Search Results (924)

Blockchain architectures increasingly rely on smart contracts as programmable execution components, yet the security challenges they introduce remain only partially addressed. Fuzz testing has emerged as one of the leading automated techniques for smart contract vulnerability discovery; however, a systematic treatment linking classical fuzzing concepts to the specific architectural constraints of smart contract execution environments has remained elusive. We conduct a large-scale systematic review of 258 publications collected from Scopus and Google Scholar. The review is structured around four research questions covering general fuzzing limitations, EVM execution constraints, cross-contract interaction challenges, and technique transferability. The contributions are fourfold. First, we develop a conceptual taxonomy establishing an explicit correspondence between classical and smart contract fuzzing. Second, we present an architectural taxonomy of representative fuzzers unified under a generalised waypoint architecture that exposes key feedback domains and composition gaps. Third, we provide a categorised technique review spanning coverage-guided, hybrid, and learning-based approaches. Fourth, we offer a reproducibility critique with a structured research roadmap. Coverage-guided greybox fuzzing emerges as the dominant paradigm (51.2% of the 121 SC fuzzer tools); learning-based approaches (machine learning, reinforcement learning, and LLMs) are a small but emerging class, with LLM-guided fuzzing the fastest-emerging subcategory by recency. Feedback-domain composition remains sparse: most surveyed fuzzers combine only a few of the eight identified feedback domains, and none approaches the full set, leaving several high-value multi-domain compositions unexplored. Open challenges include scalable stateful exploration, standardised benchmarks and oracles, and EVM-specific architectural optimisations. The complete categorised corpus, coding, and search strategy are openly available. Full article

Background: Automated assessment of relational competencies by large language models (LLMs) is gaining traction in medical education, yet its empirical reliability remains poorly documented. Methods: We evaluated ATENA, a GPT-3.5-Turbo virtual-patient simulator deployed at the Faculty of Medicine of Nice (France), in two sequential components. First, 21 university hospital professors rated seven dimensions of realism on 7-point Likert scales. Second, across 129 student simulations of a single diagnostic disclosure scenario, expert human scores based on the empathic communication coding system (ECCS) were compared with ATENA’s automated scores using agreement metrics, Bland–Altman analysis, confusion matrix, and qualitative subcorpus analysis. Results: Realism was rated high on six of seven dimensions. Evaluative agreement was poor (ICC = 0.146), with a systematic overscoring bias (+0.80 points) and wide limits of agreement [−1.57; +3.16]. Three bias patterns emerged—score compression (76% of scores within 4.2–4.3), overscoring of weak performance, and underscoring of exemplary performance—indicating that ATENA responds to formal markers of empathy present in nearly every transcript. Conclusions: In this first-generation version of ATENA, the system distinguishes stronger from weaker interviews without being able to place them at the level assigned by expert human raters. These patterns preclude any certificatory use and support a critical, formative-only deployment grounded in instructor algorithmic literacy. The audit protocol applies to any LLM evaluator scored against human coding on a validated instrument. Full article

Generative artificial intelligence (GenAI) is reshaping higher education and early-career work, raising questions about how universities can support pedagogically sustainable career readiness. This study examines undergraduate students’ perceptions of AI, automation, entry-level employment, and perceived preparedness for an AI-augmented labor market. Survey data were collected from 153 undergraduate students. The questionnaire examined awareness of AI and automation, perceived risks to traditional entry-level work, anxiety and perceived preparedness regarding post-graduation employment, skill priorities, and desired institutional support. The findings indicate substantial awareness of AI-related change, with many students expecting routine junior tasks such as data entry, basic research, report generation, customer support, and simple coding-related work to be affected. However, confidence in academic preparation was weaker and more uncertain. Students emphasized human-centered capabilities, including critical thinking, creativity, communication, and problem solving, alongside AI literacy and practical exposure to digital tools. The study identifies an awareness-preparedness gap and argues that higher education institutions should strengthen GenAI-era curriculum design, AI-authentic assessment, experiential learning, career guidance, and ethical AI literacy to support perceived preparedness and sustainable career readiness. Full article

