Teacher innovation is a key manifestation of teachers' professional competence and is essential for school development and student learning. Although previous research has identified a positive association between school climate and teacher innovation, the underlying mechanisms remain insufficiently explored, particularly in the Chinese context. Drawing on psychological empowerment theory, this study examines the relationship between school climate and teacher innovation and tests a sequential mediation model involving teacher autonomy and self-efficacy. Data were collected from 3,976 secondary school teachers in Shanghai, China, using the TALIS 2018 database. Structural equation modeling was employed for data analysis. The results indicate that school climate has a significant positive effect on teacher innovation. In addition, teacher autonomy and self-efficacy sequentially mediate this relationship, forming a chain pathway: School Climate → Teacher Autonomy → Self-Efficacy → Teacher Innovation. Specifically, a positive school climate enhances teacher autonomy, which subsequently strengthens teachers' self-efficacy and ultimately promotes innovative behavior. These findings clarify the mechanisms linking school climate to teacher innovation and extend psychological empowerment theory by highlighting the sequential interplay between autonomy and competence in shaping teachers' innovative behavior.
This study explores primary school teachers' interpretations of children's media engagement, the ways media content is reflected in children's values and behaviors, and the pedagogical positioning of media literacy within platform-dominated digital environments. Grounded in debates on platform imperialism, the study examines how teachers make sense of children's everyday encounters with global digital platforms and how they position media literacy in response to these conditions. In doing so, it addresses four interrelated questions concerning students' media engagement, value and behavioral reconfigurations, the role of global platforms, and teachers' pedagogical responses. A qualitative case study design was employed. Data were collected through semi-structured interviews with 71 primary school teachers working in public schools across different regions of Türkiye, including village, town, and city settings. Participants represented diverse age groups and levels of professional experience. The interview data were analyzed through thematic analysis. Both inductive and deductive procedures were used in order to identify recurrent patterns of meaning across teachers' accounts, while concepts such as platform imperialism, cultural autonomy, and pedagogical resistance provided an analytical frame for interpretation. According to teachers' accounts, children's media experiences were increasingly organized within algorithmically structured and platform-centered environments. Teachers described these environments as reshaping authority relations, value orientations, and everyday practices. They associated children's media engagement with patterns such as platform concentration, normalization of risk, fragile critical filtering, moral ambiguity, consumption-oriented value formation, and shifts in epistemic authority from institutional actors toward digital platforms. Teachers also described family mediation as a site of tension, sometimes reinforcing and at other times partially limiting platform influence. In response to these perceived conditions, teachers positioned media literacy not simply as a technical competence but as a pedagogical orientation involving critical verification, ethical awareness, resistance to manipulation, and participatory engagement. The findings suggest that children's media engagement is interpreted by teachers not merely as an issue of individual media use, but as part of a broader restructuring of childhood within platform-dominated environments. In this context, media literacy emerged in teachers' narratives as a contextually grounded pedagogical response to algorithmic authority, information fragility, and shifting cultural norms. The study contributes to discussions of childhood, digital culture, and education by offering an empirically situated account of how platformized environments are understood and negotiated in primary school contexts.
The Student Wellbeing Teacher-Report Scale is a 12-item instrument with three subscales (i.e., academic, social, and emotional well-being) for assessing students' subjective well-being in school-based screenings. This study examined the factor structure, validity, and reliability of the Student Wellbeing Teacher-Report Scale within a multilevel framework using responses from 81 teachers about 1,910 students in the United States. Results from multilevel confirmatory factor analysis supported a three-factor correlated model at both the within- and between-teacher levels, with one item from the social well-being removed. Cross-level metric invariance was established, suggesting that school-specific well-being was conceptualized similarly at the individual and teacher/classroom levels. Results from multilevel structural equation modeling showed that each subscale had unique, significant associations with internalizing and externalizing behaviors at the within-teacher level. However, only the social well-being subscale was associated with externalizing behaviors at the between-teacher level. The three subscales demonstrated satisfactory internal reliability across levels, supported by multilevel α and ω coefficients ranging from .81 to .95. Overall, findings provide strong psychometric support for the Student Wellbeing Teacher-Report Scale with a social well-being item (i.e., handles frustrations well at school) removed as a school-based screener and highlight directions for further refinement and application in practice. It is important to note that the relatively small teacher sample size may have limited the ability to detect significant results at the teacher/classroom level. (PsycInfo Database Record (c) 2026 APA, all rights reserved).
