise, particularly in the deepfake domain, are also the problems with any other AI application. This paper proposes an adversarial meta-learning algorithm using task-specific adaptive sample synthesis and consistency regularization, in a refinement phase. By focussing on the classifier's strengths and weaknesses, it boosts both robustness and generalization of the model. Additionally, the paper introduces a hierarchical multi-agent retrieval-augmented generation workflow with a sample synthesis module to dynamically adapt the model to new data trends by generating custom deepfake samples. The paper further presents a framework integrating the meta-learning algorithm with the hierarchical multi-agent workflow, offering a holistic solution for enhancing generalization, robustness, and adaptability. Experimental results demonstrate the model's consistent performance across various datasets, outperforming the models in comparison.

Intelligent reflecting surface (IRS) is being considered as a prospective candidate for next-generation wireless communication due to its ability to significantly improve coverage and spectral efficiency by controlling the propagation environment. One of the ways IRS increases spectral efficiency is by adjusting phase shifts to perform passive beamforming. In this letter, we integrate the concept of IRS-aided communication to the domain of multi-direction beamforming, whereby multiple receive antennas are selected to convey more information bits than existing spatial modulation (SM) techniques at any specific time. To complement this system, we also propose a successive signal detection (SSD) technique at the receiver. Numerical results show that the proposed design is able to improve the average successful bits transmitted (ASBT) by the system, which outperforms other state-of-the-art methods proposed in the literature.

In a typical \emph{billboard advertisement} technique, a number of digital billboards are owned by an \emph{influence provider}, and several commercial houses approach the influence provider for a specific number of views of their advertisement content on a payment basis. If the influence provider provides the demanded or more influence, then he will receive the full payment else a partial payment. In the context of an influence provider, if he provides more or less than the advertisers demanded influence, it is a loss for him. This is formalized as 'Regret', and naturally, in the context of the influence provider, the goal will be to allocate the billboard slots among the advertisers such that the total regret is minimized. In this paper, we study this problem as a discrete optimization problem and propose two solution approaches. The first one selects the billboard slots from the available ones in an incremental greedy manner, and we call this method the Budget Effective Greedy approach. In the second one, we introduce randomness in the first one, where we do it for a sample of slots instead of calculating the marginal gains of all the billboard slots. We analyze both algorithms to understand their time and space complexity. We implement them with real-life datasets and conduct a number of experiments. We observe that the randomized budget effective greedy approach takes reasonable computational time while minimizing the regret.

Non-Hermitian systems are widespread in both classical and quantum physics. The dynamics of such systems has recently become a focal point of research, showcasing surprising behaviors that include apparent violation of the adiabatic theorem and chiral topological conversion related to encircling exceptional points (EPs). These have both fundamental interest and potential practical applications. Yet the current literature features a number of apparently irreconcilable results. Here we develop a general theory for slow evolution of non-Hermitian systems and resolve these contradictions. We prove an analog of the adiabatic theorem for non-Hermitian systems and generalize it in the presence of uncontrolled environmental fluctuations (noise). The effect of noise turns out to be crucial due to inherent exponential instabilities present in non-Hermitian systems. Disproving common wisdom, the end state of the system is determined by the final Hamiltonian only, and is insensitive to other details of the evolution trajectory in parameter space. Our quantitative theory, leading to transparent physical intuition, is amenable to experimental tests. It provides efficient tools to predict the outcome of the system's evolution, avoiding the need to follow costly time-evolution simulations. Our approach may be useful for designing devices based on non-Hermitian physics and may stimulate analyses of classical and quantum non-Hermitian-Hamiltonian dynamics, as well as that of quantum Lindbladian and hybrid-Liouvillian systems.

The unprecedented growth in the easy availability of photo-editing tools has endangered the power of digital this http URL image was supposed to be worth more than a thousand words,but now this can be said only if it can be authenticated orthe integrity of the image can be proved to be intact. In thispaper, we propose a digital image forensic technique for JPEG images. It can detect any forgery in the image if the forged portion called a ghost image is having a compression quality different from that of the cover image. It is based on resaving the JPEG image at different JPEG qualities, and the detection of the forged portion is maximum when it is saved at the same JPEG quality as the cover image. Also, we can precisely predictthe JPEG quality of the cover image by analyzing the similarity using Structural Similarity Index Measure (SSIM) or the energyof the images. The first maxima in SSIM or the first minima inenergy correspond to the cover image JPEG quality. We created adataset for varying JPEG compression qualities of the ghost and the cover images and validated the scalability of the experimental this http URL also, experimented with varied attack scenarios, e.g. high-quality ghost image embedded in low quality of cover image,low-quality ghost image embedded in high-quality of cover image,and ghost image and cover image both at the same this http URL proposed method is able to localize the tampered portions accurately even for forgeries as small as 10x10 sized pixel this http URL technique is also robust against other attack scenarios like copy-move forgery, inserting text into image, rescaling (zoom-out/zoom-in) ghost image and then pasting on cover image.

