Background: The COVID-19 pandemic has been accompanied by an “infodemic,” where the rapid spread of misinformation has exacerbated public health challenges. Traditional fact-checking methods, though effective, are time-consuming and resource-intensive, limiting their ability to combat misinformation at scale. Large language models (LLMs) such as GPT-4 offer a more scalable solution, but their susceptibility to generating hallucinations—plausible yet incorrect information—compromises their reliability. Objective: This study aims to enhance the accuracy and reliability of COVID-19 fact-checking by integrating a retrieval-augmented generation (RAG) system with LLMs, specifically addressing the limitations of hallucination and context inaccuracy inherent in stand-alone LLMs. Methods: We constructed a context dataset comprising approximately 130,000 peer-reviewed papers related to COVID-19 from PubMed and Scopus. This dataset was integrated with GPT-4 to develop multiple RAG-enhanced models: the naïve RAG, Lord of the Retrievers (LOTR)–RAG, corrective RAG (CRAG), and self-RAG (SRAG). The RAG systems were designed to retrieve relevant external information, which was then embedded and indexed in a vector store for similarity searches. One real-world dataset and one synthesized dataset, each containing 500 claims, were used to evaluate the performance of these models. Each model’s accuracy, F1-score, precision, and sensitivity were compared to assess their effectiveness in reducing hallucination and improving fact-checking accuracy. Results: The baseline GPT-4 model achieved an accuracy of 0.856 on the real-world dataset. The naïve RAG model improved this to 0.946, while the LOTR-RAG model further increased accuracy to 0.951. The CRAG and SRAG models outperformed all others, achieving accuracies of 0.972 and 0.973, respectively. The baseline GPT-4 model reached an accuracy of 0.960 on the synthesized dataset. The naïve RAG model increased this to 0.972, and the LOTR-RAG, CRAG, and SRAG models achieved an accuracy of 0.978. These findings demonstrate that the RAG-enhanced models consistently maintained high accuracy levels, closely mirroring ground-truth labels and significantly reducing hallucinations. The CRAG and SRAG models also provided more detailed and contextually accurate explanations, further establishing the superiority of agentic RAG frameworks in delivering reliable and precise fact-checking outputs across diverse datasets. Conclusions: The integration of RAG systems with LLMs substantially improves the accuracy and contextual relevance of automated fact-checking. By reducing hallucinations and enhancing transparency by citing retrieved sources, this method holds significant promise for rapid, reliable information verification to combat misinformation during public health crises.

In the context of global climate change and the rapid development of the sports industry, increasingly frequent flooding has become a significant challenge for Chinese cities today and one of the hot issues in risk management for sports events. In order to reveal the impact of flooding on the spatial accessibility of emergency response to stadiums in the central urban area within the outer ring of Shanghai, this article evaluates the accessibility of medical emergency services in the central urban area under the impact of flooding based on flood scenario simulations and GIS network analysis. Results show that under the different flooding scenarios, as the intensity of flooding increases, urban road traffic in flood-prone areas is blocked by ponding, and some stadiums are inaccessible to emergency vehicles. The scope of emergency response services for some medical institutions along the Huangpu River is significantly reduced, while emergency response times for stadiums in the riverside area are delayed considerably. Some stadiums are unable to access emergency medical services. The study’s results can offer a case reference for upgrading the level of emergency management of stadiums under urban-scale flooding and optimizing the quality of regional medical emergency services.

The Olympic Village has evolved over time to become a full-fledged urban system that encompasses not just the temporary housing needs of athletes participating in the Olympic Games but also infrastructure development, governance coordination, sustainability planning, and long-term community transformation. Although there are many studies of Olympic Villages, they typically focus on a single aspect of them such as the physical redevelopment of an area, the way that it was governed during the Olympics, and what happened to the Village after the Games ended. Existing studies have examined various dimensions separately, but fewer studies integrate these dimensions into a unified lifecycle framework. To fill this gap, this study proposed the Physical-Relational-Legacy (P-R-L) framework to help explain the changes in form that occur in Olympic Villages over their lifecycle. Using a Straussian grounded theory approach and comparative case study methodology, the study investigated six Olympic events: the Summer Olympics of 2008, 2012, 2016, 2020, and 2024, and the Winter Olympics of 2022. The data collected for this study included official Olympic reports, operational manuals, planning documents, and other types of published materials related to legacy. Using systematic open, axial, and selective coding supported by constant comparison, the study inductively developed the three interrelated dimensions: Physical, Relational, and Legacy. The results of this study indicated that the transformation of an Olympic Village is not simply a matter of developing some areas after the games end. Rather, it involves a series of interactions between changing infrastructure systems, governance mechanisms, and ongoing legacy development processes. Moreover, each of these dimensions interacts with one another at all stages of an Olympic Village’s lifecycle instead of occurring sequentially. The study contributes by providing an integrated analytical framework explaining how physical adaptation, governance coordination, and legacy formation interact throughout the Olympic Village lifecycle.

BACKGROUND The COVID-19 pandemic has been accompanied by an “infodemic,” where the rapid spread of misinformation has exacerbated public health challenges. Traditional fact-checking methods, though effective, are time-consuming and resource-intensive, limiting their ability to combat misinformation at scale. Large language models (LLMs) such as GPT-4 offer a more scalable solution, but their susceptibility to generating hallucinations—plausible yet incorrect information—compromises their reliability. OBJECTIVE This study aims to enhance the accuracy and reliability of COVID-19 fact-checking by integrating a retrieval-augmented generation (RAG) system with LLMs, specifically addressing the limitations of hallucination and context inaccuracy inherent in stand-alone LLMs. METHODS We constructed a context dataset comprising approximately 130,000 peer-reviewed papers related to COVID-19 from PubMed and Scopus. This dataset was integrated with GPT-4 to develop multiple RAG-enhanced models: the naïve RAG, Lord of the Retrievers (LOTR)–RAG, corrective RAG (CRAG), and self-RAG (SRAG). The RAG systems were designed to retrieve relevant external information, which was then embedded and indexed in a vector store for similarity searches. One real-world dataset and one synthesized dataset, each containing 500 claims, were used to evaluate the performance of these models. Each model’s accuracy, F1-score, precision, and sensitivity were compared to assess their effectiveness in reducing hallucination and improving fact-checking accuracy. RESULTS The baseline GPT-4 model achieved an accuracy of 0.856 on the real-world dataset. The naïve RAG model improved this to 0.946, while the LOTR-RAG model further increased accuracy to 0.951. The CRAG and SRAG models outperformed all others, achieving accuracies of 0.972 and 0.973, respectively. The baseline GPT-4 model reached an accuracy of 0.960 on the synthesized dataset. The naïve RAG model increased this to 0.972, and the LOTR-RAG, CRAG, and SRAG models achieved an accuracy of 0.978. These findings demonstrate that the RAG-enhanced models consistently maintained high accuracy levels, closely mirroring ground-truth labels and significantly reducing hallucinations. The CRAG and SRAG models also provided more detailed and contextually accurate explanations, further establishing the superiority of agentic RAG frameworks in delivering reliable and precise fact-checking outputs across diverse datasets. CONCLUSIONS The integration of RAG systems with LLMs substantially improves the accuracy and contextual relevance of automated fact-checking. By reducing hallucinations and enhancing transparency by citing retrieved sources, this method holds significant promise for rapid, reliable information verification to combat misinformation during public health crises.