Sequential object-oriented software remains difficult to parallelize because several important decisions are not entirely local. A useful decomposition should consider architectural structure, execution frequency, communication latency, and how much communication can overlap with local computation. In contrast, many traditional modularization methods focus on cohesion, coupling, or communication cost without explicitly considering whether an unavoidable remote call can be issued early enough for part of its latency to be hidden by independent work. DTMC-GA addresses this problem by combining a Discrete-Time Markov Chain (DTMC) execution model, a Distributive Execution Time (DET) estimator, and a Genetic Algorithm (GA) search procedure. Guarded call-use displacement enlarges modeled overlap windows only when the relevant dependence, ordering, and safety conditions are satisfied. The public artifact contains five local benchmark fixtures, five comparison methods, 30 retained runs per algorithm–fixture pair, raw outputs, statistical scripts, regenerated figures, Docker support, continuous-integration verification, and a Zenodo DOI. Within this released five-method panel, DTMC-GA achieved the highest aggregate mean modeled speedup of 2.708 ± 0.329. A more recent NSGA-III comparator was also evaluated using matched fixtures, DTMC–DET inputs, constraints, random seeds, population size, and generation budget. NSGA-III achieved an aggregate mean modeled speedup of 2.737 ± 0.323, 1.07% above DTMC-GA, although none of the five fixture-level paired differences remained significant after Bonferroni correction. A separate sparse-graph scalability study covered 50, 100, 250, 500, 750, and 1000 classes; mean analysis time increased from about 0.024 to 0.286 s, while peak memory rose from about 119.1 to 120.9 MiB. The study also includes an executed OpenCV 4.7.0 calibration using a fixed five-convolution DNN workload. Across 30 matched pairs, native-parallel OpenCV achieved a measured ratio-of-means speedup of 1.8691×. The bounded empirical DTMC-GA/DET rerun over 30 seeds produced a mean modeled speedup of 1.0000× with an SD of 0.0000, giving an absolute modeled-versus-measured difference of 46.50%. Because the measured parallel comparator is OpenCV’s native threaded implementation rather than a DTMC-GA-generated source rewrite, this experiment is interpreted as real-runtime model calibration rather than direct measured DTMC-GA optimized speedup. Full article

Automating the identification of International Classification of Diseases (ICD) codes from electronic health records (EHRs) presents a critical challenge, particularly for rare diseases where existing computational methods severely underperform due to extreme long-tail label distributions. To address this, we propose a multi-modal deep learning framework known as MMAC-Net, designed to enhance the retrospective assignment of ICD-9 codes to admissions involving rare pathologies. The model integrates unstructured clinical narratives with structured auxiliary data, specifically pharmacological prescriptions and microbiology events, using a convolutional attention-based architecture. Through a late fusion mechanism, it synthesizes attention-weighted textual representations with dense embeddings of the structured data types. Validation on the MIMIC-III dataset shows consistent improvements over a matched text-only baseline evaluated under an identical protocol. On the full dataset of 8930 ICD codes, the framework achieved a Micro-AUC of 0.997 and Precision@8 of 0.875. On the subset of admissions carrying at least 1 of 568 rare codes, adding the two structured modalities to the text encoder raises Macro-F1 from 0.011 to 0.084 and Micro-F1 from 0.368 to 0.513 relative to the text-only baseline, corresponding to relative increases of 6.69 and 0.39, respectively, while Precision@8 rises from 0.092 to 0.159 and Micro-AUC from 0.966 to 0.985. While extreme class imbalance remains a formidable obstacle, these findings underscore that incorporating structured clinical context partially mitigates the limitations of purely natural language processing approaches. Practically, the framework is intended as a decision-support component that presents a ranked shortlist of candidate codes to a human coder or clinician; by recovering rare codes that text-only systems miss, it targets the under-coding of low-prevalence conditions that degrades registry completeness and downstream epidemiological estimates. Full article