Curriculum reforms require teachers not only to understand new curricular expectations but also to adapt them to classroom practice. In this process, teachers' cognitive resources and professional behaviors may be associated with how they interpret and respond to curriculum change. Although cognitive flexibility, innovative behavior, and perceptions of curriculum change have been examined separately in previous research, limited attention has been paid to how these constructs are jointly related within a single explanatory model. This study aims to examine the mediating role of innovative behavior in the relationship between teachers' cognitive flexibility and their perceptions of curriculum change. A cross-sectional correlational survey design was employed in the study. The sample consisted of 324 teachers working in a city in Türkiye during the 2025-2026 academic year and selected through convenience sampling. The Cognitive Flexibility Inventory, the Perception of Curriculum Change Scale, and the Innovative Behavior Scale were used as data collection instruments. SPSS PROCESS Macro version 5.0 was used for data analysis. In this context, the Errors-in-Variables (EIV) approach, which accounts for measurement error and may provide less biased parameter estimates, was adopted. In addition, the bootstrap method with 5,000 resamples and 95% confidence intervals was used to evaluate the indirect effect. Teachers' cognitive flexibility was positively and significantly associated with both their perceptions of curriculum change and their innovative behavior. Teachers' innovative behavior was also positively and significantly associated with their perceptions of curriculum change. The mediation analysis revealed a significant indirect relationship between teachers' cognitive flexibility and their perceptions of curriculum change through innovative behavior. Since the indirect effect was significant while the direct relationship became non-significant after innovative behavior was included in the model, the findings indicated an indirect-only mediation pattern. This study suggests that innovative behavior helps explain the relationship between teachers' cognitive flexibility and their perceptions of curriculum change. Teachers with higher levels of cognitive flexibility may be more likely to report higher levels of innovative behavior, which in turn is associated with more positive perceptions of curriculum change.
Pit and fissure sealant is an effective caries preventive measure, yet their global utilization in dental practice remains low. Limited knowledge of updated guidelines and technical advances, along with insufficient sealant education may contribute to this issue. This study explored the knowledge and practice of students and teachers regarding sealant use at two schools. This cross-sectional survey invited participants from West China School of Stomatology (WCSS) and Academic Centre for Dentistry Amsterdam (ACTA) to complete a questionnaire either on site or electronically. The questionnaire covered awareness of national guidelines, clinical experience, decision-making and sealant application procedures. Data were analyzed using Chi-squared or Fisher's exact tests. Statistical significance was set at P < 0.05 and multiple comparisons were adjusted using the Bonferroni correction. A total of 85 students and 25 teachers from WCSS, along with 87 students and 30 teachers from ACTA, completed the questionnaire. Over 60% of the participants reported familiarity with the national sealant guideline or advisory, with no significant differences in awareness between teachers and students within the same school or between the two schools. For clinical decision-making, most participants from both schools (> 80%) considered caries risk in sealant application decisions. ACTA participants consider sealant application until later stage of caries (Sound-Enamel cavitied lesions), whereas WCSS participants favored applying sealants at earlier stages (Sound-Opaque). Differences of clinical procedures were also detected between schools. Abrasive paste was preferred by ACTA participants, while toothpaste was chosen by more WCSS teachers and students (P < 0.001). WCSS respondents used cotton rolls for saliva control, while ACTA participants preferred rubber dam isolation (P < 0.001). Fewer WCSS students considered glass ionomer cement as the sealing materials compared to their teachers (P < 0.001) and the ACTA students (P = 0.005). This cross-section survey study revealed that in both schools, teachers and students successfully integrated caries risk into sealant decision-making. Agreement on the overall principles of sealant placement was observed; however, some discrepancies were identified in specific indications and procedural details. These observed differences suggest a need for further alignment of sealant guidelines and the operational reginal standards of detailed clinical procedures.
