Ying Lu

With rapid urbanization and expanding infrastructure, construction contract disputes are increasing in volume and complexity, challenging traditional adjudication. This study proposes a domain-specific legal artificial intelligence (AI) system for construction contract disputes via hybrid knowledge integration based on the retrieval-augmented generation (RAG) paradigm, integrating five core legal texts and 500 adjudication cases within a dual-engine architecture. The knowledge base encodes legal concepts, relations, and rules to enable structured semantic inference. The DeepSeek-R1 reasoning engine analyzes case facts and legal logic via constrained generation, while the BGE-M3 retrieval module matches legal provisions and precedents using multivector indexing. A tripartite evaluation framework—semantic similarity, legal provision citation accuracy, and issue prediction F1 score—validates system performance. The hybrid knowledge model outperforms single-source models, achieving scores of 0.736, 0.952, and 0.937, respectively, while significantly reducing judicial document generation time. This study offers a theoretical and empirical basis for legal AI in Chinese construction disputes, demonstrating how integrating diverse legal knowledge enhances intelligent judicial assistance within China’s jurisdiction. It also provides a scalable methodological reference for the advancement of smart justice, with explicit recognition of its current jurisdictional limitations.

With the widespread application of artificial intelligence in emergency management, AI-recommended evacuation paths have emerged as a crucial means of enhancing evacuation efficiency during building fires. However, the mechanisms underlying users’ adoption of AI-recommended paths in real-world disaster scenarios remain unclear. This study proposes an integrated framework that combines Q-learning-based path generation, fire dynamics simulation, virtual reality (VR) experiments, and the Technology Acceptance Model (TAM). A three-story academic building was selected as the experimental site. A dynamic fire temperature field was simulated using PyroSim and processed via grid mapping. The Q-learning algorithm was employed to generate optimal evacuation paths. Subsequently, an immersive VR experimental platform was developed, featuring six types of path conflict tasks. Data were collected from 82 participants, encompassing behavioral choices and subjective evaluations. Structural equation modeling (SEM) was conducted using AMOS, revealing that perceived ease of use and perceived usefulness significantly influenced user attitudes and behavioral intentions, which in turn positively predicted actual path adoption. The results validate the applicability of TAM in fire evacuation scenarios and elucidate the cognitive-behavioral transformation mechanism. This research provides both theoretical foundations and empirical evidence to support the design and optimization of intelligent evacuation systems.

Effective fire safety training is critical to mitigating property losses and human injuries and deaths. Augmented Reality (AR) is an emerging solution for safety training, offering a dynamic and interactive learning environment. Despite the emergence of various AR training prototypes demonstrating potential advantages, there is limited research on how teaching theories impact the efficacy of augmented reality. This study aims to integrate and test two teaching theories, directive instruction and productive failure, into AR-based fire safety training. Two novel AR prototypes, directive instruction training and productive failure training, were designed and prototyped in this work. These two prototypes were tested in a controlled experiment involving 68 participants. A comparison was carried out, focusing on knowledge acquisition, knowledge retention, intrinsic motivation, and self-efficacy. The results indicate that productive failure training has better performance in enhancing knowledge acquisition and knowledge retention. However, both prototypes perform equally in the overall learning experience. As such, these findings offer valuable insights into future augmented reality development for safety training.

With the increasing frequency of extreme weather events such as typhoons, the stability and resilience of the inter-city population mobility network is facing serious challenges. However, existing studies have failed to comprehensively reveal the dynamics of the overall regional network, especially the changes in network structure and drivers under disaster situations. Therefore, based on the Pressure-State-Response (PSR) framework and the complex network model, this paper constructs an assessment model for the characteristics of the population mobility network in the Yangtze River Delta (YRD) region under the influence of Typhoon Bebinca, in order to analyze the impact of the typhoon on the network structure and the changes in its driving factors. The study shows that the typhoon had a significant impact on the characteristics and patterns of the population mobility network in the YRD. First, during the typhoon, the network size and density decreased, long-distance mobility decreased, and mobility was concentrated among neighboring cities, showing the characteristics of “small world”. Second, after the typhoon landfall, the population inflow to the central city dropped significantly. Then, network connectivity declines during the typhoon, but modularity increases. Finally, in terms of drivers, differences in wind, air quality, residents' economic conditions, and the level of pollution control between cities significantly and negatively affect the network. In summary, this paper enriches the theoretical study of population mobility networks in disaster situations, and provides scientific basis and policy recommendations for enhancing emergency management capabilities and optimizing post-disaster recovery strategies in the YRD region.

Introduction: Expatriates are facing more stressors, such as cross-cultural adjustment, global political instability, family separation, health concern. The black swan events of the pandemic and the Russian-Ukrainian war have posed significant challenges in the current international environment. Adapting to an expatriate environment as soon as possible is critical to expatriate success. This study aims to examine the factors that affect expatriate adjustment through psychological resilience. Methods: Guided by person-environment (p-e) fit theory, an expatriate adjustment framework based on psychological resilience is proposed, and 309 valid sample data are used for structural equation model (SEM) analysis. Results: The results show that expatriate adjustment is a psychological process based on the development of resilience. Social support plays a buffering role in dealing with environmental deviations induced stressors. The person-environment transactional process is the most critical adjustment process. Discussion: The development of expatriate adjustment is divided into four stages (shock, buffer, adjustment, mastery) consistent with resilience development. Project managers can take different expatriate management strategies from multiple aspects. Finally, this study proposes the U-curve hypothesis of expatriates’ psychological resilience development aligned with the U-curve process of expatriate adjustment for future research.