Yang Liu

Globalization and informatization are reshaping human life and social behaviors. The purpose is to explore the worldwide strategies to cultivate international talents with a global vision. As a global language with the largest population, English, and especially its learning effect, have always been the major concerns of scholars and educators. This work innovatively studies the combination of immersion-based English teaching with virtual reality technology. Then, based on the experimental design mode, 106 students from a Chinese school were selected for a quasi-experimental study for 16 weeks. The collected data were analyzed by computer statistical software, and hypotheses are verified. The results showed that there is a significantly positive correlation between VR and immersion-based language teaching. There is a significantly positive correlation between immersion-based language teaching and academic achievement, and VR is positively correlated with learning outcome. Compared with other state-of-art research methods, this work modifies the students’ oral test through the analysis and comparison with the system database, and the students’ learning effect is greatly improved. Finally, some suggestions are put forward according to the research results to provide an experimental reference for English teachers and future linguistics teaching.

Legal large language models (LLMs) deployed in high-stakes judicial settings must exhibit robustness against non-substantive linguistic variations while preserving acute sensitivity to legally determinative facts and norms. This study investigates this robustness–reliability trade-off within the context of Chinese legal tasks. We curate a dataset of 5,000 Chinese judicial question–answer pairs and generate semantic-preserving adversarial rewrites, retaining only those validated by an embedding-based semantic consistency filter. Holding the total training budget and fine-tuning protocol constant, we fine-tune model variants that differ exclusively in their injection ratio of these verified rewrites, establishing seven distinct injection groups (G0–G6). We evaluate model reliability utilizing a composite protocol that incorporates objective accuracy on exam-style questions, expert evaluations of open-ended responses, and embedding-based semantic similarity. For trademark infringement reasoning tasks, we additionally assess verdict accuracy and rationale quality. Across varying model capacities (4B, 20B, and 32B backbones, including Qwen3-4B, GPT-OSS-20B, and Qwen3-VL-32B-Instruct) and both evaluated tasks, our findings reveal an inverted‑U relationship: moderate robustness data injection enhances reliability, whereas excessive injection degrades overall performance and induces characteristic failure modes, such as the attenuation of legally salient distinctions, the generation of boilerplate rationales, and overly cautious abstention. These findings substantiate “moderate robustness injection” as a practical heuristic and underscore a broader principle of differential sensitivity—achieving insensitivity to superficial variations without blunting the model’s sensitivity to legally decisive elements.

Recently, infrared human action recognition has attracted increasing attention for it has many advantages over visible light, that is, being robust to illumination change and shadows. However, the infrared action data is limited until now, which degrades the performance of infrared action recognition. Motivated by the idea of transfer learning, an infrared human action recognition framework using auxiliary data from visible light is proposed to solve the problem of limited infrared action data. In the proposed framework, we first construct a novel Cross-Dataset Feature Alignment and Generalization framework to map the infrared data and visible light data into a common feature space, where Kernel Manifold Alignment and a dual aligned-to-generalized encoders model are employed to represent the feature. Then, a support vector machine is trained, using both the infrared data and visible light data, and can classify the features derived from infrared data. The proposed method is evaluated on InfAR, which is a publicly available infrared human action dataset. To build up auxiliary data, we set up a novel visible light action dataset XD145. Experimental results show that the proposed method can achieve state-of-the-art performance compared with several transfer learning and domain adaptation methods.

