Explore the vast range of titles available.

Integrating Large Language Models (LLMs) with Knowledge Graphs (KGs) enhances the interpretability and performance of AI systems. This research comprehensively analyzes this integration, classifying approaches into three fundamental paradigms: KG-augmented LLMs, LLM-augmented KGs, and synergized frameworks. The evaluation examines each paradigm’s methodology, strengths, drawbacks, and practical applications in real-life scenarios. The findings highlight the substantial impact of these integrations in fundamentally improving real-time data analysis, efficient decision-making, and promoting innovation across various domains. In this paper, we also describe essential evaluation metrics and benchmarks for assessing the performance of these integrations, addressing challenges like scalability and computational overhead, and providing potential solutions. This comprehensive analysis underscores the profound impact of these integrations on improving real-time data analysis, enhancing decision-making efficiency, and fostering innovation across various domains.

Entrepreneurship education has evolved to meet the demands of a dynamic business environment, necessitating innovative teaching methods to prepare entrepreneurs for market uncertainties. Large Language Models (LLMs) like the Generative Pre-trained Transformer 4 (GPT-4), recognized for their exceptional performance on public datasets, are examined in this study for their potential adaptability and interactivity nature, which align well with the dynamic nature of entrepreneurship learning. The interaction with LLMs can be enhanced by using effective prompt engineering techniques (PETs) that allow for crafting precise queries to elicit accurate and relevant responses for entrepreneurial learning. Critical concerns regarding the use of GPT-4 and conversational agents in entrepreneurship courses include the reliability and accuracy of data sources, the necessity for specific, real-time data for effective decision-making, and the lack of in-depth exploration of effective prompting strategies tailored to entrepreneurship education. Addressing these issues, this study aims to identify and compare the quality output of currently available PETs, develop innovative PETs that are well-aligned with entrepreneurial learning, and provide guidelines on how to fully utilize LLMs and conversational agents with Retrieval Augmentation Generation (RAG) technology in entrepreneurship education. The combination of conversational agents and RAG technology into a hybrid innovative approach overcomes inherent limitations in each technology separately and enhances efficiency and relevancy in entrepreneurship education through exact, dynamic interactions and advanced memory capabilities. The findings of the study significantly contribute to the field of entrepreneurship by offering practical insights for students and educators on enhancing the entrepreneurship learning experience, particularly by utilizing cutting-edge technology to improve data relevance and answer accuracy in entrepreneurial queries and scenarios.

The rapid advancements in natural language processing, particularly the development of large language models (LLMs), have opened new avenues for managing complex clinical text data. However, the inherent complexity and specificity of medical texts present significant challenges for the practical application of prompt engineering in diagnostic tasks. This paper explores LLMs with new prompt engineering technology to enhance model interpretability and improve the prediction performance of pulmonary disease based on a traditional deep learning model. A retrospective dataset including 2965 chest CT radiology reports was constructed. The reports were from 4 cohorts, namely, healthy individuals and patients with pulmonary tuberculosis, lung cancer, and pneumonia. Then, a novel prompt engineering strategy that integrates feature summarization (F-Sum), chain of thought (CoT) reasoning, and a hybrid retrieval-augmented generation (RAG) framework was proposed. A feature summarization approach, leveraging term frequency-inverse document frequency (TF-IDF) and K-means clustering, was used to extract and distill key radiological findings related to 3 diseases. Simultaneously, the hybrid RAG framework combined dense and sparse vector representations to enhance LLMs' comprehension of disease-related text. In total, 3 state-of-the-art LLMs, GLM-4-Plus, GLM-4-air (Zhipu AI), and GPT-4o (OpenAI), were integrated with the prompt strategy to evaluate the efficiency in recognizing pneumonia, tuberculosis, and lung cancer. The traditional deep learning model, BERT (Bidirectional Encoder Representations from Transformers), was also compared to assess the superiority of LLMs. Finally, the proposed method was tested on an external validation dataset consisted of 343 chest computed tomography (CT) report from another hospital. Compared with BERT-based prediction model and various other prompt engineering techniques, our method with GLM-4-Plus achieved the best performance on test dataset, attaining an F1-score of 0.89 and accuracy of 0.89. On the external validation dataset, F1-score (0.86) and accuracy (0.92) of the proposed method with GPT-4o were the highest. Compared to the popular strategy with manually selected typical samples (few-shot) and CoT designed by doctors (F1-score=0.83 and accuracy=0.83), the proposed method that summarized disease characteristics (F-Sum) based on LLM and automatically generated CoT performed better (F1-score=0.89 and accuracy=0.90). Although the BERT-based model got similar results on the test dataset (F1-score=0.85 and accuracy=0.88), its predictive performance significantly decreased on the external validation set (F1-score=0.48 and accuracy=0.78). These findings highlight the potential of LLMs to revolutionize pulmonary disease prediction, particularly in resource-constrained settings, by surpassing traditional models in both accuracy and flexibility. The proposed prompt engineering strategy not only improves predictive performance but also enhances the adaptability of LLMs in complex medical contexts, offering a promising tool for advancing disease diagnosis and clinical decision-making.

This study examined undergraduate nursing students’ experiences and acceptance of NurCaseAI, a GenAI-based interactive case study system powered by retrieval-augmented generation (RAG) technology. The investigation was guided by the Artificially Intelligent Device Use Acceptance (AIDUA) model. Clinical reasoning is a critical skill for undergraduate nursing students. The emergence of generative AI (GenAI) offers opportunities to support clinical learning; however, concerns about content accuracy and conversational quality remain. Retrieval-augmented generation (RAG) has been proposed to enhance GenAI systems by grounding outputs in validated nursing knowledge, improving reliability for educational use. A qualitative exploratory design was adopted. 16 third-year nursing students from a Chinese university were purposively selected after completing three thyroid perioperative nursing case scenarios using NurCaseAI. Semi-structured interviews were conducted, transcribed and analyzed using thematic analysis. Five key themes were identified: 1) Perceived system capabilities and limitations as students viewed NurCaseAI as an effective learning aid with analytical constraints; 2) Learning enhancement benefits where RAG content aligned well with coursework and guidelines; 3) Interaction efficiency challenges including initial barriers in system adaptation; 4) Limited emotional engagement as students viewed the system as task-oriented rather than empathetic; 5) Strong system acceptance with willingness to continue using NurCaseAI despite limitations. NurCaseAI demonstrates promise as a supplementary GenAI tool for nursing education, particularly in perioperative and case-based learning. The AIDUA model offered a valuable lens to understand students’ acceptance patterns. Future improvements should enhance affective interaction and support more complex clinical communication.