Yang Wang

This chapter offers an introduction to the field of Generative AI (GenAI), providing critical foundational knowledge on neural networks, deep learning, advanced architectures, and recent innovations propelling this domain. It delineates GenAI as a branch of AI focused on creating novel, coherent content, distinguishing it from discriminative models. Tracing the origins of GenAI, the chapter elucidates the concepts of neural networks, unraveling their components like input layers, hidden layers, and output layers. Backpropagation, which facilitates training through gradient computation, is explained in detail. The chapter progresses to explore deep learning, attributed to increases in compute power and data availability. Techniques like convolutional and recurrent neural networks, which enable feature learning, are highlighted. Advanced architectures like transformers and diffusion models, based on attention mechanisms and reversed diffusion processes, respectively, are analyzed as cutting-edge innovations. The chapter concludes with promising new developments like Hinton’s Forward-Forward algorithm, Meta’s I-JEPA model, privacy-preserving federated learning, and integration of reasoning agents, painting an exciting outlook for the future. Overall, the chapter provides a layered knowledge base, spanning history, techniques, architectures, and innovations in GenAI. With its comprehensive yet accessible approach, it aims to equip readers with a holistic understanding of the foundations propelling GenAI.

This book explores the revolutionary intersection of Generative AI (GenAI) and cybersecurity. It presents a comprehensive guide that intertwines theories and practices, aiming to equip cybersecurity professionals, CISOs, AI researchers, developers, architects and college students with an understanding of GenAI’s profound impacts on cybersecurity. The scope of the book ranges from the foundations of GenAI, including underlying principles, advanced architectures, and cutting-edge research, to specific aspects of GenAI security such as data security, model security, application-level security, and the emerging fields of LLMOps and DevSecOps. It explores AI regulations around the globe, ethical considerations, the threat landscape, and privacy preservation. Further, it assesses the transformative potential of GenAI in reshaping the cybersecurity landscape, the ethical implications of using advanced models, and the innovative strategies required to secure GenAI applications. Lastly, the book presents an in-depth analysis of the security challenges and potential solutions specific to GenAI, and a forward-looking view of how it can redefine cybersecurity practices. By addressing these topics, it provides answers to questions on how to secure GenAI applications, as well as vital support with understanding and navigating the complex and ever-evolving regulatory environments, and how to build a resilient GenAI security program. The book offers actionable insights and hands-on resources for anyone engaged in the rapidly evolving world of GenAI and cybersecurity.

Targeting people living with Human Immunodeficiency Virus, this research examined the prevalence of space-focused stereotypes and their underlying mechanism on behavioral inclinations. Study 1 adopted the explicit nomination and implicit Go/No-Go association tests to explore the existence of space-focused stereotypes of people living with HIV/AIDS. The results demonstrated that space-focused stereotypes were only manifested explicitly with characteristics such as messy, dirty, and gloomy. Study 2 demonstrated a more negative evaluation and community-approaching willingness for communities that include people living with HIV/AIDS than those without HIV/AIDS. Additionally, space-focused stereotypes were found to have an indirect influence on community-approaching willingness; the influence was mediated by both emotional and cognitive factors. These results indicate the deviation of explicit and implicit space-focused stereotypes. More importantly, it revealed that space-focused stereotypes decreased community evaluation and influenced behavioral inclination. This research suggested the existence of space-focused stereotypes on another stigmatized social group. Characteristics of space might be the key to forming space-focused stereotypes.

Understanding the role of teachers’ facial expressions in students’ learning is helpful to improve online teaching. Therefore, this study explored the effects of teacher’s facial expressions on students’ learning through analyzing three groups of video lectures. Participants were 78 students enrolled in three groups: one with an enhanced-expression teacher, one with a conventional-expression teacher, and one with the teacher’s audio only. ANOVA was used to explore whether video lectures instructed by the enhanced-expression teacher were better than those instructed by the conventional-expression teacher and the audio-only teacher for facilitating students’ learning, and what is the role of the teacher’s emotions in students’ perceived social presence, arousal level, cognitive load, and learning. The results showed that the video lecture by the enhanced-expression teacher was better than those with the conventional-expression teacher and with the audio-only for facilitating students’ social presence, arousal level, and long-term learning. Interestingly, it was found that the teacher’s emotions could relieve students’ cognitive load. These results explained the inconsistency of existing studies by exploring the mechanism of teachers’ emotions in students’ learning. It also provides teachers with practical guidance for video lecture design.

