Search Results (298)

Concurrency bugs originate from complex and improper synchronization of shared resources, presenting a significant challenge for detection. Traditional static analysis relies heavily on expert knowledge and frequently fails when code is non-compilable. Conversely, large language models struggle with semantic sparsity, inadequate comprehension of concurrent semantics, and the tendency to hallucinate. To address the limitations of static analysis in capturing complex concurrency semantics and the hallucination risks associated with large language models, this study proposes ConSynergy. This novel framework integrates the structural rigor of static analysis with the semantic reasoning capabilities of large language models. The core design employs a robust task decomposition strategy that decomposes concurrency bug detection into a four-stage pipeline: shared resource identification, concurrency-aware slicing, data-flow reasoning, and formal verification. This approach fundamentally mitigates hallucinations from large language models caused by insufficient program context. First, the framework identifies shared resources and applies a concurrency-aware program slicing technique to precisely extract concurrency-related structural features, thereby alleviating semantic sparsity. Second, to enhance the large language model’s comprehension of concurrent semantics, we design a concurrency data-flow analysis based on Chain-of-Thought prompting. Third, the framework incorporates a Satisfiability Modulo Theories solver to ensure the reliability of detection results, alongside an iterative repair mechanism based on large language models that dramatically reduces dependency on code compilability. Extensive experiments on three mainstream concurrency bug datasets, including DataRaceBench, the concurrency subset of Juliet, and DeepRace, demonstrate that ConSynergy achieves an average precision and recall of 80.0% and 87.1%, respectively. ConSynergy outperforms state-of-the-art baselines by 10.9% to 68.2% in average F1 score, demonstrating significant potential for practical application. Full article

Background: The management of hard-to-heal wounds poses a major clinical challenge due to heterogeneous etiology and significant global healthcare costs (estimated at USD 148.64 billion in 2022). Large Language Models (LLMs), such as ChatGPT and Gemini, are emerging as potential decision-support tools. This study aimed to rigorously assess the accuracy and reliability of ChatGPT and Gemini in the visual description and initial therapeutic management of complex wounds based solely on clinical images. Methods: Twenty clinical images of complex wounds from diverse etiologies were independently analyzed by ChatGPT (version dated 15 October 2025) and Gemini (version dated 15 October 2025). The models were queried using two standardized, concise prompts. The AI responses were compared against a clinical gold standard established by the unanimous consensus of an expert panel of three plastic surgeons. Results: Statistical analysis showed no significant difference in overall performance between the two models and the expert consensus. Gemini achieved a slightly higher percentage of perfect agreement in management recommendations (75.0% vs. 60.0% for ChatGPT). Both LLMs demonstrated high proficiency in identifying the etiology of vascular lesions and recognizing critical “red flags,” such as signs of ischemia requiring urgent vascular assessment. Noted divergences included Gemini’s greater suspicion of potential neoplastic etiology and the models’ shared error in suggesting Negative Pressure Wound Therapy (NPWT) in a case potentially contraindicated by severe infection. Conclusions: LLMs, particularly ChatGPT and Gemini, demonstrate significant potential as decision-support systems and educational tools in wound care, offering rapid diagnosis and standardized initial management, especially in non-specialist settings. Instances of divergence in systemic treatments or in atypical presentations highlight the limitations of relying on image-based reasoning alone. Ultimately, LLMs serve as powerful, scalable assets that, under professional supervision, can enhance diagnostic speed and improve care pathways. Full article

This paper evaluates multi-agent coordination strategies against single-agent retrieval-augmented generation (RAG) for open-source language models. Four coordination strategies (collaborative, sequential, competitive, hierarchical) were tested across Mistral 7B, Llama 3.1 8B, and Granite 3.2 8B using 100 domain-specific question–answer pairs (3100 total evaluations). Performance was assessed using Composite Performance Score (CPS) and Threshold-aware CPS (T-CPS), aggregating nine metrics spanning lexical, semantic, and linguistic dimensions. Under the tested conditions, all 28 multi-agent configurations showed degradation relative to single-agent baselines, ranging from −4.4% to −35.3%. Coordination overhead was identified as a primary contributing factor. Llama 3.1 8B tolerated Sequential and Hierarchical coordination with minimal degradation (−4.9% to −5.3%). Mistral 7B with shared context retrieval achieved comparable results. Granite 3.2 8B showed degradation of 14–35% across all strategies. Collaborative coordination exhibited the largest degradation across all models. Study limitations include evaluation on a single domain (agriculture), use of 7–8B parameter models, and homogeneous agent architectures. These findings suggest that single-agent RAG may be preferable for factual question-answering tasks in local deployment scenarios with computational constraints. Future research should explore larger models, heterogeneous agent teams, role-specific prompting, and advanced consensus mechanisms. Full article

