The evolution of the traditional power grid into the "smart grid" has resulted in a fundamental shift in energy management, which allows the integration of renewable energy sources with modern communication technology. However, this interconnection has increased smart grids' vulnerability to attackers, which might result in privacy breaches, operational interruptions, and massive outages. The SCADA-based smart grid protocols are critical for real-time data collection and control, but they are vulnerable to attacks like unauthorized access and denial of service (DoS). This research proposes a hybrid deep learning-based Intrusion Detection System (IDS) intended to improve the cybersecurity of smart grids. The suggested model takes advantage of Convolutional Neural Networks' (CNN) feature extraction capabilities as well as Long Short-Term Memory (LSTM) networks' temporal pattern recognition skills. DNP3 and IEC104 intrusion detection datasets are employed to train and test our CNN-LSTM model to recognize and classify the potential cyber threats. Compared to other deep learning approaches, the results demonstrate considerable improvements in accuracy, precision, recall, and F1-score, with a detection accuracy of 99.70%.

Motivated by the high resource costs and privacy concerns associated with centralized machine learning, federated learning (FL) has emerged as an efficient alternative that enables clients to collaboratively train a global model while keeping their data local. However, in real-world deployments, client data distributions often evolve over time and differ significantly across clients, introducing heterogeneity that degrades the performance of standard FL algorithms. In this work, we introduce Fed-REACT, a federated learning framework designed for heterogeneous and evolving client data. Fed-REACT combines representation learning with evolutionary clustering in a two-stage process: (1) in the first stage, each client learns a local model to extracts feature representations from its data; (2) in the second stage, the server dynamically groups clients into clusters based on these representations and coordinates cluster-wise training of task-specific models for downstream objectives such as classification or regression. We provide a theoretical analysis of the representation learning stage, and empirically demonstrate that Fed-REACT achieves superior accuracy and robustness on real-world datasets.

Figure 3: Distribution of Structural Similarity Index (SSIM), Peak Signal-to-Noise Ratio (PSNR), and Haar wavelet-based Perceptual Similarity Index (HaarPSI) scores for DeepFoqus-Accelerate reconstructions: (a-c) show results across 408 samples at 2x, 3x, and 4x acceleration, and (d-f) present distributions for the 36 image sets evaluated by reviewers. Figure 4: (A-B) Representative standard-of-care (SOC) images (first row) and DeepFoqus-Accelerate reconstructions from accelerated scans (second row), with corresponding quantitative and qualitative scores presented in the third row. Panel (B) shows two slices of the worst-case scenario in the qualitative dataset, characterized by wrap-around and motion artifacts. Discussion This evaluation of DeepFoqus-Accelerate demonstrates that this FDA-cleared k-space-based DL reconstruction software can reliably enable up to fourfold accelerated brain MRI acquisition without compromising diagnostic image quality. Both expert review and quantitative image similarity metrics confirm that AI-reconstructed images are clinically equivalent to fully sampled standards.

Abstract--Rule-Based Moral Principles for Explaining Uncertainty in Natural Language Generation As large language models (LLMs) are increasingly used in high-stakes applications, the challenge of explaining uncertainty in natural language generation has become both a technical and moral imperative. Traditional approaches rely on probabilistic methods that are often opaque, difficult to interpret, and misaligned with human expectations of transparency and accountability. In response to these limitations, this paper introduces a novel framework based on rule-based moral principles--simple, human-inspired ethical guidelines--for responding to uncertainty in LLM-generated text. Drawing on insights from experimental moral psychology and virtue ethics, we define a set of symbolic behavioral rules such as precaution, deference, and responsibility to guide system responses under conditions of epistemic or aleatoric uncertainty. These rules are implemented declaratively and are designed to generate adaptive, context-sensitive explanations even in the absence of precise confidence metrics. The moral principles are encoded as symbolic rules within a lightweight Prolog-based engine, where each uncertainty tag (low, medium, high) activates an ethically aligned system action along with an automatically generated, plain-language rationale. We evaluate the framework through scenario-based simulations that benchmark rule coverage, assess fairness implications, and analyze trust calibration. An interpretive explanation module is integrated to reveal both the assigned uncertainty level and its underlying justification in a transparent and accessible way. We illustrate the framework through hypothetical yet plausible use cases in clinical and legal domains, demonstrating how rule-based moral reasoning can enhance user trust, promote fairness, and improve the interpretability of AI-generated language. By offering a lightweight, philosophically grounded alternative to probabilistic uncertainty modeling, our approach paves the way for more ethical, human-aligned, and socially responsible natural language generation.

