Tao Yu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Tao Yu has a total of 22 co-authors. Here are the top co-authors by number of shared publications:
- Younghwan Pan
Kookmin University
6 shared publications - Junping Xu
College of Media and International Culture, Zhejiang University, Hangzhou, China
5 shared publications - Yihuan Tian
Culture Design Lab, Graduate School of Techno Design, Kookmin University, Seoul 02707, Republic of Korea
5 shared publications - Yang Huang
Department of Smart Experience Design, Graduate School of Techno Design, Kookmin University, Seoul 02707, Republic of Korea
4 shared publications - Zuling Cheng
Department of Global Convergence, Kangwon National University, Chuncheon-si 24341, Republic of Korea
3 shared publications - Wansok Jang
Division of Global Studies, Sungkonghoe University, Seoul 08359, Republic of Korea
2 shared publications
Latest Publications
- Open Access
- 20 December 2025
- Published by MDPI AG in Applied Sciences
- Vol. 16 (1) , 63
Smart Learning with Generative AI Tools in Higher Education: An Integrated SOR–SDT Model of Student Creative Confidence and Engagement
We investigate how generative AI tools function in smart learning by estimating a structural path model that combines the Stimulus–Organism–Response (SOR) framework with Self-Determination Theory (SDT). Using survey data from N = 540 university students and covariance-based SEM, we examine whether perceptions of these tools—usefulness (PU), ease of use (PEU), creative benefit (PCB), and personalization (PP)—align with SDT’s motivational states of perceived autonomy (PA) and perceived competence (PC) and, in turn, relate to creative confidence (CC) and creative engagement (CE). All four perceptions show positive links to PA and PC, with PP exhibiting the largest association with PA. PA precedes PC, indicating a sequential motivational route. At the behavioral level, PC relates more strongly to CC, whereas PA shows a comparatively larger association with CE. In aggregate, the results support integrating SOR with SDT to explain students’ psychological responses to generative AI tools and inform course designs that cultivate autonomy and competence to sustain creative confidence and engagement in smart-learning contexts.
- 20 December 2025
- Published by MDPI AG in Applied Sciences
- Vol. 16 (1) , 63
Open AccessSmart Learning with Generative AI Tools in Higher Education: An Integrated SOR–SDT Model of Student Creative Confidence and Engagement
We investigate how generative AI tools function in smart learning by estimating a structural path model that combines the Stimulus–Organism–Response (SOR) framework with Self-Determination Theory (SDT). Using survey data from N = 540 university students and covariance-based SEM, we examine whether perceptions of these tools—usefulness (PU), ease of use (PEU), creative benefit (PCB), and personalization (PP)—align with SDT’s motivational states of perceived autonomy (PA) and perceived competence (PC) and, in turn, relate to creative confidence (CC) and creative engagement (CE). All four perceptions show positive links to PA and PC, with PP exhibiting the largest association with PA. PA precedes PC, indicating a sequential motivational route. At the behavioral level, PC relates more strongly to CC, whereas PA shows a comparatively larger association with CE. In aggregate, the results support integrating SOR with SDT to explain students’ psychological responses to generative AI tools and inform course designs that cultivate autonomy and competence to sustain creative confidence and engagement in smart-learning contexts.
- Open Access
- 30 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 190054-190074
Which Visual Features Influence Perceived Authenticity in AI-Generated Portrait Photography? A Mixed-Methods Study
As AI-generated portrait images become increasingly photorealistic, understanding how viewers assess their authenticity is critical for applications in digital media, content verification, and generative system design. This study investigates which visual cues most significantly affect perceived authenticity in AI-generated portrait photography (AIGPP). Employing a mixed-methods approach, we first identified eight candidate visual features—skin texture (ST), spatial consistency (SC), facial proportion (FP), pose naturalness (PN), expression naturalness (EN), limb realism (LR), styling (STY), and composition (CP)—through a systematic literature review (SLR) and expert interviews. A subsequent survey involving 508 participants was conducted using AI-generated portraits as stimuli. Statistical analyses, including correlation and multiple regression, revealed that facial-related features, particularly EN, FP, and ST, were the strongest predictors of authenticity perception. These findings provide empirical insights into how micro-level visual cues influence realism judgment and may inform the optimization of AI image generation systems and visual evaluation protocols.