The rapid development of large artificial intelligence (AI) models—large language models (LLMs), multimodel large language models (MLLMs) and vision–language models (VLMs)—has enabled instruction-driven visual understanding, where a single foundation model can recognize and localize arbitrary objects from natural-language prompts. However, predictions from individual models remain inconsistent—LLMs hallucinate nonexistent entities, while VLMs exhibit limited recall and unstable calibration compared to purpose-trained detectors. To address these limitations, a new paradigm termed “multiple large AI model’s consensus” has emerged. In this approach, multiple heterogeneous LLMs, MLLMs or VLMs process a shared visual–textual instruction and generate independent structured outputs (bounding boxes and categories). Next, their results are merged through consensus mechanisms. This cooperative inference improves spatial accuracy and semantic correctness, making it particularly suitable for generating high-quality training datasets for fast real-time object detectors. This survey provides a comprehensive overview of the large multi-AI model’s consensus for object detection. We formalize the concept, review related literature on ensemble reasoning and multimodal perception, and categorize existing methods into four frameworks: prompt-level, reasoning-to-detection, box-level, and hybrid consensus. We further analyze fusion algorithms, evaluation metrics, and benchmark datasets, highlighting their strengths and limitations. Finally, we discuss open challenges—vocabulary alignment, uncertainty calibration, computational efficiency, and bias propagation—and identify emerging trends such as consensus-aware training, structured reasoning, and collaborative perception ecosystems. Full article

Recent advances in reproductive science have prompted a profound reexamination of some of the most fundamental anthropological aspects of human life: the value of nascent human life, the meanings of motherhood and fatherhood, and the concept of family. Abrahamic religious traditions in particular offer a rich moral heritage, developed over centuries, that can significantly contribute to ethical reflection on assisted reproductive technologies. This article examines the Judeo-Christian and Islamic traditions, which are predominant in the Western cultural context and greatly influence the lives and moral frameworks of more than half of the world’s population. The study underscores the strength of the ethical foundations shared across these religious traditions and common values, principles, and moral concerns, while also seeking to understand and integrate the distinctive nuances that differentiate them. Full article

Stroke is the second-largest cause of death and disability worldwide, and many patients require intensive care for airway compromise, hemodynamic instability, cerebral edema, or systemic complications. This review summarizes key aspects of ICU management in both acute ischemic stroke (AIS) and hemorrhagic stroke (HS). Priorities are airway protection, oxygenation, individualized blood pressure targets, and strict control of temperature and glucose. Neurological monitoring and prompt management of intracranial pressure (ICP), together with timely surgical interventions (hemicraniectomy or hematoma evacuation), are central to acute care. Seizures are treated promptly, while routine prophylaxis is not recommended. Prevention of aspiration pneumonia, venous thromboembolism, infections, and other intensive care unit (ICU) complications is essential, along with early nutrition, mobilization, and rehabilitation. Prognosis and decisions about intensity of care require shared discussions with families and involvement of palliative services, when appropriate. Many practices remain based on observational data or extrapolation from other populations, underlining the need for stroke-specific clinical trials. Outcomes are consistently better when patients are managed in specialized stroke or neurocritical care units with a multidisciplinary treatment approach Full article

