Search Results (1,644)

The interaction between humans and artificial intelligence has become a critical channel for information exchange, yet no quantitative, theoretically grounded framework exists for measuring information efficiency in human–AI communication. This study empirically validated an info-metrics framework operationalizing information efficiency through three dimensions—information density (D), relevance (R), and redundancy (Q)—synthesized into an information efficiency metric (IEM). We analyzed 60 AI responses from ChatGPT 5.2 and Claude Opus 4.5 across factual, analytical, and creative question types using combined coding, automated structural measures, and human evaluation of informational units. The results showed that information density and relevance positively contributed to IEM, while redundancy had a negative contribution. Efficiency varied by task type, with factual prompts showing the highest variability across models and highest efficiency. Contrary to expectations, creative responses did not exhibit higher redundancy, suggesting that expressive diversity does not necessarily constitute informational noise. The framework offers a task-sensitive, theoretically grounded approach to evaluating human–AI information exchange beyond correctness or subjective quality judgment, supporting systems-oriented optimization of conversational AI protocols. Full article

Vulnerability disclosures are outpacing manual remediation capacity. We present a Zero-Touch Vulnerability Remediation Framework combining OpenVAS scanning, multi-source threat intelligence, and Large Language Models (LLMs) enhanced through Retrieval-Augmented Generation (RAG). The Scanning Layer normalizes findings into structured JSON; the AI Decision Layer applies hybrid FAISS + BM25 retrieval, dual-LLM verification (a primary generator checked by a gpt-4o auxiliary verifier), and confidence-based routing; the Orchestration Layer executes validated patches via CI/CD pipelines with automated rollback. On 350 real-world vulnerability cases across five GPT-family models, the full Prompt + RAG pipeline raised accuracy from 52.0% to 76.7–82.6% (all p < 0.001, Cohen’s h = 0.51–0.68) and reduced hallucination from 23.4% to 7.8%. Confidence routing routed 34.9% of cases to the high-confidence auto-execution tier, yielding a 4.1% rollback rate and zero service outages. The framework addresses the most relevant categories of the OWASP LLM Top 10 and lays groundwork for enterprise-scale, Zero-Touch vulnerability management. Full article

Recent breakthroughs in large language models have made it feasible to effectively summarize cross-lingual dialogue information, proving essential for the global communication context. However, existing methodologies encounter difficulties in maintaining factual consistency across multiple dialogue exchanges and lack clear explanations of the summarization process. This paper presents a novel factually consistent prompting technology with large language models to address these challenges in cross-lingual dialogue summarization. First, we propose a factual replacement mechanism to enhance information analysis by incorporating noise information into summarization candidates. We adopt a self-guidance framework to enforce factual consistency, enhancing information flow tracking in cross-lingual hybrid dialogue scenarios with the assistance of GPT-based models. Furthermore, we introduce a view-aware chain-of-thought-driven architecture to improve the interpretability and transparency of the cross-lingual dialogue summarization process. Comprehensive experimental evaluations on cross-lingual summarization tasks, spanning English, French, Spanish, Russian, Chinese, and Arabic, and hybrid cross-lingual tasks substantiate that the proposed model achieves superior performance relative to state-of-the-art baselines. Full article

This study examines the impact of ChatGPT-assisted learning on the academic achievement of hospitality and tourism students in Egyptian public universities, with particular emphasis on the mediating roles of perceived usefulness and self-regulated learning. Drawing conceptually on the Technology Acceptance Model (TAM), the study adopts a contextualized framework that emphasizes perceived usefulness while incorporating ChatGPT-assisted learning effectiveness as a learning-oriented driver within generative AI-supported educational environments. A quantitative research design was employed using an online survey administered to students who actively used ChatGPT for academic purposes. A total of 689 valid responses were collected from nine public universities and analyzed using Partial Least Squares Structural Equation Modeling (PLS-SEM) to test the proposed hypotheses. The findings indicate that ChatGPT-Assisted Learning Effectiveness (CALE) has a statistically significant and positive direct effect on academic achievement (AA; β = 0.386, T = 3.946, p < 0.001, 95% CI = 0.192–0.561) and strongly predicts perceived usefulness (β = 0.673, T = 9.274, p < 0.001, 95% CI = 0.581–0.742) and self-regulated learning (β = 0.707, T = 10.734, p < 0.001, 95% CI = 0.621–0.779). In turn, PU (β = 0.281, T = 3.854, p < 0.001, 95% CI = 0.142–0.417) and SRL (β = 0.220, T = 2.418, p = 0.016, 95% CI = 0.041–0.356) significantly enhance academic achievement. Mediation analyses further confirm that PU (β = 0.189, T = 2.366, p = 0.018, 95% CI = 0.031–0.284) and SRL (β = 0.156, T = 3.699, p < 0.001, 95% CI = 0.102–0.301) partially mediate the relationship between CALE and academic achievement. These findings offer important theoretical insights by contextualizing TAM’s performance-related logic within generative AI-driven learning environments and refining its application to academic outcome settings, while highlighting self-regulated learning as a critical explanatory mechanism. From a practical perspective, the study provides valuable implications for educators and policymakers by emphasizing the need to promote students’ perceived usefulness of ChatGPT and foster learner autonomy, positioning generative AI as a powerful pedagogical support tool for enhancing academic success in hospitality and tourism education. Full article