Background/Objectives: Interparental conflict increases adolescents' risk of peer bullying victimization, yet its underlying mechanisms remain underexplored. This study sought to examine whether depressive symptoms mediate the longitudinal relationship between interparental conflict and peer bullying victimization among adolescents, and whether teacher-student relationships moderate this mediating pathway. Methods: A two-wave longitudinal design spaced six months apart was adopted, with 759 Chinese adolescents participating across both waves. Interparental conflict, depressive symptoms, teacher-student relationships, and peer bullying victimization were measured using validated scales. Gender, age, and Wave 1 depressive symptoms and peer bullying victimization were controlled. Data were analyzed using SPSS PROCESS macro (Models 4 and 58). Results: Interparental conflict was positively associated with peer bullying victimization six months later (r = 0.14, p < 0.001). Depressive symptoms significantly mediated this relationship (indirect effect = 0.03, 95% CI = [0.01, 0.05]). Teacher-student relationships significantly moderated both the path from interparental conflict to depressive symptoms (b = -0.08, p < 0.05) and the path from depressive symptoms to peer bullying victimization (b = -0.16, p < 0.01). For adolescents with low teacher-student relationships, interparental conflict was positively associated with depressive symptoms (b = 0.10, p < 0.001), and depressive symptoms were positively associated with peer bullying victimization (b = 0.45, p < 0.001). For those with high teacher-student relationships, interparental conflict was not significantly associated with depressive symptoms (b = 0.01, p > 0.05), while depressive symptoms remained significantly but weakly associated with peer bullying victimization (b = 0.26, p < 0.001). Conclusions: Positive teacher-student relationships buffer the indirect association linking interparental conflict to peer bullying victimization via depressive symptoms. This occurs mainly by attenuating the association between interparental conflict and depressive symptoms to non-significance and lowering the magnitude of the association between depressive symptoms and peer bullying victimization. Higher-quality teacher-student relationships may weaken the correlational pathway connecting family conflict to peer bullying victimization.
With youth well-being at historic lows and recent marked increases in teacher burnout this perspective paper examines the evidence that suggests by "making room" in classrooms and curriculums for artmaking activities coupled with socioemotional learning there may be improvements in teacher and student wellbeing. This paper reviews the literature on artmaking activities and human dignity education in school settings and proposes a novel framework for an inclusive school-based mental health prevention and intervention program for teachers and students. The framework recognizes that youth with subclinical mental health concerns may benefit from lower-intensity programming compared to students experiencing clinical mental health symptoms. Anticipated impacts for teachers and students include improved communication, a greater sense of connection and belonging, meaning making, self-expression, emotional regulation, and improved overall wellbeing. The paper concludes with implementation strategies for piloting this novel mental health prevention/intervention program in varied school settings with innovative collaborations among therapists, homeroom teachers, and artists. Keywords: socio-emotional learning, artmaking, dignity, teacher, student, wellness.
Learning disabilities (LDs) can adversely affect a child's academic and social development if not identified and addressed early. Teachers are pivotal in recognizing LDs and implementing classroom strategies. Their knowledge and practices significantly influence inclusive education outcomes. To assess the knowledge and practices of schoolteachers regarding learning disabilities and determine associated factors. An observational study was conducted among 100 randomly selected teachers from government and private schools in Ernakulam district, Kerala. A self-structured, validated questionnaire assessed knowledge (15 items), and practices (12 items). Teachers with prior LD training were excluded. Responses were categorized as excellent (>80%), good (50%-80%), or poor (<50%). Data was analyzed using SPSS version 20. Most participants were female (72%) and aged 31-40 years. The majority held B. Ed qualifications and taught both primary and secondary levels. About 72% had identified students with LDs, and 66% reported the presence of school counselors. Knowledge scores were good in 77%, excellent in 12%, and poor in 11%. Knowledge was significantly associated with sex and subject taught (P = 0.011 and P = 0.002). Practice levels were good in 76%, poor in 21%, and excellent in only 3%. While practice was not linked to demographic factors, it was significantly associated with knowledge levels (P = 0.013). Most teachers demonstrated good knowledge and practice, though excellence was limited. The strong association between knowledge and practice underscores the importance of training. Notably, 80% of teachers expressed interest in further LD education, indicating readiness for professional development.