In the digital education system, the increasing demand for measuring lexical and semantic text similarity has led to the need for developing scalable and intelligent automated grading systems. Traditional assessment methods used for short-answer grading are labor-intensive, time-consuming and prone to human-evaluator bias. Furthermore, computer science curricula have different evaluation challenges as programming-centric courses require rigorous syntactical and structural validation, whereas theoretical courses require deep conceptual and semantic comprehension. To address these dual challenges, we introduce the Virtual University Automatic Short-Answer Grading (VUASAG) framework, a model addressing both statistical-based lexical analysis and Transformer-based semantic architectures. Utilizing an expanded Mohler dataset spanning multiple computer science domains (including Data Structures, Introduction to Programming, Object-Oriented Programming, and Software Engineering), we evaluated the efficiency of traditional string-matching metrics alongside state-of-the-art Transformer-based models (T5, BERT, XLNet, and SBERT). The empirical results demonstrate that token-based statistical models perform robustly on keyword-dependent syntax validation while Transformer-based models excel by capturing deep conceptual context. Specifically, Sentence-BERT (SBERT) achieved superior predictive accuracy across courses, yielding a minimum Root Mean Square Error (RMSE) of 0.9511 for the Introduction to Programming course, paired with moderate positive Pearson and Spearman correlation coefficients. Confidence intervals are estimated using Fisher’s z-transformation with a 95 percentile confidence level. Question-level five-fold and two-fold cross-validation is conducted for all four courses, and SBERT is found to perform better. Finally, to demonstrate the practical utility, we deployed a web-based portal for students’ evaluation leveraging a LLaMA-based pre-trained model to deliver real-time, personalized pedagogical feedback measuring code functionality, quality, documentation, and error handling. This reduces the educator workload while preserving grading integrity. Full article

The Future of Jobs Report 2025 identifies technological change, geoeconomic fragmentation, demographic shifts, and the green transition as compounding forces restructuring global labour markets. Rather than merely adding new risks, these pressures erode the foundational assumptions upon which labour rights rest, widening the gap between workers’ lived conditions and the reach of existing legal frameworks across international, regional, national, and platform levels. Algorithmic management serves as the focal case through which these pressures converge. In platform and conventional workplaces alike, technical systems now allocate tasks, price labour, rank workers, enforce sanctions, and determine classification, cutting across traditional boundaries separating labour law, private law, data protection, and AI regulation. Two analytical concepts organise the argument. Mediated autonomy denotes a condition in which workers retain nominal choice, yet that choice is structured in advance by systems governing incentives, visibility, risk, and legal categorisation. Algorithmic subordination captures the resulting form of control: indirect, informationally asymmetric, and legally dispersed dependence on data flows, automated rankings, and code-driven decisions. Drawing on the philosophy of technology and classificatory looping theory, the article demonstrates that legal and technical categories do not merely describe workers; they actively constitute the field within which workers bear risk and assert entitlements. Biometric monitoring and affective computing intensify this dynamic by extending managerial authority to bodily signals and inferred emotional states. China’s regulatory approach to non-standard employment is examined as a case study and compared with the EU’s status-and-algorithm model, situating both within broader patterns of global regulatory divergence. The article concludes that effective labour protection must move beyond employment-status classification toward control-sensitive and technology-sensitive frameworks, encompassing transparency obligations, contestation rights, human review mechanisms, fair remuneration, data minimisation for biometric and affective information, and meaningful worker participation in platform governance. Full article