The emergence of Generative Artificial Intelligence (GenAI), capable of interpreting and reasoning with human language, has catalyzed a paradigm shift across various societal sectors. Within the tourism industry, GenAI is increasingly utilized to facilitate personalized itinerary planning, destination recommendations, and the provision of optimal route information. This study evaluates the reliability of GenAI in identifying the nearest metro station within a walking distance from tourist attractions in Busan, South Korea. Furthermore, it aims to empirically verify the determinants influencing the correctness of AI-generated responses compared to network-based shortest-path analyses. The empirical results demonstrate that Google’s Gemini 3 Pro model achieved superior performance, recording an accuracy rate of 65.0%. Regression analysis revealed that for both Gemini and GPT models, the volume of news articles associated with an attraction—representing media visibility—significantly increased the likelihood of accurate information provision. Notably, the Gemini model exhibited distinct sensitivity to geographic factors and text similarity metrics, suggesting a difference in how it processes spatial context compared to other models. Consequently, this study underscores the importance of high-quality AI-generated tourism data and offers significant contributions to the advancement of sophisticated personalized travel planning systems and GeoAI research focused on spatial problem-solving. Full article

Generative artificial intelligence and foundation models have changed machine learning by allowing systems to produce readable text, realistic images, and other multimodal content with little direct input from a user. Foundation models are large neural networks trained on very large and varied datasets, and they form the core of many current generative AI (GenAI) systems. Their rapid development has led to major advances in areas like natural language processing, computer vision, multimodal learning, and robotics. Examples include GPT, LLaMA, and diffusion-based architectures, such as models often used for image generation. Systems such as Stable Diffusion show this shift by illustrating how AI can interpret information, draw basic inferences, and produce new outputs using more than one type of data. This review surveys common foundation model architectures and examines what they can do in generative tasks. It reviews Transformer, diffusion, and multimodal architectures, focusing on methods that support scaling and transfer across domains. The paper also reviews key approaches to pretraining and fine-tuning, including self-supervised learning, instruction tuning, and parameter-efficient adaptation, which support these systems’ ability to generalize across tasks. In addition to the technical details, this review discusses how GenAI is being used for text generation, image synthesis, robotics, and biomedical research. The study also notes continuing challenges, such as the high computing and energy demands of large models, ethical concerns about data bias and misinformation, and worries about privacy, reliability, and responsible use of AI in real settings. This review brings together ideas about model design, training methods, and social implications to point future research toward GenAI systems that are efficient, easy to interpret, and reliable, while supporting scientific progress and ethical responsibility. Full article

Background: Large Language Models (LLMs) are reshaping medical research workflows. Objective: This narrative review synthesizes evidence on LLM applications across systematic reviews, scientific writing, and clinical research. Methods: We reviewed literature from 2023–2025 examining LLM applications in medical research, identified through PubMed, Scopus, Web of Science, arXiv, medRxiv, and Google Scholar. Studies reporting empirical findings, methodological evaluations, or systematic analyses of LLM applications were included; editorials and commentaries without empirical data were excluded. Results: In systematic reviews, LLMs achieve 80–94% data extraction accuracy and 40% reduction in screening workload, but show only slight-to-moderate agreement (κ = 0.16–0.43) in risk-of-bias assessment. In scientific writing, hallucination rates of 47–55% for fabricated references and over 90% prevalence of demographic bias require rigorous verification. For clinical research, LLMs assist with statistical coding and protocol development but require human validation. Critically, excessive reliance on automated tools may cause cognitive offloading that compromises analytical capabilities. Conclusions: LLMs are powerful but unstable tools requiring constant verification. Success depends on maintaining human-in-the-loop approaches that preserve critical thinking while leveraging AI efficiency. Full article