The prevalence of gambling among adolescents is rising, yet teachers often lack awareness of its associated risks. Special educational needs (SEN) students may experience increased vulnerability to gambling-related harms, although it remains unclear whether teachers recognise this. To explore this issue, 15 UK mainstream secondary school teachers were recruited through purposive sampling and participated in online semi-structured interviews exploring their perceptions of gambling behaviours among typically developing and SEN students. Interviews were audio-recorded, transcribed verbatim and analysed using an exploratory interpretive approach. Reflexive thematic analysis identified three themes: understanding of gambling-related risks, SEN students' vulnerability and secondary concern. Teachers described gambling as a potential risk and perceived SEN students to be more vulnerable, citing cognitive and social deficits as contributing factors. However, gambling was generally viewed as less serious than other risky behaviours, such as substance use. Participants reported that limited training and knowledge contributed to gambling being deprioritised relative to other concerns, leaving them feeling underprepared to address gambling. Although participants perceived SEN students as particularly vulnerable, many expressed low confidence in recognising and responding to gambling-related risks. This study underscores the need for targeted professional development to challenge misconceptions and strengthen teachers' competence in addressing gambling-related risks, particularly in supporting SEN students who may experience additional vulnerabilities in online contexts. It further calls for specialised research and tailored interventions within educational settings to ensure risks related to gambling are not overlooked.
Childhood trauma is a risk factor for non-suicidal self-injury (NSSI), but school-based factors in this pathway remain unclear. This study examined whether depression indirectly linked childhood trauma to NSSI and whether perceived teacher legitimacy moderated the trauma-depression association. A cross-sectional survey was conducted in 2025 with 576 students from two senior high schools and two universities in northern and southern mainland China. The sample exceeded the minimum size estimated using the standard cross-sectional survey formula, n = Z 2 p(1 - p)/d 2. Childhood trauma, depression, perceived teacher legitimacy, and NSSI were measured using standardized self-report questionnaires. Gender, grade level, and parental educational attainment were controlled. Pearson correlations and PROCESS Model 7 with 5,000 bootstrap resamples were used. Childhood trauma was positively associated with depression and NSSI. Depression partially mediated the association between childhood trauma and NSSI. Perceived teacher legitimacy moderated the trauma-depression association; unexpectedly, this association was stronger at higher levels of perceived teacher legitimacy. Depression may be one important, but not exclusive, pathway linking childhood trauma to NSSI. Perceived teacher legitimacy appears to have context-dependent effects rather than being uniformly protective. These cross-sectional findings support the need for trauma-informed school support, early screening for depression and NSSI risk, and longitudinal research.
This study explores the key determinants shaping foreign language majors' adoption intention toward Chinese generative artificial intelligence (GenAI) tools. Using an extended Technology Acceptance Model (TAM), it examines the underexplored roles of two critical social factors: perceived teacher attitude (PTA) and peer influence (PI). A quantitative methodology was adopted, utilizing an online questionnaire to collect data from 423 foreign language majors across multiple universities in China. The proposed theoretical model and hypothesized relationships were rigorously analyzed through structural equation modeling (SEM). The results identified PI and attitude as direct positive predictors of adoption intention. Notably, restrictive PI was not significantly associated with adoption intention, whereas supportive PTA showed a significant positive association. This research extends the TAM and implied that simple prohibitions or passive advocacy are insufficient to promote the effective application of AI. Educational stakeholders should take practical measures to go with the future trends in Human-Machine Collaboration.
Residents serve as frontline educators in contemporary healthcare systems, yet formal preparation for this role remains uncommon globally. The persistent gap between teaching responsibility and structured institutional support represents a significant concern for educational quality, resident professional development, and patient safety. To synthesize the current evidence on residents' involvement in teaching, their perceptions of the educator role, factors influencing teaching effectiveness, and the outcomes of formal Residents as Teachers (RaT) programs. A structured narrative review was conducted using PubMed, Embase, and ERIC from 2010 to 2025, guided by a PCO (Population, Concept, Outcome) framework. Studies involving medical or surgical residents in postgraduate years 1-6 examining teaching roles, perceptions, or RaT program outcomes were included. Systematic reviews and studies focused on faculty or undergraduate populations were excluded. The screening and reporting process was informed by PRISMA principles; narrative synthesis with thematic organization was applied. Thirty-two studies met inclusion criteria, yielding four interconnected themes: (1) residents universally recognize teaching as central to their professional identity yet carry substantial informal teaching loads without formal preparation; (2) institutional culture, prior experience, and motivational profiles significantly shape teaching effectiveness; (3) residents across specialties and regions consistently request structured training with explicit objectives; and (4) longitudinal, integrated programs produce more durable skill gains than brief stand-alone workshops. Despite substantial informal teaching activity, residents remain underprepared for their educator role. Structured, longitudinal RaT programs are needed, particularly in under-researched healthcare systems, where unique cultural, institutional, and workforce characteristics call for locally developed approaches.