- 30 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 190054-190074
Open AccessWhich Visual Features Influence Perceived Authenticity in AI-Generated Portrait Photography? A Mixed-Methods Study
As AI-generated portrait images become increasingly photorealistic, understanding how viewers assess their authenticity is critical for applications in digital media, content verification, and generative system design. This study investigates which visual cues most significantly affect perceived authenticity in AI-generated portrait photography (AIGPP). Employing a mixed-methods approach, we first identified eight candidate visual features—skin texture (ST), spatial consistency (SC), facial proportion (FP), pose naturalness (PN), expression naturalness (EN), limb realism (LR), styling (STY), and composition (CP)—through a systematic literature review (SLR) and expert interviews. A subsequent survey involving 508 participants was conducted using AI-generated portraits as stimuli. Statistical analyses, including correlation and multiple regression, revealed that facial-related features, particularly EN, FP, and ST, were the strongest predictors of authenticity perception. These findings provide empirical insights into how micro-level visual cues influence realism judgment and may inform the optimization of AI image generation systems and visual evaluation protocols.
- Open Access
- 1 October 2025
- Published by MDPI AG in Journal of Theoretical and Applied Electronic Commerce Research
- Vol. 20 (4) , 257
Modeling Consumer Reactions to AI-Generated Content on E-Commerce Platforms: A Trust–Risk Dual Pathway Framework with Ethical and Platform Responsibility Moderators
With the widespread integration of Artificial Intelligence-Generated Content (AIGC) into e-commerce platforms, understanding how users perceive, evaluate, and respond to such content has become a critical issue for both academia and industry. This study examines the influence mechanism of AIGC Content Quality (AIGCQ) on users’ Purchase Intention (PI) by constructing a cognitive model centered on Trust (TR) and Perceived Risk (PR). Additionally, it introduces two moderating variables—Ethical Concern (EC) and Perceived Platform Responsibility (PLR)—to explore higher-order psychological influences. The research variables were identified through a systematic literature review and expert interviews, followed by structural equation modeling based on data collected from 507 e-commerce users. The results indicate that AIGCQ significantly reduces users’ PR and enhances TR, while PR negatively and TR positively influence PI, validating the fundamental dual-pathway structure. However, the moderating effects reveal unexpected complexities: PLR simultaneously amplifies both the negative effect of PR and the positive effect of TR on PI, presenting a “dual amplification” pattern; meanwhile, EC weakens the strength of both pathways, manifesting a “dual attenuation” effect. These findings highlight the nonlinear cognitive mechanisms underlying users’ acceptance of AIGC, suggesting that PLR and EC influence decision-making in more intricate ways than previously anticipated. By uncovering the unanticipated patterns in moderation, this study extends the boundary conditions of the trust–risk theoretical framework within AIGC contexts. In practical terms, it reveals that PLR acts as a “double-edged sword,” providing more nuanced guidance for platform governance of AI-generated content, including responsibility frameworks and ethical labeling strategies.
- 1 October 2025
- Published by MDPI AG in Journal of Theoretical and Applied Electronic Commerce Research
- Vol. 20 (4) , 257
Open AccessModeling Consumer Reactions to AI-Generated Content on E-Commerce Platforms: A Trust–Risk Dual Pathway Framework with Ethical and Platform Responsibility Moderators
With the widespread integration of Artificial Intelligence-Generated Content (AIGC) into e-commerce platforms, understanding how users perceive, evaluate, and respond to such content has become a critical issue for both academia and industry. This study examines the influence mechanism of AIGC Content Quality (AIGCQ) on users’ Purchase Intention (PI) by constructing a cognitive model centered on Trust (TR) and Perceived Risk (PR). Additionally, it introduces two moderating variables—Ethical Concern (EC) and Perceived Platform Responsibility (PLR)—to explore higher-order psychological influences. The research variables were identified through a systematic literature review and expert interviews, followed by structural equation modeling based on data collected from 507 e-commerce users. The results indicate that AIGCQ significantly reduces users’ PR and enhances TR, while PR negatively and TR positively influence PI, validating the fundamental dual-pathway structure. However, the moderating effects reveal unexpected complexities: PLR simultaneously amplifies both the negative effect of PR and the positive effect of TR on PI, presenting a “dual amplification” pattern; meanwhile, EC weakens the strength of both pathways, manifesting a “dual attenuation” effect. These findings highlight the nonlinear cognitive mechanisms underlying users’ acceptance of AIGC, suggesting that PLR and EC influence decision-making in more intricate ways than previously anticipated. By uncovering the unanticipated patterns in moderation, this study extends the boundary conditions of the trust–risk theoretical framework within AIGC contexts. In practical terms, it reveals that PLR acts as a “double-edged sword,” providing more nuanced guidance for platform governance of AI-generated content, including responsibility frameworks and ethical labeling strategies.