The advent of advanced Generative AI (GenAI) models such as DeepSeek and ChatGPT has significantly reshaped the cybersecurity landscape, introducing both promising opportunities and critical risks. This study investigates how GenAI powered chatbot services can be exploited via jailbreaking techniques to bypass ethical safeguards, enabling the generation of phishing content, recommendation of hacking tools, and orchestration of phishing campaigns. In ethically controlled experiments, we used ChatGPT 4o Mini selected for its accessibility and status as the latest publicly available model at the time of experimentation, as a representative GenAI system. Our findings reveal that the model could successfully guide novice users in executing phishing attacks across various vectors, including web, email, SMS (smishing), and voice (vishing). Unlike automated phishing campaigns that typically follow detectable patterns, these human-guided, AI assisted attacks are capable of evading traditional anti phishing mechanisms, thereby posing a growing security threat. We focused on DeepSeek and ChatGPT due to their widespread adoption and technical relevance in 2025. The study further examines common jailbreaking techniques and the specific vulnerabilities exploited in these models. Finally, we evaluate a range of mitigation strategies such as user education, advanced authentication mechanisms, and regulatory policy measures and discuss emerging trends in GenAI facilitated phishing, outlining future research directions to strengthen cybersecurity defenses in the age of artificial intelligence.

We report on experimental observations of the bending of a dust acoustic shock wave around a dust void region. This phenomenon occurs as a planar shock wavefront encounters a compressible obstacle in the form of a void whose size is larger than the wavelength of the wave. As they collide, the central portion of the wavefront, that is the first to touch the void, is blocked while the rest of the front continues to propagate, resulting in an inward bending of the shock wave. The bent shock wave eventually collapses, leading to the transient trapping of dust particles in the void. Subsequently, a Coulomb explosion of the trapped particles generates a bow shock. The experiments have been carried out in a DC glow discharge plasma, where the shock wave and the void are simultaneously created as self-excited modes of a three-dimensional dust cloud. The salient features of this phenomenon are reproduced in molecular dynamics simulations, which provide valuable insights into the underlying dynamics of this interaction.

Motivated by recent experiments of motile bacteria crossing liquid-liquid interfaces of isotropic- nematic coexistence (Cheon et al., Soft Matter 20: 7313-7320, 2024), we study the dynamics of prolate microswimmers traversing clean liquid-liquid interfaces. Using large-scale lattice Boltzmann simulations, we observe that neutrally wetting swimmers can be either trapped or cross the in- terface, depending on their initial angle, swimming speed and the interfacial tension between the two fluids. The simulation results are rationalized by considering a competition between interfacial (thermodynamic) and active (hydrodynamic) forces. The swimmers get trapped at the interface due to a thermodynamic trapping force, akin to Pickering effect, when the forces from interfacial tension dominate over the swimming forces. The trapping behavior can be captured by calculating a critical capillary number by balancing the interfacial and active energies. This prediction agrees remarkably well with the numerical simulations as well as the bacterial experiments of Cheon et al., (Soft Matter 20: 7313-7320, 2024). Finally, our results demonstrate that the torque resulting in a reorientation of the swimmers parallel to the interface have both hydro and thermodynamic components.

We analyze the stability of an inverse problem for determining the time-dependent matrix potential appearing in the Dirichlet initial-boundary value problem for the wave equation in an unbounded cylindrical waveguide. The observation is given by the input-output map associated with the wave equation. Considering a suitable geometric optics solution and with the help of light ray transform, we demonstrate the stability estimate in the determination of the time-dependent matrix potential from the given input-output map.

In this letter, using energy transfers, we demonstrate a route to thermalization in an isolated ensemble of realistic gas particles. We performed a grid-free classical molecular dynamics simulation of two-dimensional Lenard-Jones gas. We start our simulation with a large-scale vortex akin to a hydrodynamic flow and study its non-equilibrium behavior till it attains thermal equilibrium. In the intermediate phases, small wavenumbers ($k$) exhibit $E(k) \propto k^{-3}$ kinetic energy spectrum whereas large wavenumbers exhibit $E(k) \propto k$ spectrum. Asymptotically, $E(k) \propto k$ for the whole range of $k$, thus indicating thermalization. These results are akin to those of Euler turbulence despite complex collisions and interactions among the particles.