This study aims to compare and analyze the effects of the flipped classroom model and web-based instruction on pre-service classroom teachers' attitudes toward cooperative learning and their learning strategies in the Life Science Teaching course. The study employed a quasi-experimental pretest-posttest control-group design. A total of 74 third-year pre-service classroom teachers participated in the study. The implementation lasted 6 weeks and included 18 hours of instruction. Data were analyzed by comparing the experimental and control groups in terms of cooperative learning attitudes and learning strategies. The findings revealed that instruction based on the flipped classroom model was more effective than web-based instruction in improving pre-service classroom teachers' attitudes toward cooperative learning. In addition, both the flipped classroom model and web-based instruction were significantly more effective than the traditional teaching method used in the control group in developing students' learning strategies. The results suggest that technology-supported instructional approaches, particularly the flipped classroom model, can contribute positively to cooperative learning attitudes and learning strategies in teacher education. These findings highlight the importance of integrating innovative and student-centered instructional models into teacher training programs.
Medical postgraduates face a growing volume of knowledge, heavy research workload, and inefficient traditional teaching methods. Cardiovascular ultrasound teaching is particularly challenging due to its high theoretical and clinical requirements. This exploratory study evaluated the effect of a flipped classroom combined with the Feynman Learning Method in cardiovascular ultrasound teaching for postgraduates, aiming to improve teaching quality and teachers' pedagogical skills. Seventy-eight third-year postgraduates majoring in ultrasound medicine under standardized residency training at Qilu Medical College of Shandong University from September 2024 to February 2025 were divided into two groups based on training order: a control group (n=38) receiving conventional teaching plus bedside instruction, and an experimental group (n=40) receiving integrated teaching with pre-method training. Both groups used the same case library and followed the same teaching schedule for a 20-week intervention. Data from theoretical and practical exams, as well as questionnaires, were statistically analyzed using SPSS 20.0 software. There were no significant differences in gender or age between the two groups at baseline (p > 0.05). The experimental group achieved significantly higher scores in theoretical (85.3±8.17) and practical (83.4±8.58) exams, as well as higher overall exam pass rates (all p < 0.001). Eighty-five percent of students and 87.5% of teachers accepted the integrated model and confirmed its role in enhancing learning outcomes and teaching skills. While 70% of students reported considerable out-of-class time investment (only 37.5% were willing to continue using the model), 87.5% of teachers recognized its high value for promoting clinical teaching. The student-centered integrated teaching model effectively improves postgraduates' academic performance and teachers' teaching competence through the "learning-by-teaching" approach, making it a valuable pedagogical reform for cardiovascular ultrasound education. It provides a solid foundation for further validation; however, its large-scale application requires full consideration of postgraduates' practical challenges, such as heavy research pressure and time limitations.
Continuing professional development (CPD) about social processes and structures (eg, power relations, organizational policies) in health care has been minimally explored. This study examined the teaching of social practices and structures in the domains of interprofessional collaboration and quality improvement, and the education paradigms used, with particular attention to transformative education. This case study included three CPD cases at two university-affiliated Centres. Data collection included observations (n = 42.5 hours), interviews with program/Centre leads and guest presenters (n = 13), and documents. Paradigms of education informed the interpretive thematic analysis conducted. A range of concepts related to social practices and structures were introduced across sessions (eg, systemic racism, unconscious bias). An examination of education purpose, role of teacher-learner, and teaching modalities, demonstrated that varied paradigms of education (eg, cognitivist, constructivist) were drawn on to teach these topics, with efforts toward transformative approaches. Programs were largely oriented to practical learning and application, with the teacher positioned as expert or learners being encouraged to interact with, and learn from each other. However, participants recognized additional education purposes aligned with a transformative paradigm, to question and shift learners' perspectives, making efforts to align teacher-learner roles and learning modalities. There were tensions, though, working across paradigms; session formats and learner/sponsor expectations influenced paradigms used. Findings provide insights to how to incorporate content about social practices and structures into interprofessional collaboration and quality improvement CPD. Given the multiple and sometimes competing education goals, positioning paradigms of education as coexisting layers may be more effective than viewing them as mutually exclusive.