- 31 August 2025
- Published by Springer Nature
- p. 262-268
Deep Integration: The Revolutionary Applications and Challenges of Artificial Intelligence in Video Production
- 31 August 2025
- Published by Springer Nature
- p. 262-268
Deep Integration: The Revolutionary Applications and Challenges of Artificial Intelligence in Video Production
- Open Access
Design of Realistic and Artistically Expressive 3D Facial Models for Film AIGC: A Cross-Modal Framework Integrating Audience Perception Evaluation
The rise of virtual production has created an urgent need for both efficient and high-fidelity 3D face generation schemes for cinema and immersive media, but existing methods are often limited by lighting–geometry coupling, multi-view dependency, and insufficient artistic quality. To address this, this study proposes a cross-modal 3D face generation framework based on single-view semantic masks. It utilizes Swin Transformer for multi-level feature extraction and combines with NeRF for illumination decoupled rendering. We utilize physical rendering equations to explicitly separate surface reflectance from ambient lighting to achieve robust adaptation to complex lighting variations. In addition, to address geometric errors across illumination scenes, we construct geometric a priori constraint networks by mapping 2D facial features to 3D parameter space as regular terms with the help of semantic masks. On the CelebAMask-HQ dataset, this method achieves a leading score of SSIM = 0.892 (37.6% improvement from baseline) with FID = 40.6. The generated faces excel in symmetry and detail fidelity with realism and aesthetic scores of 8/10 and 7/10, respectively, in a perceptual evaluation with 1000 viewers. By combining physical-level illumination decoupling with semantic geometry a priori, this paper establishes a quantifiable feedback mechanism between objective metrics and human aesthetic evaluation, providing a new paradigm for aesthetic quality assessment of AI-generated content.
Open AccessDesign of Realistic and Artistically Expressive 3D Facial Models for Film AIGC: A Cross-Modal Framework Integrating Audience Perception Evaluation
The rise of virtual production has created an urgent need for both efficient and high-fidelity 3D face generation schemes for cinema and immersive media, but existing methods are often limited by lighting–geometry coupling, multi-view dependency, and insufficient artistic quality. To address this, this study proposes a cross-modal 3D face generation framework based on single-view semantic masks. It utilizes Swin Transformer for multi-level feature extraction and combines with NeRF for illumination decoupled rendering. We utilize physical rendering equations to explicitly separate surface reflectance from ambient lighting to achieve robust adaptation to complex lighting variations. In addition, to address geometric errors across illumination scenes, we construct geometric a priori constraint networks by mapping 2D facial features to 3D parameter space as regular terms with the help of semantic masks. On the CelebAMask-HQ dataset, this method achieves a leading score of SSIM = 0.892 (37.6% improvement from baseline) with FID = 40.6. The generated faces excel in symmetry and detail fidelity with realism and aesthetic scores of 8/10 and 7/10, respectively, in a perceptual evaluation with 1000 viewers. By combining physical-level illumination decoupling with semantic geometry a priori, this paper establishes a quantifiable feedback mechanism between objective metrics and human aesthetic evaluation, providing a new paradigm for aesthetic quality assessment of AI-generated content.
