The rapid integration of generative artificial intelligence (AI) into creative workflows is transforming design from a human-driven activity into a synergistic process between humans and AI systems. Yet, most current tools still operate as linear “executors” of user commands, which fundamentally clashes with the non-linear, iterative, and ambiguous nature of human creativity. Addressing this gap, this article introduces a conceptual framework of five irreducible paradoxes—ambiguity vs. precision, control vs. serendipity, speed vs. reflection, individual vs. collective, and originality vs. remix—as core design tensions that shape human–AI co-creative systems. Rather than treating these tensions as problems to solve, we argue they should be understood as design drivers that can guide the creation of next-generation co-creative environments. Through a critical synthesis of existing literature, we show how current executor-based AI tools (e.g., Microsoft 365 Copilot, Midjourney) fail to support non-linear exploration, refinement, and human creative agency. This study contributes a novel theoretical lens for critically analyzing existing systems and a generative framework for designing human–AI collaboration environments that augment, rather than replace, human creative agency.

Engaging with social robots has been recognized as potentially beneficial for children facing social communication challenges, especially in developing skills such as joint attention. This systematic review examines the effectiveness of robot-mediated learning in enhancing children’s cognitive abilities, emotional resilience, and self-confidence, based on an analysis of research conducted between 2018 and 2025. While social robots show promise in delivering personalized and engaging educational experiences, challenges such as emotional dependence, accessibility, and ethical concerns persist. The review concludes that further research, particularly longitudinal studies and cross-cultural comparisons, is necessary to fully understand the long-term impacts of robot-mediated learning on children’s cognitive and social development. It explores the application of robotic technology in relation to children’s educational progress across different age groups. The review includes studies of various types, such as experimental, non-experimental, and research articles, covering diverse research fields, robots, contexts of use, target audiences, and outcomes. The findings highlight the growing potential of social robots to support children’s cognitive development by providing interactive and stimulating learning experiences. This review offers valuable insights for researchers and designers aiming to improve robotic interventions and tailor them to meet diverse developmental needs and educational settings.

The rapid integration of generative Artificial Intelligence (AI) into creative workflows is transforming design from a human-driven activity into a synergistic process between humans and AI systems. Yet, most current tools still operate as linear “executors” of user commands, which fundamentally clashes with the non-linear, iterative, and ambiguous nature of human creativity. Addressing this gap, this article introduces a conceptual framework of five irreducible paradoxes—Ambiguity vs. Precision, Control vs. Serendipity, Speed vs. Reflection, Individual vs. Collective, and Originality vs. Remix—as core design tensions that shape Human–AI co-creative systems. Rather than treating these tensions as problems to solve, we argue they should be understood as design drivers that can guide the creation of next-generation co-creative environments. Through a critical synthesis of existing literature, we show how current executor-based AI tools (e.g., Microsoft 365 Copilot, Midjourney) fail to support non-linear exploration, refinement, and human creative agency. This study contributes a theoretical lens for analyzing existing systems and a generative framework for designing Human–AI collaboration environments that augment, rather than replace, human creativity.

Engaging with social robots has been identified as possibly advantageous for children with social communication challenges, particularly in practicing social skills like shared attention. This study included systematic review on articles from 2018 describing the robotic applications and technologies used in improving children’s cognitive develop-ment. By reporting on contemporary study types, modes of cognition, and the role of children, important factors and key considering for social robot’s design also discussed. This review aims to give a broad picture of the state of "cognitive development" research especially in recent years. It investigates the use of robotic technology based on the ed-ucational attainment of children at various ages. The review includes articles including experimental, non-experimental, article type, research field, robots, context of use, target audience and results or conclusions. The results indicate the increasing potential of social robots to support kids' cognitive development, especially by providing them with in-teresting and engaging learning opportunities. This review offers valuable insight for researchers, and designers looking to further enhance robotic therapies and modify them to fit a variety of developmental requirements and educational environments.

Children with autism spectrum disorder (ASD) often experience difficulties in maintaining engagement and completing structured tasks, particularly when instructional complexity and sensory load are not appropriately adapted. Digital interventions, including serious games, offer potential support; however, limited research has examined how specific design features influence usability and task completion in real-world contexts. This pilot study evaluates a sensory-adapted serious game designed to support engagement, comprehension, and task completion in children with ASD aged 5 to 12 years. Thirty-three participants from five educational settings interacted with six game levels that varied in instructional clarity and sensory complexity. Performance was assessed using structured observational measures alongside qualitative feedback from educators. Findings revealed a non-linear pattern of performance, with a marked decline at one level characterized by increased sensory and instructional demands. Higher performance was consistently associated with greater independence in task completion. Qualitative analysis highlighted the importance of clear instructions, reduced visual complexity, and predictable feedback in supporting engagement. These findings provide preliminary evidence that small increases in task complexity may disproportionately affect performance in children with ASD. The study offers practical recommendations for clinicians and educators using digital tools to support engagement and autonomy in structured learning and therapeutic activities.

Federated learning (FL) has emerged as a promising paradigm for privacy-preserving medical image analysis, enabling collaborative model training across distributed institutions without sharing sensitive patient data. However, two key challenges remain: instability of model optimization under non-independent and identically distributed(non-IID)data, and the need for client-specific adaptation. In this paper, we propose a hybrid federated learning framework that integrates proximal regularization (FedProx) with personalized model adaptation(FedPer) to address these challenges jointly. Unlike prior work that treats these mechanisms independently, we investigate their interaction and demonstrate how their combination improves both convergence stability and generalization. We evaluate the proposed framework on the MURA dataset under simulated heterogeneous client distributions. Experimental results show that the hybrid approach consistently outperforms standard federated baselines, achieving up to 96. These findings highlight the importance of jointly considering optimization stability and personalization in federated learning systems, particularly for real-world medical imaging applications.