This meta-analysis investigates the outcomes associated with ChatGPT-integrated English instruction on non-linguistic outcomes in South Korea. Following the PRISMA 2020 guidelines, 22 experimental studies (k = 26, N = 1303) were synthesized using a random-effects model, which revealed a significant medium overall effect size (g = 0.55). Subgroup analyses showed medium-to-large effects in Affective (g = 0.67) and AI Literacy (g = 0.59) domains, but a small effect on Cognitive/Metacognitive outcomes (g = 0.17). Moderator analyses (e.g., educational level, duration) yielded no significant differences, suggesting a meaningful overall trend across contexts. However, the descriptive disparity between affective gains and cognitive growth indicates an ‘Affective–Cognitive Gap.’ Findings suggest that while ChatGPT is associated with a reduction in psychological barriers—partly by reducing mental effort—it may lead to ‘cognitive offloading’ without intentional pedagogical scaffolding. The study concludes that mere tool adoption is insufficient. To prevent a ‘plateau effect’ after the initial novelty wears off, instruction must evolve from passive use to ‘agentic engagement’ through structured metacognitive routines. These results emphasize the necessity of teacher-mediated integration to repurpose AI-afforded efficiency toward higher-order evaluative and critical thinking tasks. Full article

Background/Objectives: To date, limited focus has been given to the possible contribution of attachment theory to the comprehension of how paranoia and conspiracy beliefs may develop. Our study aimed to examine the potential mediating effects of the different adult attachment styles on the relationship between childhood trauma and paranoid/conspiracy thinking. Methods: This is a cross-sectional study that was conducted during September–January 2025 among Tunisian young adults (aged 18–35 years) from the general population. The Child Abuse Self Report Scale (CASRS-12), the Relationship Questionnaire (RQ), the eight-item Green et al., Paranoid Thoughts Scale (GPTS-8), and the Generic Conspiracist Beliefs Scale-5 (GCB-5) were administered to participants. Results: After controlling for potential confounders, analyses showed that secure attachment partially mediated the link between childhood trauma and paranoia (indirect effect: Beta = 0.001; Boot SE = 0.001) and conspiracy beliefs (indirect effect: Beta = 0.024; Boot SE = 0.01). On the other hand, preoccupied attachment acted as a significant mediator in the relationship between childhood trauma and paranoid thinking (indirect effect: Beta = 0.001; Boot SE = 0.001). In all these models, greater childhood trauma was directly related to higher paranoia and/or conspiracy thinking. Conclusions: Findings suggest that interventions and policies aimed at promoting a more secure attachment and addressing insecure attachment representations are likely to be effective in diminishing paranoia and conspiracy beliefs, especially for victims of childhood adversity. Full article

This paper presents a standard-oriented architecture for automating information security risk management (ISRM) using artificial intelligence. The study first evaluates eight international frameworks (including COBIT 2019, NIST SP 800-53, and ISO 31000) for automation suitability, identifying ISO/IEC 27005 as the optimal structural foundation. Based on these findings, an architecture integrating Natural Language Processing and machine learning to automate risk identification, assessment, and treatment is proposed. A core component is a decision-making module that combines expert reasoning with a Multi-LLM consensus mechanism to ensure reliability. To provide exploratory support for the proposed architecture, a comparative study using five state-of-the-art Large Language Models (ChatGPT, Gemini Advanced, Grok, Microsoft Copilot, and DeepSeek Chat) was conducted on a standardized risk identification task. The results highlight strong cross-model consensus patterns, providing exploratory evidence that LLMs may support expert-informed risk identification and reasoning tasks while acknowledging the current limitations in complex reasoning. This approach proposes a transparent architectural foundation for AI-driven ISRM whose scalability must be established through future prototype-based evaluation, thereby bridging the gap between rigid compliance standards and generative AI capabilities. Full article

Background: Accurate dietary assessment is vital for preventing malnutrition in aging populations, particularly in home-care settings. Although Large Multimodal Models (LMMs) for nutrient estimation are evolving, their nutrient-specific accuracy requires rigorous validation. Methods: Fifteen standardized hospital meals were photographed under controlled conditions (90-degree angle, 500 lux). Ground truth values were determined by direct weighing. Estimates for energy and macronutrients were performed by 10 registered dietitians (RDs) and 10 AI models (including ChatGPT-4o and Gemini 1.5 Pro). Accuracy was assessed using Pearson’s correlation, Mean Absolute Error (MAE), and Bland–Altman analysis to quantify systematic bias. Results: For energy and carbohydrates, RDs and top-performing AI models (notably ChatGPT-4o and Gemini 1.5 Pro) demonstrated practical accuracy (r > 0.8, frequently within ±10% range). However, accuracy for protein and lipids was significantly lower across all AI models. Specifically, all AI models exhibited a substantial systematic overestimation of lipids (Mean Bias > +20%, p < 0.01), highlighting a critical “invisible nutrient” bias. Conclusions: Current AI tools show potential for caloric and carbohydrate monitoring but struggle with lipid and protein density. These findings emphasize the need for human–AI collaboration (“human-in-the-loop”) and the integration of cooking metadata to improve clinical utility in geriatric nutrition. Full article