To examine theoretically informed association patterns linking different types of academic stress to depression among high school students and to investigate the roles of parent-child communication and self-efficacy within a proposed moderated serial mediation model. Using convenience sampling, a cross-sectional survey was conducted among 967 high school students in northwestern China. Measures included four types of academic stress (parental, self-imposed, teacher, and social stress), perceived stress (PSS-14), loneliness (UCLA-LS3), parent-child communication (PACS), self-efficacy (GSES), and depression (KADS-11). A moderated serial mediation model was tested with PROCESS Model 83 and 5,000 bootstrap resamples, specifying perceived stress and loneliness as sequential mediators and parent-child communication and self-efficacy as moderators. An EBICglasso psychometric network analysis was further used to provide supplementary descriptive evidence to support the findings. All four types of academic stress were indirectly associated with depression through the serial pathway involving perceived stress and loneliness, with all 95% confidence intervals excluding zero. Self-imposed stress showed the strongest indirect effect (b = 0.036), whereas teacher stress showed the weakest (b = 0.013). Social stress additionally showed a distinctive association with loneliness beyond perceived stress (β = 0.22, p < 0.001), and this indirect effect (b = 0.176) was the largest among all pathways. Parent-child communication significantly attenuated the association between self-imposed stress and perceived stress (β = -0.19, p < 0.001). Self-efficacy significantly attenuated the association between perceived stress and loneliness (β = -0.22, p < 0.001), showing a stronger moderating association than parent-child communication. In the network analysis, perceived stress showed the highest strength and closeness centrality, self-efficacy showed the highest betweenness centrality, and social stress retained a direct edge with loneliness. The findings suggest that different types of academic stress may be linked to adolescent depression through a cognitive-interpersonal pathway involving perceived stress and loneliness. Parent-child communication and self-efficacy may serve as protective factors within this association pattern. These results provide a basis for more targeted interventions for adolescent depression in school contexts.
Kidney injury and chronic kidney disease progression are accompanied by spatially heterogeneous activation of programmed cell death (PCD), yet existing approaches have limited ability to jointly infer cell-type composition, death-program activity, and their spatial organization from spatial transcriptomics (ST) data. We present CoDeST (Confidence-weighted Dual-teacher Spatial Training), a multi-task spatial inference framework that predicts, for each ST spot, a 34-class kidney cell-type composition vector together with continuous activity scores for four PCD programs: apoptosis, pyroptosis, necroptosis, and ferroptosis. CoDeST uses a three-stage pseudo-to-real training strategy that combines self-supervised pretraining on real ST slices, supervised deconvolution on donor-matched pseudo-spots, and real-ST domain adaptation with teacher-student distillation, marker-based weak constraints, confidence-weighted AUCell/ssGSEA PCD supervision, and boundary-preserving spatial regularization. In donor-held-out pseudo-spot benchmarks, CoDeST shows competitive recovery of cell-type proportions compared with representative deconvolution methods. On real kidney ST data, marker-consistency analysis and a Visium HD-derived benchmark further support its ability to transfer deconvolution signals from pseudo-spots to real spatial tissue settings. Ablation and sensitivity analyses indicate that the three-stage design, confidence weighting, and spatial graph modeling contribute to stable deconvolution and PCD mapping while balancing spatial coherence with boundary contrast. CoDeST provides a kidney-focused framework for joint spatial mapping of cell composition and PCD-related transcriptional programs.