The main purpose of this paper is to design a fully discrete local discontinuous Galerkin (LDG) scheme for the generalized Benjamin-Ono equation. First, we proved the $L^2$-stability for the proposed semi-discrete LDG scheme and obtained a sub-optimal order of convergence for general nonlinear flux. We develop a fully discrete LDG scheme using the Crank-Nicolson (CN) method and fourth-order fourth-stage Runge-Kutta (RK) method in time. Adapting the methodology established for the semi-discrete scheme, we demonstrate the stability of the fully discrete CN-LDG scheme for general nonlinear flux. Additionally, we consider the fourth-order RK-LDG scheme for higher order convergence in time and prove that it is strongly stable under an appropriate time step constraint by establishing a \emph{three-step strong stability} estimate for linear flux. Numerical examples associated with soliton solutions are provided to validate the efficiency and optimal order of accuracy for both methods.

Brain tumor segmentation plays a crucial role in computer-aided diagnosis. This study introduces a novel segmentation algorithm utilizing a modified nnU-Net architecture. Within the nnU-Net architecture's encoder section, we enhance conventional convolution layers by incorporating omni-dimensional dynamic convolution layers, resulting in improved feature representation. Simultaneously, we propose a multi-scale attention strategy that harnesses contemporary insights from various scales. Our model's efficacy is demonstrated on diverse datasets from the BraTS-2023 challenge. Integrating omni-dimensional dynamic convolution (ODConv) layers and multi-scale features yields substantial improvement in the nnU-Net architecture's performance across multiple tumor segmentation datasets. Remarkably, our proposed model attains good accuracy during validation for the BraTS Africa dataset. The ODconv source code along with full training code is available on GitHub.

In the bilayer ReS$_{2}$ channel of a field-effect transistor (FET), we demonstrate using Raman spectroscopy that electron doping (n) results in softening of frequency and broadening of linewidth of the in-plane vibrational modes, leaving out-of-plane vibrational modes unaffected. Largest change is observed for the in-plane Raman mode at $\sim$ 151 cm$^{-1}$, which also shows doping induced Fano resonance with the Fano parameter 1/q = -0.17 at doping concentration of $\sim 3.7\times10^{13}$ cm$^{-2}$. A quantitative understanding of our results is provided by first-principles density functional theory (DFT), showing that the electron-phonon coupling (EPC) of in-plane modes is stronger than that of out-of-plane modes, and its variation with doping is independent of the layer stacking. The origin of large EPC is traced to 1T to 1T$^{\prime}$ structural phase transition of ReS$_{2}$ involving in-plane displacement of atoms whose instability is driven by the nested Fermi surface of the 1T structure. Results are also compared with the isostructural trilayer ReSe$_{2}$.

Digital technology has made possible unimaginable applications come true. It seems exciting to have a handful of tools for easy editing and manipulation, but it raises alarming concerns that can propagate as speech clones, duplicates, or maybe deep fakes. Validating the authenticity of a speech is one of the primary problems of digital audio forensics. We propose an approach to distinguish human speech from AI synthesized speech exploiting the Bi-spectral and Cepstral analysis. Higher-order statistics have less correlation for human speech in comparison to a synthesized speech. Also, Cepstral analysis revealed a durable power component in human speech that is missing for a synthesized speech. We integrate both these analyses and propose a machine learning model to detect AI synthesized speech.

Fake speech detection systems have become a necessity to combat against speech deepfakes. Current systems exhibit poor generalizability on out-of-domain speech samples due to lack to diverse training data. In this paper, we attempt to address domain generalization issue by proposing a novel speech representation using self-supervised (SSL) speech embeddings and the Modulation Spectrogram (MS) feature. A fusion strategy is used to combine both speech representations to introduce a new front-end for the classification task. The proposed SSL+MS fusion representation is passed to the AASIST back-end network. Experiments are conducted on monolingual and multilingual fake speech datasets to evaluate the efficacy of the proposed model architecture in cross-dataset and multilingual cases. The proposed model achieves a relative performance improvement of 37% and 20% on the ASVspoof 2019 and MLAAD datasets, respectively, in in-domain settings compared to the baseline. In the out-of-domain scenario, the model trained on ASVspoof 2019 shows a 36% relative improvement when evaluated on the MLAAD dataset. Across all evaluated languages, the proposed model consistently outperforms the baseline, indicating enhanced domain generalization.

There are many approaches in mobile data ecosystem that inspect network traffic generated by applications running on user's device to detect personal data exfiltration from the user's device. State-of-the-art methods rely on features extracted from HTTP requests and in this context, machine learning involves training classifiers on these features and making predictions using labelled packet traces. However, most of these methods include external feature selection before model training. Deep learning, on the other hand, typically does not require such techniques, as it can autonomously learn and identify patterns in the data without external feature extraction or selection algorithms. In this article, we propose a novel deep learning based end-to-end learning framework for prediction of exposure of personally identifiable information (PII) in mobile packets. The framework employs a pre-trained large language model (LLM) and an autoencoder to generate embedding of network packets and then uses a triplet-loss based fine-tuning method to train the model, increasing detection effectiveness using two real-world datasets. We compare our proposed detection framework with other state-of-the-art works in detecting PII leaks from user's device.