Z-curve analysis is intended to diagnose the credibility of research results, but its interpretation and statistical properties are often misunderstood. We clarify that the Expected Discovery Rate (EDR; i.e. average observed power) is conceptually distinct from average pre-data power and lacks a clear link to credibility because it reflects both the average pre-data power and the estimated average population effect size. In our review of 37 articles reporting 278 Z-curve applications, 77.3% concluded publication bias, yet 48.2% did not state whether the p-values analyzed reflected focal findings, and 69.1% may have violated the assumption of independent p-values. Simulations further demonstrate that Z-curve estimators can be biased and inconsistent, failing to follow the Law of Large Numbers and potentially producing misleading conclusions. We also question claims made by Soto and Schimmack [Credibility of results in emotion science: A Z-curve analysis of results in the journals Cognition & Emotion and Emotion. Cognition and Emotion, (2025), 39(8), 1803–1819], regarding the credibility of emotion-science findings, noting that such conclusions should be interpreted cautiously given the limitations of Z-curve estimates. Overall, we do not recommend using Z-curve to evaluate research findings. Traditional meta-analytic methods remain more appropriate and reliable for statistical conclusions about focal research findings.

In late 2019, the COVID-19 pandemic began to spread over the world, causing millions of deaths. In the first few months of the pandemic, several countries prevented the spread of the pandemic successfully. By contrast, the pandemic in many other countries was not controlled well. For example, India encountered a second serious outbreak of COVID-19 from April 2021 due to the poor resistance measures implemented by the government. To figure out the effective countermeasures to the pandemic, this research proposes a COVID-19 pandemic and its response system, which consists of the infection subsystem, the quarantine subsystem, and the medical subsystem. On this basis, an improved SEIR-SD model is established which is utilized to analyze the response measures to the pandemic quantitatively. This model successfully simulates the actual epidemic scenarios in Wuhan, which verifies its effectiveness. Afterward, the impact of hospital administration rate, quarantine rate, average contact number, and contact infection rate on the cumulative number of infections and deaths are analyzed by simulation. The results show that both the medical and administrative efforts, especially in the early stage of the epidemic, are significant in reducing the number of infections and shortening the epidemic period. In the medical aspect, the more stringent quarantine brings the earlier inflection point of the epidemic; more importantly, improving the treatment rate significantly reduces the scale of the epidemic. In the administrative aspect, enforcing individual protection and strict community closure can effectively cut off the transmission of the virus and curb the spread of the epidemic. Finally, this research proposes several practical suggestions in response to the COVID-19 pandemic. The main contribution of this research is that the effects of different response measures on the number of new infections daily and the cumulative number of deaths of a country or region in the COVID-19 pandemic are estimated quantitatively based on modeling and simulation.

The recent rapid development of information technology, such as sensing technology, communications technology, and database, allows us to use simulation experiments for analyzing serious accidents caused by hazardous chemicals. Due to the toxicity and diffusion of hazardous chemicals, these accidents often lead to not only severe consequences and economic losses, but also traffic jams at the same time. Emergency evacuation after hazardous chemical accidents is an effective means to reduce the loss of life and property and to smoothly resume the transport network as soon as possible. This paper considers the dynamic changes of the hazardous chemicals’ concentration after their leakage and simulates the diffusion process. Based on the characteristics of emergency evacuation of hazardous chemical accidents, we build a mixed-integer programming model and design a heuristic algorithm using network optimization and diffusion simulation. We then verify the validity and feasibility of the algorithm using Jinan, China, as a computational example. In the end, we compare the results from different scenarios to explore the key factors affecting the effectiveness of the evacuation process.

Interpersonal physiological synchrony has been consistently found during collaborative tasks. However, few studies have applied synchrony to predict collaborative learning quality in real classroom. To explore the relationship between interpersonal physiological synchrony and collaborative learning activities, this study collected electrodermal activity and heart rate during naturalistic class sessions and compared the physiological synchrony between independent task and group discussion task. The students were recruited from a renowned university in China. Since each student learn differently and not everyone prefers collaborative learning, participants were sorted into collaboration and independent dyads based on their collaborative behaviors before data analysis. The result showed that, during group discussions, high collaboration pairs produced significantly higher synchrony than low collaboration dyads. Given the equivalent engagement level during independent and collaborative tasks, the difference of physiological synchrony between high and low collaboration dyads was triggered by collaboration quality. Building upon this result, the classification analysis was conducted, indicating that EDA synchrony can identify different levels of collaboration quality.