Whole-of-school physical activity initiatives have strong potential for population-level impact, yet few have been disseminated or scaled nationally. Given the relative lack of dissemination research compared with implementation research in this field, and the limited exploration of third-party perspectives, this study aimed to examine how cross-sector partner organizations disseminated TransformUs Secondary to schools across Australia. This qualitative study was incorporated into the TransformUs Secondary type 2 hybrid implementation-effectiveness trial. Sixteen partner organizations were invited to participate in online semi-structured interviews conducted 16-months post-launch. Interview transcripts were analyzed using reflexive thematic analysis, with findings organized into three major themes: dissemination strategies (types, frequency, pathways), barriers and enablers influencing partners' dissemination efforts, and mechanisms underlying perceived effective dissemination strategies. Out of 16 invited partner organizations representing diverse sectors, 10 participated in interviews. Partners employed multiple dissemination strategies, including different types (i.e. material development and distribution, educational meetings, and network mobilization), frequency (i.e. ranging from one-off email blasts to ongoing dissemination), and pathways (i.e. education-sector direct school engagement versus health-sector intermediary-led dissemination, each characterized by distinct channels and message framing). Partners' dissemination efforts were shaped by teacher-level factors (perceived complexity, relevance, workload pressures), partner organizational factors (bureaucratic processes, role boundaries, limited capacity), and system-level factors (funding and competing priorities). Partners perceived that effective dissemination required not only access to resources, but also strategies that built implementer capacity, strengthened positive attitudes, and fostered school-level engagement. These findings are novel in highlighting how national dissemination of a school-based physical activity initiative operates across diverse partners, showing that dissemination pathways, channels, and messages are shaped by the actors involved, and extend beyond simply sharing materials. Future research should design and test dissemination strategies, as well as explore how they interact with contextual barriers to optimize dissemination outcomes. Policy should support partner organizations to navigate organizational and system challenges, while researchers strengthen partner dissemination efforts. This study was prospectively registered with the Australian New Zealand Clinical Trials Registry (ANZCTR) under the identifier ACTRN12622000600741 on 21 April 2022.
Developing generalizable robotic policies that balance inference efficiency, manipulation accuracy, and robustness remains a formidable challenge. Existing Vision-Language-Action models demand prohibitive data scales, while keyframe-based approaches struggle to reconcile the expressivity of generative models with the latency of iterative sampling. To address this trilemma, we present Flow2Act, a unified framework that integrates agglomerative perception with a deterministic one-step generative policy. Unlike prior methods relying on separate semantic encoders or iterative diffusion processes, our approach introduces three key innovations. First, we employ an agglomerative multi-teacher visual backbone that distills complementary strengths from diverse foundation models, capturing semantics, spatial structure, and segmentation to yield robust representations without task-specific pretraining. Second, we propose a conditional MeanFlow policy that parameterizes the interval-averaged velocity field. This formulation enables genuine single-step action generation, eliminating the discretization errors and computational overhead inherent in ODE-based flow matching. Third, we devise a curriculum region-aware mechanism via a Spatial-Grounded State Space Duality architecture, which progressively shifts attention from global flow stability to fine-grained contact precision. We evaluate Flow2Act on challenging simulation benchmarks and real-world robotic tasks, demonstrating significant gains in policy performance, robustness to environmental perturbations, and cross-task real-world applicability. Videos, code, and more details are available at project page.
Automatic waste classification is an important enabling technology for cleaner cities, source-level recycling, and low-cost smart-bin systems. Although modern convolutional neural networks achieve strong recognition performance, their deployment on affordable edge devices remains constrained by memory footprint, computational cost, and response latency. This paper presents an edge-oriented compact CNN framework for waste image classification, combining a high-accuracy MobileNetV4 reference model with three lightweight student architectures: EfficientNet-Lite0, LCNet-0.5, and MobileNetV3-Small-0.5. All models are evaluated on TrashNet under a unified preprocessing, training, and size-accounting protocol, allowing a clear comparison of accuracy-efficiency trade-offs. On the main stratified train/validation/test split, the MobileNetV4 teacher achieves 97.09% top-1 accuracy, while the compact students retain strong performance with substantially smaller footprints: EfficientNet-Lite0 reaches 93.99% with 3.38 M parameters, LCNet-0.5 reaches 94.18% with only 0.61 M parameters, and MobileNetV3-Small-0.5 reaches 87.73% with 0.57 M parameters. A complementary stratified five-fold evaluation, including both knowledge-distilled and non-distilled student variants, provides a robust assessment of model behavior across data partitions and confirms LCNet-0.5 as the most suitable sub-megabyte candidate under the proposed size-accuracy selection rule. The selected LCNet-0.5 model achieves a macro-F1 score of 0.9247 on the main TrashNet test split and is integrated into a self-contained Raspberry Pi 3 Model B+ prototype that performs local camera-to-display inference with an observed end-to-end latency of approximately 1.0 s per image. Cross-dataset evaluation on RealWaste further shows that the compact model can be adapted effectively to cluttered real-world imagery through short fine-tuning. Overall, the results demonstrate that careful lightweight architecture selection, supported by knowledge distillation analysis and edge-prototype validation, can deliver accurate, compact, and practically deployable waste classifiers for resource-constrained environments.