Text-to-SQL models allow users to interact with a database more easily by generating executable SQL statements from natural-language questions. Despite recent successes on simpler databases and questions, current Text-to-SQL methods still suffer from low execution accuracy on industry-scale databases and complex questions involving domain-specific business logic. We present \emph{PaVeRL-SQL}, a framework that combines \emph{Partial-Match Rewards} and \emph{Verbal Reinforcement Learning} to drive self-improvement in reasoning language models (RLMs) for Text-to-SQL. To handle practical use cases, we adopt two pipelines: (1) a newly designed in-context learning framework with group self-evaluation (verbal-RL), using capable open- and closed-source large language models (LLMs) as backbones; and (2) a chain-of-thought (CoT) RL pipeline with a small backbone model (OmniSQL-7B) trained with a specially designed reward function and two-stage RL. These pipelines achieve state-of-the-art (SOTA) results on popular Text-to-SQL benchmarks -- Spider, Spider 2.0, and BIRD. For the industrial-level Spider2.0-SQLite benchmark, the verbal-RL pipeline achieves an execution accuracy 7.4\% higher than SOTA, and the CoT pipeline is 1.4\% higher. RL training with mixed SQL dialects yields strong, threefold gains, particularly for dialects with limited training data. Overall, \emph{PaVeRL-SQL} delivers reliable, SOTA Text-to-SQL under realistic industrial constraints. The code is available at https://github.com/PaVeRL-SQL/PaVeRL-SQL.

The sympathetic nervous system (SNS) plays a central role in regulating the body's responses to stress and maintaining physiological stability. Its dysregulation is associated with a wide range of conditions, from cardiovascular disease to anxiety disorders. Skin nerve activity (SKNA) extracted from high-frequency electrocardiogram (ECG) recordings provides a noninvasive window into SNS dynamics, but its measurement is highly susceptible to electromyographic (EMG) contamination. Traditional preprocessing based on bandpass filtering within a fixed range (e.g., 500--1000 Hz) is susceptible to overlapping EMG and SKNA spectral components, especially during sustained muscle activity. We present a denoising approach using a lightweight one-dimensional convolutional autoencoder with a long short-term memory (LSTM) bottleneck to reconstruct clean SKNA from EMG-contaminated recordings. Using clean ECG-derived SKNA data from cognitive stress experiments and EMG noise from chaotic muscle stimulation recordings, we simulated contamination at realistic noise levels (--4 dB, --8 dB signal-to-noise ratio) and trained the model in the leave-one-subject-out cross-validation framework. The method improved signal-to-noise ratio by up to 9.65 dB, increased cross correlation with clean SKNA from 0.40 to 0.72, and restored burst-based SKNA features to near-clean discriminability (AUROC $\geq$ 0.96). Classification of baseline versus sympathetic stimulation (cognitive stress) conditions reached accuracies of 91--98\% across severe noise levels, comparable to clean data. These results demonstrate that deep learning--based reconstruction can preserve physiologically relevant sympathetic bursts during substantial EMG interference, enabling more robust SKNA monitoring in naturalistic, movement-rich environments.