Our current work seeks to provide direct empirical evidence on whether Chinese international students’ experiences studying abroad promote dialectical thinking. We collected behavioral data from 258 Chinese international students studying in multiple regions. We found that there was a main effect among the four conditions. More specifically, when primed with studying abroad or typical day, participants were more likely to show tolerance for contradiction by deeming both sides of contradictory scientific statements as convincing and rating them more favorably. Therefore, it is plausible that Chinese international students’ experiences studying abroad promote their dialectical thinking. More work is needed to further this line of research by extending these effects with other measures of dialectical thinking such as perception of interconnectedness and prediction of change, adopting differing paradigms to provide more robust findings, and probing the underlying processes as to why experiences studying abroad promote dialectical thinking.

The rapid advancement of new digital technologies, such as smart technology, artificial intelligence (AI) and automation, robotics, cloud computing, and the Internet of Things (IoT), is fundamentally changing the nature of work and increasing concerns about the future of jobs and organizations. To keep pace with rapid disruption, companies need to update and transform business models to remain competitive. Meanwhile, the growth of advanced technologies is changing the types of skills and competencies needed in the workplace and demanded a shift in mindset among individuals, teams and organizations. The recent COVID-19 pandemic has accelerated digitalization trends, while heightening the importance of employee resilience and well-being in adapting to widespread job and technological disruption. Although digital transformation is a new and urgent imperative, there is a long trajectory of rigorous research that can readily be applied to grasp these emerging trends. Recent studies and reviews of digital transformation have primarily focused on the business and strategic levels, with only modest integration of employee-related factors. Our review article seeks to fill these critical gaps by identifying and consolidating key factors important for an organization’s overarching digital transformation. We reviewed studies across multiple disciplines and integrated the findings into a multi-level framework. At the individual level, we propose five overarching factors related to effective digital transformation among employees: technology adoption; perceptions and attitudes toward technological change; skills and training; workplace resilience and adaptability, and work-related wellbeing. At the group-level, we identified three factors necessary for digital transformation: team communication and collaboration; workplace relationships and team identification, and team adaptability and resilience. Finally, at the organizational-level, we proposed three factors for digital transformation: leadership; human resources, and organizational culture/climate. Our review of the literature confirms that multi-level factors are important when planning for and embarking on digital transformation, thereby providing a framework for future research and practice.

In the information era, the instant and diversified broadcasting of the COVID-19 pandemic has played an important role in stabilizing the societal mental state and avoiding inter-group conflicts. The presentation of visual graphics was considered as an innovative information form and broadly utilized in news reports. However, its effects on the audiences' cognition and behaviors have received little empirical attention. The current study applied real-time and retrospective priming paradigms to examine the impacts of information framing (positive vs. negative) and form (plain text vs. pie chart) on individuals' risk perception (cognition), positive emotion (emotion), and willingness to help others (behavioral intention) during the outbreak and post-pandemic period in China. The results indicated the “amplification effect” of the innovative form of information in the real-time priming condition, which increased the effect of the information framing on cognition, emotion, and behavioral intention. However, in the retrospective priming condition, the amplification effect on cognition and emotion were weakened, while its effect on behavioral intention disappeared. In conclusion, the study found the “amplification effect” of innovative information forms. Further, the difference in the results in the real-time and retrospective priming paradigms suggested the constraint of the context of the “amplification effect,” and indicated the possible deviation of the retrospective paradigm in studies about disaster-related news. This study provides empirical support for how subtle changes in information presentation influence public mental and behavioral responses during a pandemic and has important implications for media psychology and social governance.

This paper proposes a novel robust fixed-time control for the robot manipulator system with uncertainties. Based on the uniform robust exact differentiator algorithm, a robust control term is constructed. Then, a robust fixed-time inverse dynamics control is proposed. For the proposed control method, the fixed-time stability of a closed-loop system with uncertainties is strictly proved. The newly proposed method exhibits the following two attractive features. First, the proposed control scheme extends the existing fixed-time IDC for the robot manipulator system to the robust control scheme. Second, the proposed method is strictly nonsingular rather than the commonly used approximate approach. Simulation result demonstrates the effectiveness of the proposed control scheme.

As the most frequently used tool for measuring empathy, the Interpersonal Reactivity Index (IRI) is often scored by researchers arbitrarily and casually. Many commonly used IRI scoring approaches and their corresponding measurement models are unverified, which may make the conclusions of subsequent variable relation studies controversial and even misleading. We make the first effort to summarize these measurement models and to evaluate rationality of the common scoring methods of the IRI by confirmatory factor analysis, focusing on model fitting, factor loading, and model-based reliability simultaneously. The results show that most of these models do not fit well, indicating that the scoring approaches of the IRI corresponding to these models may be problematic. Relatively speaking, better scoring approaches of the IRI include summing empathic concern and perspective taking as the total score of the IRI and reporting the score of perspective taking as cognitive empathy.