Agentic artificial intelligence is extending bioinformatics beyond conversational assistance by enabling systems to select tools, execute code, revise analytical plans, and interpret biological data. These capabilities may accelerate research, but they also redistribute decisions that determine whether biological conclusions are valid. We conducted a targeted, structured PubMed search in July 2026 and identified 11 peer-reviewed agentic bioinformatics systems for descriptive review based on predefined eligibility criteria for analytical decision-making, tool or code execution, iterative evaluation, or coordinated agent activity. The evidence base covered single-cell transcriptomics, microbial genomics, cancer genomics, and omics applications, together with methodological literature on reproducibility and biological validation. We examined how current systems report delegated authority, provenance, validation, evidence, abstention, and human oversight. Existing platforms implement safeguards such as sandboxed execution, restricted commands, interaction logs, evidence identifiers, automated checks, critic agents, quality scores, and expert assessment. However, published reports rarely provide a connected account linking the original biological question to samples, reference resources, analytical decisions, computational actions, statistical results, supporting evidence, validation outcomes, and final claims. We distinguish inherited bioinformatics errors, errors amplified through autonomous action, and emergent failures arising from memory, retrieval, tool interaction, or agent coordination. We further propose a multidimensional decision-rights profile, consequence-sensitive validation gates, and a claim-to-evidence provenance architecture organized through the Traceable History of Research Evidence, Agent Actions, and Decisions in Bioinformatics (THREAD-Bio) framework. Illustrative cases show that technically successful execution may still support misleading inference. Trustworthy agentic bioinformatics therefore requires claims to remain reconstructible, challengeable, validated, and proportionate to the evidence. Full article

Large language models (LLMs) increasingly repair software vulnerabilities, but most evaluations judge only similarity to a developer fix or removal of the weakness. Neither reveals whether working code was broken. We evaluate eight commercial and open LLMs on 922 JavaScript vulnerability patches, scoring neutralisation and functional preservation. Lacking tests, we score at scale with a reference-based LLM judge, calibrated against execution on a 144-patch benchmark and 254 Java-CVE patches, plus a cross-family judge. The best model fixes 23% of vulnerabilities (judge-based), and cost-efficiency inverts the accuracy ranking. Our central finding concerns the instrument: both judges flag more over-fixes than execution confirms (precision 5–10%), yet on the functional axis agree far more with each other ($\kappa =0.75$) than with execution ($\kappa \le 0.26$), so judge–judge agreement measures reliability, not validity. On real Java code the over-reporting persists, while the judges’ correctness estimates diverge, leaving no single judge trustworthy. Over-fixing is genuine but, under execution, uncommon: a few percent of vulnerability-removing patches, or under 2%, excluding one artefact-prone scenario, both lower bounds. Only adequately tested execution measures the functional-preservation rate, so security-patch evaluation must run the code, use a judge only to rank models, and weigh costs. We release the harness and executable benchmark. Full article

This paper presents an automated computational workflow for the acquisition, instrument-response correction and three-dimensional visualization of multi-component seismic records obtained from the IRIS Federation of Digital Seismograph Networks (FDSNs) using the open-source ObsPy (v1.5.0) package. The workflow targets engineering applications and consists of four stages: (i) selection of three-component (3C) broadband stations, (ii) bandpass filtering and spectral deconvolution of the instrument response to obtain ground displacement in physical units, (iii) calculation of theoretical P- and S-wave arrival times with the Tau-P kinematic algorithm based on the IASP91 reference Earth velocity model, and (iv) construction of an interactive 3D particle motion visualization in which segments associated with the P-wave, S-wave and background are color-coded. The pipeline is demonstrated on three seismic events recorded in February 2023 by the broadband station KO.BNN, including the destructive Mw 7.8 Kahramanmaraş earthquake. The workflow yields the absolute three-dimensional displacement vector and produces interactive visualizations that are intended for use by structural engineers as a complement to traditional one-dimensional acceleration records. Full article