- Top Cited PapersOpen Access
How Do Ethical Factors Affect User Trust and Adoption Intentions of AI-Generated Content Tools? Evidence from a Risk-Trust Perspective
With the widespread application of AI-generated content (AIGC) tools in creative domains, users have become increasingly concerned about the ethical issues they raise, which may influence their adoption decisions. To explore how ethical perceptions affect user behavior, this study constructs an ethical perception model based on the trust–risk theoretical framework, focusing on its impact on users’ adoption intention (ADI). Through a systematic literature review and expert interviews, eight core ethical dimensions were identified: Misinformation (MIS), Accountability (ACC), Algorithmic Bias (ALB), Creativity Ethics (CRE), Privacy (PRI), Job Displacement (JOD), Ethical Transparency (ETR), and Control over AI (CON). Based on 582 valid responses, structural equation modeling (SEM) was conducted to empirically test the proposed paths. The results show that six factors significantly and positively influence perceived risk (PR): JOD (β = 0.216), MIS (β = 0.161), ETR (β = 0.150), ACC (β = 0.137), CON (β = 0.136), and PRI (β = 0.131), while the effects of ALB and CRE were not significant. Regarding trust in AI (TR), six factors significantly negatively influence it: CRE (β = −0.195), PRI (β = −0.145), ETR (β = −0.148), CON (β = −0.133), ALB (β = −0.113), and ACC (β = −0.098), while MIS and JOD were not significant. In addition, PR has a significant negative effect on TR (β = −0.234), which further impacts ADI. Specifically, PR has a significant negative effect on ADI (β = −0.259), while TR has a significant positive effect (β = 0.187). This study not only expands the applicability of the trust–risk framework in the context of AIGC but also proposes an ethical perception model for user adoption research, offering empirical evidence and practical guidance for platform design, governance mechanisms, and trust-building strategies.
Top Cited PapersOpen AccessHow Do Ethical Factors Affect User Trust and Adoption Intentions of AI-Generated Content Tools? Evidence from a Risk-Trust Perspective
With the widespread application of AI-generated content (AIGC) tools in creative domains, users have become increasingly concerned about the ethical issues they raise, which may influence their adoption decisions. To explore how ethical perceptions affect user behavior, this study constructs an ethical perception model based on the trust–risk theoretical framework, focusing on its impact on users’ adoption intention (ADI). Through a systematic literature review and expert interviews, eight core ethical dimensions were identified: Misinformation (MIS), Accountability (ACC), Algorithmic Bias (ALB), Creativity Ethics (CRE), Privacy (PRI), Job Displacement (JOD), Ethical Transparency (ETR), and Control over AI (CON). Based on 582 valid responses, structural equation modeling (SEM) was conducted to empirically test the proposed paths. The results show that six factors significantly and positively influence perceived risk (PR): JOD (β = 0.216), MIS (β = 0.161), ETR (β = 0.150), ACC (β = 0.137), CON (β = 0.136), and PRI (β = 0.131), while the effects of ALB and CRE were not significant. Regarding trust in AI (TR), six factors significantly negatively influence it: CRE (β = −0.195), PRI (β = −0.145), ETR (β = −0.148), CON (β = −0.133), ALB (β = −0.113), and ACC (β = −0.098), while MIS and JOD were not significant. In addition, PR has a significant negative effect on TR (β = −0.234), which further impacts ADI. Specifically, PR has a significant negative effect on ADI (β = −0.259), while TR has a significant positive effect (β = 0.187). This study not only expands the applicability of the trust–risk framework in the context of AIGC but also proposes an ethical perception model for user adoption research, offering empirical evidence and practical guidance for platform design, governance mechanisms, and trust-building strategies.
- Open Access
Advancing Traditional Dunhuang Regional Pattern Design with Diffusion Adapter Networks and Cross-Entropy
To promote the inheritance of traditional culture, a variety of emerging methods rooted in machine learning and deep learning have been introduced. Dunhuang patterns, an important part of traditional Chinese culture, are difficult to collect in large numbers due to their limited availability. However, existing text-to-image methods are computationally intensive and struggle to capture fine details and complex semantic relationships in text and images. To address these challenges, this paper proposes the Diffusion Adapter Network (DANet). It employs a lightweight adapter module to extract visual structural information, enabling the diffusion model to generate Dunhuang patterns with high accuracy, while eliminating the need for expensive fine-tuning of the original model. The attention adapter incorporates a multihead attention module (MHAM) to enhance image modality cues, allowing the model to focus more effectively on key information. A multiscale attention module (MSAM) is employed to capture features at different scales, thereby providing more precise generative guidance. In addition, an adaptive control mechanism (ACM) dynamically adjusts the guidance coefficients across feature layers to further enhance generation quality. In addition, incorporating a cross-entropy loss function enhances the model’s capability in semantic understanding and the classification of Dunhuang patterns. The DANet achieves state-of-the-art (SOTA) performance on the proposed Diversified Dunhuang Patterns Dataset (DDHP). Specifically, it attains a perceptual similarity score (LPIPS) of 0.498, a graph matching score (CLIP score) of 0.533, and a feature similarity score (CLIP-I) of 0.772.