The behavior of somatic stem cells is regulated by their niche. Interaction between hematopoietic stem cells (HSCs) and their niches are a representative model to understand stem cell‐niche interplay. Here, we provide an overview of crosstalk between HSCs and their niches in bone marrow and extramedullary organs following the life journey of HSCs from emergence, development, maturation until aging. We highlight the unique differences of HSC niches in different life stages within various organs focusing on recent literature to propose new speculations and hypotheses.

Trend and change-point analyses of water quality time series data have important implications for pollution control and environmental decision-making. This paper developed a new approach to assess trends and change-points of water quality parameters by integrating locally weighted polynomial regression and segmented regression. Firstly, LWPR was used to pretreat the original water quality data into a smoothed time series to represent the long-term trend of water quality. Then, SegReg was used to identify the long-term trends and change-points of the smoothed time series. Finally, statistical tests were applied to determine the significance of the long-term trends and change-points. The efficacy of this approach was validated using a 10-year record of total nitrogen and chemical oxygen demand from Shanxi Reservoir watershed in eastern China. Results showed that this approach was straightforward and reliable for assessment of long-term trends and change-points on irregular water quality datasets. The reliability was verified by statistical tests and practical considerations for Shanxi Reservoir watershed. The newly developed integrated LWPR-SegReg approach is not only limited to the assessment of trends and change-points of water quality parameters but also has a broad application to other fields with long-term time series records.

Researchers have suggested that infants exhibiting baby schema are considered cute. These similar studies have mainly focused on changes in overall baby schema facial features. However, whether a change in only eye size affects the perception of cuteness across different facial expressions and ages has not been explicitly evaluated until now. In the present study, a paired comparison method and 7-point scale were used to investigate the effects of eye size on perceived cuteness across facial expressions and ages. The results show that stimuli with large eyes were perceived to be cuter than both unmanipulated eyes and small eyes across all facial expressions and age groups. This suggests not only that the effect of baby schema on cuteness is based on changes in a set of features but also that eye size as an individual feature can affect the perception of cuteness.

Copyright © 2015 by ASME.This paper replaces the hinged pivots of an eight-bar linkage with flexure joints in order to achieve a flexure-connected linkage system that guides rectilinear movement of its end-effector. The goal is a linkage design that can be reduced in size to provide a suspension for the proof masses of a MEMS gyroscope. The symmetric design of the linkage and its long travel relative to other MEMS suspensions has the potential to provide a number of advantages, such as the reduction of quadrature error. The design presented yields 0.1% deviation over its range of movement. An example also presents the driving linkage of the MEMS gyroscope, which is also designed as flexure connected linkage.

© 2015 by The American Society for Biochemistry and Molecular Biology Inc. Plasminogen activator inhibitor 1 level is extremely elevated in the edematous fluid of acutely injured lungs and pleurae. Elevated PAI-1 specifically inactivates pulmonary urokinase-type and tissue-type plasminogen activators. We hypothesized that plasminogen activation and fibrinolysis may alter epithelial sodium channel activity, a key player in clearing edematous fluid. Two-chain urokinase has been found to strongly stimulate heterologous human αβγ ENaC activity in a dose- and time-dependent manner. This activity of tcuPA was completely ablated by PAI-1. Furthermore, a mutation of the active site of the enzyme also prevented ENaC activation. By comparison, three truncation mutants of the amino-terminal fragment of tcuPA still activated ENaC. uPA enzymatic activity was positively correlated with ENaC current amplitude prior to reaching the maximal level. In sharp contrast to uPA, neither single-chain tPA nor derivatives, including two-chain tPA and tenecteplase, affected ENaC activity. Furthermore, γ but not α subunit of ENaC was proteolytically cleaved at by tcuPA. In summary, the underlying mechanisms of urokinase-mediated activation of ENaC include release of self-inhibition, proteolysis of γ ENaC, incremental increase in opening rate, and activation of closed channels. This study for the first time demonstrates multifaceted mechanisms for uPA-mediated up-regulation of ENaC, which form the cellular and molecular rationale for the beneficial effects of urokinase in mitigating mortal pulmonary edema and pleural effusions.