Large language models (LLMs) show increasing potential in education, yet benchmarks for non-English languages in specialized domains remain scarce. We introduce MedBench-IT, the first comprehensive benchmark for evaluating LLMs on Italian medical university entrance examinations. Sourced from Edizioni Simone, a leading preparatory materials publisher, MedBench-IT comprises 17,410 expert-written multiple-choice questions across six subjects (Biology, Chemistry, Logic, General Culture, Mathematics, Physics) and three difficulty levels. We evaluated diverse models including proprietary LLMs (GPT-4o, Claude series) and resource-efficient open-source alternatives (<30B parameters) focusing on practical deployability. Beyond accuracy, we conducted rigorous reproducibility tests (88.86% response consistency, varying by subject), ordering bias analysis (minimal impact), and reasoning prompt evaluation. We also examined correlations between question readability and model performance, finding a statistically significant but small inverse relationship. MedBench-IT provides a crucial resource for Italian NLP community, EdTech developers, and practitioners, offering insights into current capabilities and standardized evaluation methodology for this critical domain.

The widespread use of generative AI has shown remarkable success in producing highly realistic deepfakes, posing a serious threat to various voice biometric applications, including speaker verification, voice biometrics, audio conferencing, and criminal investigations. To counteract this, several state-of-the-art (SoTA) audio deepfake detection (ADD) methods have been proposed to identify generative AI signatures to distinguish between real and deepfake audio. However, the effectiveness of these methods is severely undermined by anti-forensic (AF) attacks that conceal generative signatures. These AF attacks span a wide range of techniques, including statistical modifications (e.g., pitch shifting, filtering, noise addition, and quantization) and optimization-based attacks (e.g., FGSM, PGD, C \& W, and DeepFool). In this paper, we investigate the SoTA ADD methods and provide a comparative analysis to highlight their effectiveness in exposing deepfake signatures, as well as their vulnerabilities under adversarial conditions. We conducted an extensive evaluation of ADD methods on five deepfake benchmark datasets using two categories: raw and spectrogram-based approaches. This comparative analysis enables a deeper understanding of the strengths and limitations of SoTA ADD methods against diverse AF attacks. It does not only highlight vulnerabilities of ADD methods, but also informs the design of more robust and generalized detectors for real-world voice biometrics. It will further guide future research in developing adaptive defense strategies that can effectively counter evolving AF techniques.

Blockchain and smart contracts have garnered significant interest in recent years as the foundation of a decentralized, trustless digital ecosystem, thereby eliminating the need for traditional centralized authorities. Despite their central role in powering Web3, their complexity still presents significant barriers for non-expert users. To bridge this gap, Artificial Intelligence (AI)-based agents have emerged as valuable tools for interacting with blockchain environments, supporting a range of tasks, from analyzing on-chain data and optimizing transaction strategies to detecting vulnerabilities within smart contracts. While interest in applying AI to blockchain is growing, the literature still lacks a comprehensive survey that focuses specifically on the intersection with AI agents. Most of the related work only provides general considerations, without focusing on any specific domain. This paper addresses this gap by presenting the first Systematization of Knowledge dedicated to AI-driven systems for blockchain, with a special focus on their security and privacy dimensions, shedding light on their applications, limitations, and future research directions.

Generated Scalable Vector Graphics (SVG) images demand evaluation criteria tuned to their symbolic and vectorial nature: criteria that existing metrics such as FID, LPIPS, or CLIPScore fail to satisfy. In this paper, we introduce SVGauge, the first human-aligned, reference-based metric for text-to-SVG generation. SVGauge jointly measures (i) visual fidelity, obtained by extracting SigLIP image embeddings and refining them with PCA and whitening for domain alignment, and (ii) semantic consistency, captured by comparing BLIP-2-generated captions of the SVGs against the original prompts in the combined space of SBERT and TF-IDF. Evaluation on the proposed SHE benchmark shows that SVGauge attains the highest correlation with human judgments and reproduces system-level rankings of eight zero-shot LLM-based generators more faithfully than existing metrics. Our results highlight the necessity of vector-specific evaluation and provide a practical tool for benchmark-ing future text-to-SVG generation models.