Companion animals, including dogs and cats, share close living environments with humans, making antimicrobial stewardship essential to prevent zoonotic transmission of resistant pathogens. The overuse and misuse of conventional antibiotics in veterinary medicine have accelerated the emergence of multidrug-resistant (MDR) microorganisms, prompting the need for alternative strategies. Natural compounds, such as antimicrobial peptides (AMPs), phytochemicals, chitosan-based polymers, and nutraceuticals, offer promising solutions due to their broad-spectrum activity, low resistance potential, and additional health-promoting properties. This review provides a comprehensive analysis of recent advances of the aforementioned compounds for companion animals, including their mechanisms of action, applications in feed and nutraceuticals, and therapeutic use in dermatological, gastrointestinal, and systemic infections. We discuss the current challenges related to bioavailability, safety, standardization, and regulatory frameworks, as well as future perspectives for integrating these agents into veterinary practice. Emphasis is placed on clinical evidence in dogs and cats, highlighting how natural antimicrobials can contribute to sustainable infection control and antimicrobial resistance mitigation under the One Health paradigm. Full article

Background: Kaposiform hemangioendothelioma (KHE) is a rare, locally aggressive vascular tumor that shares histological features with Kaposi’s sarcoma. It usually occurs in infancy or early childhood and is seldom reported in adults. The most common sites are the skin and retroperitoneum, whereas the head, neck, and mediastinum are less frequently involved. KHE rarely regresses spontaneously, and metastasis is uncommon, but up to 70% of cases may develop Kasabach–Merritt Syndrome (KMS), a life-threatening coagulopathy. Here, we present an unusual case of KHE in an adult patient, emphasizing the importance of early recognition and management. Methods: A 39-year-old woman with systemic lupus erythematosus presented with an exophytic lesion in the left retromolar region. Clinical and radiological evaluations were followed by both incisional and excisional biopsies. Histopathological and immunohistochemical analyses were performed, and surgical resection with wide margins was undertaken according to recommendations from a multidisciplinary tumor board. Results: Histology revealed spindle cell clusters, slit-like vascular spaces, endothelial cells with eosinophilic cytoplasm, and immunopositivity for CD31, CD34, and smooth muscle actin, confirming the diagnosis of KHE. Given the tumor’s locally aggressive behavior and potential risk of KMS, extended surgical excision was performed. Conclusions: This case underscores the diagnostic challenges of KHE in adults and highlights the essential role of histopathology, immunohistochemistry, and multidisciplinary evaluation. Prompt diagnosis and radical surgical management are critical to preventing complications and improving patient outcomes. Full article

Grid-connected photovoltaic (PV)–battery systems require advanced control to maintain stable operation, efficient energy exchange, and minimal conversion losses under variable generation and load conditions. This study proposes a dual-loop Energy Management System (EMS) integrated with a Hybrid Grey Wolf Optimizer–Particle Swarm Optimization (GWO–PSO) algorithm for coordinated control of a low-voltage PV–battery–grid system (380 V AC, ≈800 V DC bus). The hybrid optimizer was chosen due to the limitations of standalone GWO and PSO methods, which frequently experience slow convergence and local stagnation; the integrated GWO–PSO strategy enhances both exploration and exploitation during the real-time adjustment of PI controller gains. The rapid inner loop effectively balances instantaneous power among the PV, battery, and grid, while the outer optimization loop aims to minimize the ITAE criterion to enhance transient response. Simulation outcomes validate stable DC-bus voltage regulation, quicker transitions between power import and export, and prompt power balance with deviations maintained below 2.5%, signifying reduced converter losses and improved power-sharing efficiency. The battery’s state of charge is sustained within the range of 20–80%, ensuring safe operational conditions. The proposed hybrid EMS offers faster convergence, smoother power regulation, and enhanced dynamic stability compared to standalone metaheuristic controllers, establishing it as an effective and reliable solution for grid-connected PV–battery systems. Full article