Open AccessAdvancing Traditional Dunhuang Regional Pattern Design with Diffusion Adapter Networks and Cross-Entropy
To promote the inheritance of traditional culture, a variety of emerging methods rooted in machine learning and deep learning have been introduced. Dunhuang patterns, an important part of traditional Chinese culture, are difficult to collect in large numbers due to their limited availability. However, existing text-to-image methods are computationally intensive and struggle to capture fine details and complex semantic relationships in text and images. To address these challenges, this paper proposes the Diffusion Adapter Network (DANet). It employs a lightweight adapter module to extract visual structural information, enabling the diffusion model to generate Dunhuang patterns with high accuracy, while eliminating the need for expensive fine-tuning of the original model. The attention adapter incorporates a multihead attention module (MHAM) to enhance image modality cues, allowing the model to focus more effectively on key information. A multiscale attention module (MSAM) is employed to capture features at different scales, thereby providing more precise generative guidance. In addition, an adaptive control mechanism (ACM) dynamically adjusts the guidance coefficients across feature layers to further enhance generation quality. In addition, incorporating a cross-entropy loss function enhances the model’s capability in semantic understanding and the classification of Dunhuang patterns. The DANet achieves state-of-the-art (SOTA) performance on the proposed Diversified Dunhuang Patterns Dataset (DDHP). Specifically, it attains a perceptual similarity score (LPIPS) of 0.498, a graph matching score (CLIP score) of 0.533, and a feature similarity score (CLIP-I) of 0.772.
- Open Access
- 7 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 87600-87620
Revolutionizing Imaging Design Content With AIGC: User-Centered Challenges, Opportunities, and Workflow Evolution
This study examines the influence of Artificial Intelligence Generated Content (AIGC) technology on the workflows and career trajectories of designers. Using the Technology-Organization-Environment (TOE) framework, a mixed-methods approach was employed, integrating expert interviews and quantitative surveys. Semi-structured interviews identified ten primary influencing factors, which were further analyzed through a survey of 531 users to quantify the interrelationships among these factors. Results indicate that career paths and industry environment exert the most significant positive impact on designers’ behavioral intentions, highlighting the rising demand for AIGC and its potential to enhance career prospects. Output availability and cross-functional collaboration demonstrate practical benefits in improving content quality and team efficiency. Technological maturity and public acceptance serve as key adoption drivers, while social construction and innovation underscore AI’s capacity to foster creativity and design diversity. Ethics and privacy reflect user concerns over data security, whereas learning costs negatively affect adoption due to perceived complexity. This study confirms the utility of the TOE framework in guiding the deployment of AIGC technology and underscores the necessity of promoting AI’s sustainable development in design. It provides theoretical, practical, and policy recommendations, including improving the usability and innovation of AIGC tools, strengthening support for team collaboration, and offering career development training for designers. Furthermore, policymakers are urged to address privacy and ethical challenges to ensure the long-term sustainable adoption of AI. This research delivers critical theoretical and practical insights into integration of AIGC in the design field.
- 7 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 87600-87620
Open AccessRevolutionizing Imaging Design Content With AIGC: User-Centered Challenges, Opportunities, and Workflow Evolution
This study examines the influence of Artificial Intelligence Generated Content (AIGC) technology on the workflows and career trajectories of designers. Using the Technology-Organization-Environment (TOE) framework, a mixed-methods approach was employed, integrating expert interviews and quantitative surveys. Semi-structured interviews identified ten primary influencing factors, which were further analyzed through a survey of 531 users to quantify the interrelationships among these factors. Results indicate that career paths and industry environment exert the most significant positive impact on designers’ behavioral intentions, highlighting the rising demand for AIGC and its potential to enhance career prospects. Output availability and cross-functional collaboration demonstrate practical benefits in improving content quality and team efficiency. Technological maturity and public acceptance serve as key adoption drivers, while social construction and innovation underscore AI’s capacity to foster creativity and design diversity. Ethics and privacy reflect user concerns over data security, whereas learning costs negatively affect adoption due to perceived complexity. This study confirms the utility of the TOE framework in guiding the deployment of AIGC technology and underscores the necessity of promoting AI’s sustainable development in design. It provides theoretical, practical, and policy recommendations, including improving the usability and innovation of AIGC tools, strengthening support for team collaboration, and offering career development training for designers. Furthermore, policymakers are urged to address privacy and ethical challenges to ensure the long-term sustainable adoption of AI. This research delivers critical theoretical and practical insights into integration of AIGC in the design field.