The accelerating use of algorithmic systems in public administration exposes a standardization gap between technical AI assurance and institutional accountability. This article develops a tiered accountability and reporting standards framework for decisions made under public authority. A qualitative design science method combines comparative institutional analysis, structured absence analysis, and standards architecture design. The empirical basis comprises international standards, the United States Department of Government Efficiency (DOGE)–Treasury access episode as an institutional control precursor, Australia’s Robodebt scheme as an automated-decision failure, and public sector governance arrangements in Estonia, Singapore, Japan, South Korea, Canada, and the United States. A replicable coding protocol traces documented accountability gaps to five auditable primitives: provenance tracking, decision logging, role attribution, contestability, and post-deployment audit. The primitives are organized into minimum, heightened, and systemic/constitutional tiers according to material influence, rights and essential service effects, civil service integrity, institutional independence, and substitutive capacity. The article also specifies a Public Sector Algorithmic Accountability Statement (PAAS), crosswalks its ten disclosure fields to GRI 1, GRI 2, and GRI 3, and demonstrates its operation through a fully worked hypothetical benefits eligibility application. A tier assignment decision aid, an assurance cycle, and a cost–feasibility model support implementation, including in small institutions. The framework’s novelty lies not in claiming new lifecycle controls, but in consolidating those controls around the public decision configuration, escalating them according to public authority consequences, and joining internal evidence to comparable public reporting. The proposal shifts standardization from AI system certification alone toward auditable institutional answerability, governance sustainability, and constitutional integrity. Full article

The treatment planning stage accounts for most reported patient safety events in radiation therapy, and automated quality assurance (QA) checklists are a common countermeasure; few of these tools, however, undergo formal human factors evaluation before they are released into clinical use. Building on our prior work demonstrating suboptimal usability of an institutional dosimetry QA checklist (DQC) and its participatory, theory-driven redesign, this study evaluated an enhanced DQC using Borycki and Kushniruk’s multi-phase, multi-method usability evaluation framework, which integrates cognitive and socio-technical perspectives to create a “safety net” against usability problems and technology-induced errors. Three phases were carried out in sequence at one academic medical center: (1) rapid think-aloud usability testing with two cohorts of dosimetrists and physicists (n = 10) separated by an improvement cycle; (2) remote simulation-based testing in which dosimetrists (n = 7) worked through ten high-fidelity synthetic treatment plans containing embedded errors; and (3) six weeks of near-live testing with weekly iterative refinement, involving 15 interview participants and 21 users who provided in-tool feedback (dosimetrists, physicists, and trainees). Reported usability, usefulness, and safety issues decreased by 49% between think-aloud cohorts (59 to 30), and perceived usability met recommended standards (System Usability Scale > 80). Simulation-based testing was feasible; perceived usability, completion time, embedded-error performance, and perceived realism all significantly favored easy over hard plans (participant-level exact Wilcoxon signed-rank tests, p ≤ 0.031; rank-biserial correlation 1.00). Near-live testing surfaced predominantly deeper usefulness and safety issues (77% of 142 coded comments), and the new checklist accounted for 43.6–67.5% of all checklist runs across user groups (relative use frequency). The framework provided complementary, progressively deeper insights and readied the enhanced DQC for clinical implementation. Full article

The escalating impacts of climate change and anthropogenic pressures on vulnerable ecosystems demand digital tools that make advanced modeling more accessible to conservation practitioners. This study presents a TRL-4 prototype that integrates a configurable Digital Twin (DT) core with a generative AI conversational interface for conservation-oriented modeling in Doñana National Park, Spain, a UNESCO World Heritage site facing significant environmental challenges. The main contribution is not the training of specific ecological forecasting models, but the validation of an end-to-end workflow that allows users to configure, execute, inspect, and interpret a predictive system through natural language. The prototype supports the prediction of conservation-relevant ecological indicators, including Iberian lynx population dynamics and waterbird abundance, using heterogeneous environmental, climatic, hydrological, and socio-demographic datasets. The architecture connects a structured YAML configuration, heterogeneous environmental and biological datasets, automated machine learning training, database-backed traceability, dashboard visualization, and SHAP-based interpretability. Through representative executions, the prototype demonstrates that non-technical users can select target and explanatory variables, configure preprocessing options, launch model training, generate predictions, and review their outputs without directly editing configuration files or running code. Although the predictive metrics obtained in selected runs remain preliminary and should be interpreted as diagnostics rather than evidence of general forecasting skill, the results show that conversational DTs can substantially reduce technical barriers to ecological modeling. By combining generative AI, cloud infrastructure, reproducible machine learning workflows, and explainable AI, the proposed architecture provides a strong foundation for future conservation decision-support systems that augment expert judgment while preserving human oversight, transparency, and critical interpretation. Full article