Contemporary cities and metropolises, as complex spatial and social structures, require innovative tools for promotion, education, and the identification of development potential. The search for such tools prompted the authors to conduct the research. This article attempts to assess the effectiveness of urban and metropolitan games as tools of territorial marketing and as means of supporting spatial education and social participation. The research is based on the analysis of 42 games with urban and metropolitan themes, selected according to defined criteria. Both qualitative and quantitative research methods were applied, including documentation analysis, comparative analysis techniques, statistical methods, case studies, and coding of games across seven parameters (dimensions), using five descriptors (coded 1–5) per parameter. The research results indicate a high diversity among the analyzed games in terms of structure, function, and application. The proposed typology of games allowed for an in-depth, systematic comparison. The identification of five typological clusters allowed for an assessment of the advantages and limitations of individual game forms. This provided data on the suitability of individual game types for various purposes, including their application in territorial marketing and urban education. The findings confirm that urban and metropolitan games can play a significant role in building spatial awareness, supporting planning processes, and promoting urban areas. They represent an innovative tool supporting the sustainable development of cities and metropolises, particularly in the areas of resident engagement in decision-making processes, collaboration between authorities, residents, and non-governmental organizations, planning with consideration for future generations, as well as fostering—even among the younger generation—a sense of shared responsibility for urban space and the decisions undertaken. Full article

In light of urbanization, climate change, and the escalation of extreme weather events, flood management is becoming more and more important. Improving community resilience and reducing flood risks require prompt decision-making and effective communication. This study investigates how flood management systems can incorporate Large Language Models (LLMs), especially those that use Retrieval-Augmented Generation (RAG) architectures. We suggest a multimodal framework that uses a Flood Knowledge Graph to aggregate data from various sources, such as social media, hydrological, and meteorological inputs. Although LLMs have the potential to be transformative, we also address important drawbacks like governance issues, hallucination risks, and a lack of physical modeling capabilities. When compared to text-only LLMs, the RAG system significantly improves the reliability of flood-related decision support by reducing factual inconsistency rates by more than 75%. Our suggested architecture includes expert validation and security layers to guarantee dependable, useful results, like flood-constrained evacuation route planning. In areas that are vulnerable to flooding, this strategy seeks to strengthen warning systems, enhance information sharing, and build resilient communities. Full article

Text data is indispensable for modern machine learning and natural language processing but often contains sensitive information that must be protected before sharing or release. Differential privacy (DP) provides rigorous guarantees for privacy preservation, yet applying DP to text rewriting poses unique challenges. Existing approaches frequently assume white-box access to large language models (LLMs), relying on internal signals such as logits or gradients. These assumptions limit practicality, since real-world users typically interact with LLMs only through black-box APIs. We introduce PrivRewrite, a framework for differentially private text rewriting that operates entirely under black-box access. PrivRewrite constructs a diverse pool of candidate rewrites through randomized prompting and pruning and then employs the exponential mechanism to select a single release with end-to-end $ϵ$-DP. A key contribution is our refined sensitivity analysis of the utility function, which yields tighter bounds than naive estimates and thereby strengthens the accuracy guarantees of the exponential mechanism. The framework requires no fine-tuning, internal model access, or local inference, making it lightweight and deployable in practical API-based settings. Experimental results on benchmark datasets demonstrate that PrivRewrite achieves strong privacy–utility trade-offs, producing fluent and semantically faithful outputs while upholding formal privacy guarantees. These results highlight the feasibility of black-box DP text rewriting and show how refined sensitivity analysis can further improve utility under strict privacy constraints. Full article

CCTV safety monitoring demands anomaly detectors combine reliable clip-level accuracy with predictable per-clip latency despite weak supervision. This work investigates compact vision–language models (VLMs) as practical detectors for this regime. A unified evaluation protocol standardizes preprocessing, prompting, dataset splits, metrics, and runtime settings to compare parameter-efficiently adapted compact VLMs against training-free VLM pipelines and weakly supervised baselines. Evaluation spans accuracy, precision, recall, F1, ROC-AUC, and average per-clip latency to jointly quantify detection quality and efficiency. With parameter-efficient adaptation, compact VLMs achieve performance on par with, and in several cases exceeding, established approaches while retaining competitive per-clip latency. Adaptation further reduces prompt sensitivity, producing more consistent behavior across prompt regimes under the shared protocol. These results show that parameter-efficient fine-tuning enables compact VLMs to serve as dependable clip-level anomaly detectors, yielding a favorable accuracy–efficiency trade-off within a transparent and consistent experimental setup. Full article