Phishing websites continue to pose a significant security challenge, making the development of robust detection mechanisms essential. Brand Domain Identification (BDI) serves as a crucial step in many phishing detection approaches. This study systematically evaluates the effectiveness of features employed over the past decade for BDI, focusing on their weighted importance in phishing detection as of 2025. The primary objective is to determine whether the identified brand domain matches the claimed domain, utilizing popular features for phishing detection. To validate feature importance and evaluate performance, we conducted two experiments on a dataset comprising 4,667 legitimate sites and 4,561 phishing sites. In Experiment 1, we used the Weka tool to identify optimized and important feature sets out of 5: CN Information(CN), Logo Domain(LD),Form Action Domain(FAD),Most Common Link in Domain(MCLD) and Cookie Domain through its 4 Attribute Ranking Evaluator. The results revealed that none of the features were redundant, and Random Forest emerged as the best classifier, achieving an impressive accuracy of 99.7\% with an average response time of 0.08 seconds. In Experiment 2, we trained five machine learning models, including Random Forest, Decision Tree, Support Vector Machine, Multilayer Perceptron, and XGBoost to assess the performance of individual BDI features and their combinations. The results demonstrated an accuracy of 99.8\%, achieved with feature combinations of only three features: Most Common Link Domain, Logo Domain, Form Action and Most Common Link Domain,CN Info,Logo Domain using Random Forest as the best classifier. This study underscores the importance of leveraging key domain features for efficient phishing detection and paves the way for the development of real-time, scalable detection systems.

The concept of Aadhaar came with the need for a unique identity for every individual. To implement this, the Indian government created the authority UIDAI to distribute and generate user identities for every individual based on their demographic and biometric data. After the implementation, came the security issues and challenges of Aadhaar and its authentication. So, our study focuses on the journey of Aadhaar from its history to the current condition. The paper also describes the authentication process, and the updates happened over time. We have also provided an analysis of the security attacks witnessed so far as well as the possible countermeasure and its classification. Our main aim is to cover all the security aspects related to Aadhaar to avoid possible security attacks. Also, we have included the current updates and news related to Aadhaar.

User profiling, the practice of collecting user information for personalized recommendations, has become widespread, driving progress in technology. However, this growth poses a threat to user privacy, as devices often collect sensitive data without their owners' awareness. This article aims to consolidate knowledge on user profiling, exploring various approaches and associated challenges. Through the lens of two companies sharing user data and an analysis of 18 popular Android applications in India across various categories, including $\textit{Social, Education, Entertainment, Travel, Shopping and Others}$, the article unveils privacy vulnerabilities. Further, the article propose an enhanced machine learning framework, employing decision trees and neural networks, that improves state-of-the-art classifiers in detecting personal information exposure. Leveraging the XAI (explainable artificial intelligence) algorithm LIME (Local Interpretable Model-agnostic Explanations), it enhances interpretability, crucial for reliably identifying sensitive data. Results demonstrate a noteworthy performance boost, achieving a $75.01\%$ accuracy with a reduced training time of $3.62$ seconds for neural networks. Concluding, the paper suggests research directions to strengthen digital security measures.

Carbon-deficient giants (CDGs) are a rare and chemically peculiar class of stars whose origins remain under active investigation. We present an asteroseismic analysis of the entire known CDG population, selecting 129 stars observed by $Kepler$, K2, and TESS to obtain seismic constraints. We detect solar-like oscillations in 43 CDGs. By measuring $\nu_{\rm max}$ and applying seismic scaling relations, we determine precise masses for these stars, finding that 79\% are low-mass ($M \lesssim 2~M_\odot$). The luminosity distribution is bimodal, and the CDGs separate into three chemically and evolutionarily distinct groups, characterized by clear trends in sodium and CNO abundances, $\alpha$-element enhancement, and kinematics. We find that two of these groups are only distinguished by their initial $\alpha$-element abundances, thus effectively reducing the number of groups to two. Lithium enrichment is common across all groups, linking CDGs to lithium-rich giants and suggesting a shared evolutionary origin. We find that spectroscopic $\log g$ is systematically offset from seismic values. Group~1 CDG patterns are most consistent with formation through core He-flash mixing, while the more massive and more chemically processed Groups~2 and 2$\alpha$ likely formed through mergers involving helium white dwarfs, possibly in hierarchical triples. Pollution from AGB stars appears very unlikely, given the unchanged [C+N+O] abundance across all groups.