Multimodal named entity recognition (MNER) leverages both textual and visual information to improve entity recognition, particularly in unstructured scenarios such as social media. While existing approaches predominantly rely on raster images (e.g., JPEG, PNG), scalable vector graphics (SVG) offer unique advantages in resolution independence and structured semantic representation—an underexplored potential in multimodal learning. To fill this gap, we propose MNER-SVG, the first framework that incorporates SVG as a visual modality and enhances it with ChatGPT-generated auxiliary knowledge. Specifically, we introduce a Multimodal Similar Instance Perception Module that retrieves semantically relevant examples and prompts ChatGPT to generate contextual explanations. We further construct a Full-Text Graph and a Multimodal Interaction Graph, which are processed via Graph Attention Networks (GATs) to achieve fine-grained cross-modal alignment and feature fusion. Finally, a Conditional Random Field (CRF) layer is employed for structured decoding. To support evaluation, we present SvgNER, the first MNER dataset annotated with SVG-specific visual content. Extensive experiments demonstrate that MNER-SVG achieves state-of-the-art performance with an F1 score of 82.23%, significantly outperforming both text-only and existing multimodal baselines. This work validates the feasibility and potential of integrating vector graphics and large language model-generated knowledge into multimodal NER, opening a new research direction for structured visual semantics in fine-grained multimodal understanding. Full article

This study proposes a Natural Language Processing (NLP)-based evaluation framework to examine the linguistic consistency of large language model (LLM)-generated personalized texts over time. NLP metrics were used to quantify and compare linguistic patterns across repeated generations produced using identical prompts. In Experiment 1, internal reliability was examined across 10 repeated generations from four LLMs (Claude, Llama, Gemini, and ChatGPT), applied to 10 scientific texts tailored for a specific reader profile. Linear mixed-effects models showed no effect of repeated generation on linguistic features (e.g., cohesion, syntactic complexity, lexical sophistication), suggesting short-term consistency across repeatedly generated outputs. Experiment 2 examined linguistic variation across model updates of GPT-4o (October 2024 vs. June 2025) and GPT-4.1 (June 2025). Significant variations were observed across outputs from different model versions. GPT-4o (June 2025) generated more concise but cohesive texts, whereas GPT-4.1 (June 2025) generated outputs that are more academic, lexically sophisticated, and complex in syntax. Given the rapid evolution of LLMs and the lack of standardized methods for tracking output consistency, the current work demonstrates one of the applications of NLP-based evaluation approaches for monitoring meaningful linguistic shifts across model updates over time. Full article

Item description information plays a crucial role in helping users understand the basic situation of an item and is also vital auxiliary information in recommendation systems. Traditional methods obtain this data through platform backend data or web scraping techniques, but these data are often static, relatively fixed, and insufficiently descriptive. In recent years, large language models (LLMs) like generative pre-trained transformer (GPT) have become powerful tools in natural language processing, bringing new hope for LLM-based recommendations. However, does the text information generated by large language models help improve recommendation accuracy? How can the information produced by generative artificial intelligence be integrated with existing multi-source heterogeneous information? In this paper, we propose a novel deep hybrid recommendation method for multimodal information integrating content generated by large language models (DML). We first explore the use of large language models to generate detailed descriptive information about movies. Next, we perform a weighted fusion of the generated text information with existing movie category information and user demographic data, among other multi-source heterogeneous information. Finally, we use the fused information to predict movie ratings. The results indicate that the multimodal information deep hybrid recommendation method, which integrates content generated by large language models, provides substantial evidence of superior performance relative to existing baseline models. Full article

Task analysis is a foundational component of instructional design in special education, yet it can impose substantial time and cognitive demands on teachers. Artificial intelligence (AI) tools such as ChatGPT may provide support for instructional planning tasks by assisting educators in generating and organizing task sequences. This study examined the effectiveness, readability, time efficiency, and acceptability of ChatGPT-assisted task analysis compared to a traditional task analysis method. Thirty-two special education teachers participated in a randomized between-groups study in which they developed task analyses using either a traditional approach or ChatGPT supported by a structured interaction protocol. Task analyses were evaluated based on alignment with expert-developed models, readability, and development time, and teachers’ perceptions of acceptability were also examined. Results indicated that ChatGPT-assisted task analyses required significantly less development time while demonstrating strong alignment with expert-generated models. Readability levels and the number of task steps were similar across groups. Teachers who used ChatGPT also reported positive perceptions regarding the usefulness and acceptability of AI assistance in instructional planning. These findings suggest that AI-assisted tools may support teachers in developing task analyses more efficiently while maintaining instructional clarity. However, given the exploratory nature of the study and the limited sample, further research is needed to examine how AI-assisted task analysis may influence instructional practice and student learning outcomes in special education. Full article