South Asia’s worsening air pollution crisis represents one of the most urgent public health and environmental challenges of the 21st century. Nearly two billion people—over one-quarter of the global population—reside in this region, where air quality levels routinely exceed World Health Organization (WHO) guidelines by factors of 10 to 15. This has translated into an unprecedented health burden, with approximately two million premature deaths annually, widespread chronic respiratory and cardiovascular disease, and rising economic losses. According to recent World Bank estimates, welfare losses amount to over 5% of regional GDP, a figure far exceeding the projected costs of coordinated mitigation. Despite this, South Asia continues to lack a binding regional framework capable of addressing its shared airshed. Existing cooperative efforts—such as the Malé Declaration on Control and Prevention of Air Pollution (1998)—have provided a useful platform for dialog and pilot monitoring, but they remain voluntary, under-resourced, and insufficient to manage the transboundary nature of the crisis. National-level programs, including India’s National Clean Air Programme (NCAP), Bangladesh’s National Air Quality Management Plan (NAQMP), and Nepal’s National Air Quality Management Action Plan (AQMAP), demonstrate domestic commitment but are constrained by fragmentation, limited financing, and lack of regional integration. This gap represents the central knowledge and governance challenge that prompted the present commentary. To address it, we propose a dual-track architecture designed to institutionalize binding regional cooperation. Track A would establish a United Nations-anchored South Asian Transboundary Air Pollution Protocol, under the auspices of the United Nations Environment Programme, the World Health Organization (WHO), and the United Nations Economic and Social Commission for Asia and the Pacific (UNESCAP). This protocol would codify legally enforceable emission standards, compliance committees, financial mechanisms, and harmonized monitoring. Track B would establish a South Asian Association for Regional Cooperation (SAARC) Prime Ministers’ Council on Air Quality (SPMCAQ) to provide political leadership, align domestic implementation, and authorize rapid responses to cross-border haze events. Lessons from the Indian Ocean Experiment, the ASEAN Agreement on Transboundary Haze Pollution, and Europe’s Convention on Long-Range Transboundary Air Pollution demonstrate that legally binding agreements combined with high-level political ownership can achieve durable reductions in pollution despite geopolitical tensions. By situating South Asia within these global precedents, the proposed framework provides a pragmatic, enforceable, and politically resilient pathway to protect health, reduce economic losses, and deliver cleaner air for nearly one-quarter of humanity. Full article

In the realm of Industry 5.0, the incorporation of Artificial Intelligence (AI) in overseeing workers, machinery, and industrial systems is essential for fostering a human-centric, sustainable, and resilient industry. Despite technological advancements, the construction industry remains largely labor intensive, with site management and interventions predominantly reliant on manual judgments, leading to inefficiencies and various challenges. This research emphasizes identifying unsafe behaviors and risks within construction environments by employing AI. Given the continuous emergence of unsafe behaviors that requires certain caution, it is imperative to adapt to these novel categories while retaining the knowledge of existing ones. Although deep convolutional neural networks have shown excellent performance in behavior recognition, they traditionally function as predefined multi-way classifiers, which exhibit limited flexibility in accommodating emerging unsafe behavior classes. Addressing this issue, this study proposes a versatile and efficient recognition model capable of expanding the range of unsafe behaviors while maintaining the recognition of both new and existing categories. Adhering to the continual learning paradigm, this method integrates two types of complementary prompts into the pre-trained model: task-invariant prompts that encode knowledge shared across tasks, and task-specific prompts that adapt the model to individual tasks. These prompts are injected into specific layers of the frozen backbone to guide learning without requiring a rehearsal buffer, enabling effective recognition of both new and previously learned unsafe behaviors. Additionally, this paper introduces a benchmark dataset, Split-UBR, specifically constructed for continual unsafe behavior recognition on construction sites. To rigorously evaluate the proposed model, we conducted comparative experiments using average accuracy and forgetting as metrics, and benchmarked against state-of-the-art continual learning baselines. Results on the Split-UBR dataset demonstrate that our method achieves superior performance in terms of both accuracy and reduced forgetting across all tasks, highlighting its effectiveness in dynamic industrial environments. Full article