Assessing whether existing lattice towers can survive extreme wind when they are located on ridgelines or at saddle points requires three questions to be answered simultaneously: how the local wind field is modified by the surrounding topography, how the structural geometry recorded in legacy computer-aided design (CAD) drawings can be recovered accurately, and which member fails first and by what mechanism. This paper couples a Physics-Informed Neural Network (PINN) wind solver, jointly constrained by mass and momentum conservation, with a graph-theory-based Drawing Exchange Format (DXF) parser and a closed-loop vulnerability screening module, so that all three questions are answered in a single automated pass. The PINN reconstructs the three-dimensional steady-state wind field over irregular topography in approximately 0.12 s, holding the root-mean-square (RMS) velocity divergence below 2.1 × 10⁻³ (more than two orders of magnitude lower than that of linear interpolation) while recovering the pressure-gradient-driven acceleration that mass-consistent variational solvers cannot represent. On the CAD side, k-dimensional tree (KD-Tree) spatial indexing combined with breadth-first search (BFS) connected-component analysis resolves the pseudo-disconnections, spurious intersections, and multi-level nested block references that are common in production DXF files, achieving 100% node-merging accuracy across fifteen tower drawings. A unified Vulnerability Index (VI) that combines strength, member stability, and plate buckling into a single scalar, updated through Sherman–Morrison rank-one perturbation at a millisecond cost, closes the diagnose–strengthen–verify loop without re-solving the full stiffness system. Applied to a 220 kV line struck by Super Typhoon Meranti, the pipeline identified seven at-risk members that code-based checking had missed, a result consistent with the recorded field damage, and completed the full assessment in 34.4 s, over three orders of magnitude faster than conventional practice. Full article

The transition toward fully autonomous Digital Building Permitting (DBP) requires Automated Compliance Checking (ACC) systems to verify Building Information Models (BIM) against natural-language regulations. While Large Language Models (LLMs) offer strong semantic comprehension, integrating them into ACC introduces “Spatial Hallucinations” and “Serialization Bottlenecks” when processing high-dimensional BIM graphs. This research proposes an Agentic Neuro-Symbolic Framework that decouples semantic interpretation from geometric verification. Instead of relying on generative text for spatial reasoning, an Agentic LLM acts as a dynamic logic synthesizer orchestrating a deterministic geometry kernel ($IfcOpenShell$). The artifact was evaluated against the Australian National Construction Code (NCC 2022) across three stratified tiers: Semantic-Geometric Alignment, Multi-Parametric Dependencies, and Relational Topology Reasoning. Results demonstrate the framework autonomously resolves ontological ambiguity and synthesizes execution logic dynamically. By implementing connectivity graph traversal, the system isolates structural sub-graphs, reducing computational complexity from O(N) to O(K) and bypassing context-window limits. Offloading calculations to a deterministic environment achieves a highly deterministic accuracy rate for spatial queries, yielding immutable BIM Collaboration Format ($.bcfzip$) audit trails. Ultimately, this research establishes a scalable foundation for autonomous compliance, demonstrating that AI reliability in engineering significantly improves when probabilistic models orchestrate deterministic tools rather than predicting physical realities. Full article