Background: Anopheles sinensis is a major malaria vector in China and other Southeast Asian countries, and it is becoming increasingly resistant to the insecticides used for agriculture, net impregnation, and indoor residual spray. Very limited genomic information on this species is available, which has hindered the development of new tools for resistance surveillance and vector control. We used the 454 GS FLX system and generated expressed sequence tag databases of various life stages of An. sinensis, and we determined the transcriptional differences between deltamethrin resistant and susceptible mosquitoes.Results: The 454 GS FLX transcriptome sequencing yielded a total of 624,559 reads with the pooled An. sinensis mosquitoes across various development stages. The de novo assembly generated 33,411 contigs with average length of 493 bp. A total of 8,057 ESTs were generated with Gene Ontology and Kyoto Encyclopedia of Genes and Genomes annotation. A total of 2,131 ESTs were differentially expressed between deltamethrin resistant and susceptible mosquitoes collected from the same field site in Jiangsu, China. Among these differentially expressed ESTs, a total of 294 pathways were mapped to the KEGG database, with the predominant ESTs belonging to metabolic pathways. Furthermore, a total of 2,408 microsatellites and 15,496 single nucleotide polymorphisms were identified.Conclusions: The annotated EST and transcriptome databases provide a valuable genomic resource for further genetic studies of this important malaria vector species. The differentially expressed ESTs associated with insecticide resistance identified in this study lay an important foundation for further functional analysis. The identified microsatellite and SNP markers will provide useful tools for future population genetic and comparative genomic analyses of malaria vectors. © 2014 Zhu et al.; licensee BioMed Central Ltd.

© 2014 Li et al. Aedes albopictus is a very invasive and aggressive insect vector that causes outbreaks of dengue fever, chikungunya disease, and yellow fever in many countries. Vector ecology and disease epidemiology are strongly affected by environmental changes. Urbanization is a worldwide trend and is one of the most ecologically modifying phenomena. The purpose of this study is to determine how environmental changes due to urbanization affect the ecology of Aedes albopictus. Aquatic habitats and Aedes albopictus larval population surveys were conducted from May to November 2013 in three areas representing rural, suburban, and urban settings in Guangzhou, China. Ae. albopictus adults were collected monthly using BG-Sentinel traps. Ae. albopictus larva and adult life-table experiments were conducted with 20 replicates in each of the three study areas. The urban area had the highest and the rural area had the lowest number of aquatic habitats that tested positive for Ae. albopictus larvae. Densities in the larval stages varied among the areas, but the urban area had almost two-fold higher densities in pupae and three-fold higher in adult populations compared with the suburban and rural areas. Larvae developed faster and the adult emergence rate was higher in the urban area than in suburban and rural areas. The survival time of adult mosquitoes was also longer in the urban area than it was in suburban and rural areas. Study regions, surface area, water depth, water clearance, surface type, and canopy coverage were important factors associated with the presence of Ae. albopictus larvae. Urbanization substantially increased the density, larval development rate, and adult survival time of Ae. albopictus, which in turn potentially increased the vector capacity, and therefore, disease transmissibility. Mosquito ecology and its correlation with dengue virus transmission should be compared in different environmental settings.

According to upper echelon theory, this research seeks to explore how environmental leadership (EL) influences environmental performance through green intellectual capital (GIC), while also considering environmental climate as a moderator. We analyze the data from 317 Chinese manufacturing companies in two waves to test these hypotheses. The findings suggest that EL positively affects environmental performance via green structural capital and green relational capital (GRC). Moreover, environmental climate strengthens the impacts of EL on green human capital and GRC. This research enriches the existing literature on EL and GIC by revealing the vital roles of EL in promoting environmental performance and GIC, as well as explaining the mechanism between these relationships.