Hasil untuk "Vocational guidance. Career development"

Zhen Wang, Youcan Xu, Jun Xiao et al.

Video motion transfer aims to generate a target video that inherits motion patterns from a source video while rendering new scenes. Existing training-free approaches focus on constructing motion guidance based on the intermediate outputs of pre-trained T2V models, which results in heavy computational overhead and limited flexibility. In this paper, we present FlowMotion, a novel training-free framework that enables efficient and flexible motion transfer by directly leveraging the predicted outputs of flow-based T2V models. Our key insight is that early latent predictions inherently encode rich temporal information. Motivated by this, we propose flow guidance, which extracts motion representations based on latent predictions to align motion patterns between source and generated videos. We further introduce a velocity regularization strategy to stabilize optimization and ensure smooth motion evolution. By operating purely on model predictions, FlowMotion achieves superior time and resource efficiency as well as competitive performance compared with state-of-the-art methods.

Mehdi Noroozi, Alberto Gil Ramos, Luca Morreale et al.

Current sampling mechanisms for conditional diffusion models rely mainly on Classifier Free Guidance (CFG) to generate high-quality images. However, CFG requires several denoising passes in each time step, e.g., up to three passes in image editing tasks, resulting in excessive computational costs. This paper introduces a novel conditioning technique to ease the computational burden of the well-established guidance techniques, thereby significantly improving the inference time of diffusion models. We present Explicit Conditioning (EC) of the noise distribution on the input modalities to achieve this. Intuitively, we model the noise to guide the conditional diffusion model during the diffusion process. We present evaluations on image editing tasks and demonstrate that EC outperforms CFG in generating diverse high-quality images with significantly reduced computations.

Yongxin Guo, Wenbo Deng, Zhenglin Cheng et al.

Reinforcement Learning with Verifiable Rewards (RLVR) has markedly enhanced the reasoning abilities of large language models (LLMs). Its success, however, largely depends on strong base models with rich world knowledge, yielding only modest improvements for small-size language models (SLMs). To address this limitation, we investigate Guided GRPO, which injects ground-truth reasoning steps into roll-out trajectories to compensate for SLMs' inherent weaknesses. Through a comprehensive study of various guidance configurations, we find that naively adding guidance delivers limited gains. These insights motivate G$^2$RPO-A, an adaptive algorithm that automatically adjusts guidance strength in response to the model's evolving training dynamics. Experiments on mathematical reasoning and code-generation benchmarks confirm that G$^2$RPO-A substantially outperforms vanilla GRPO. Our code and models are available at https://github.com/T-Lab-CUHKSZ/G2RPO-A.

Yuxuan Luo, Zhengkun Rong, Lizhen Wang et al.

While recent image-based human animation methods achieve realistic body and facial motion synthesis, critical gaps remain in fine-grained holistic controllability, multi-scale adaptability, and long-term temporal coherence, which leads to their lower expressiveness and robustness. We propose a diffusion transformer (DiT) based framework, DreamActor-M1, with hybrid guidance to overcome these limitations. For motion guidance, our hybrid control signals that integrate implicit facial representations, 3D head spheres, and 3D body skeletons achieve robust control of facial expressions and body movements, while producing expressive and identity-preserving animations. For scale adaptation, to handle various body poses and image scales ranging from portraits to full-body views, we employ a progressive training strategy using data with varying resolutions and scales. For appearance guidance, we integrate motion patterns from sequential frames with complementary visual references, ensuring long-term temporal coherence for unseen regions during complex movements. Experiments demonstrate that our method outperforms the state-of-the-art works, delivering expressive results for portraits, upper-body, and full-body generation with robust long-term consistency. Project Page: https://grisoon.github.io/DreamActor-M1/.

Xiaoran Wu

Clutter in photos is a distraction preventing photographers from conveying the intended emotions or stories to the audience. Photography amateurs frequently include clutter in their photos due to unconscious negligence or the lack of experience in creating a decluttered, aesthetically appealing scene for shooting. We are thus motivated to develop a camera guidance system that provides solutions and guidance for clutter identification and removal. We estimate and visualize the contribution of objects to the overall aesthetics and content of a photo, based on which users can interactively identify clutter. Suggestions on getting rid of clutter, as well as a tool that removes cluttered objects computationally, are provided to guide users to deal with different kinds of clutter and improve their photographic work. Two technical novelties underpin interactions in our system: a clutter distinguishment algorithm with aesthetics evaluations for objects and an iterative image inpainting algorithm based on generative adversarial nets that reconstructs missing regions of removed objects for high-resolution images. User studies demonstrate that our system provides flexible interfaces and accurate algorithms that allow users to better identify distractions and take higher quality images within less time.

Naval Kishore Mehta, Arvind, Shyam Sunder Prasad et al.

Monitoring complex assembly processes is critical for maintaining productivity and ensuring compliance with assembly standards. However, variability in human actions and subjective task preferences complicate accurate task anticipation and guidance. To address these challenges, we introduce the Multi-Modal Transformer Fusion and Recurrent Units (MMTFRU) Network for egocentric activity anticipation, utilizing multimodal fusion to improve prediction accuracy. Integrated with the Operator Action Monitoring Unit (OAMU), the system provides proactive operator guidance, preventing deviations in the assembly process. OAMU employs two strategies: (1) Top-5 MMTF-RU predictions, combined with a reference graph and an action dictionary, for next-step recommendations; and (2) Top-1 MMTF-RU predictions, integrated with a reference graph, for detecting sequence deviations and predicting anomaly scores via an entropy-informed confidence mechanism. We also introduce Time-Weighted Sequence Accuracy (TWSA) to evaluate operator efficiency and ensure timely task completion. Our approach is validated on the industrial Meccano dataset and the largescale EPIC-Kitchens-55 dataset, demonstrating its effectiveness in dynamic environments.

Hu Guo, Lily Patel, Rohan Gupt

Optical see-through augmented reality (OST-AR) overlays digital targets and annotations on the physical world, offering promising guidance for hands-on tasks such as medical needle insertion or assembly. Recent work on OST-AR depth perception shows that target opacity and tool visualization significantly affect accuracy and usability; opaque targets and rendering the real instrument reduce depth errors, whereas transparent targets and absent tools impair performance. However, reliance on visual overlays may overload attention and leaves little room for depth cues when occlusion or lighting hampers perception. To address these limitations, we explore multimodal feedback that combines OST-AR with wrist-based vibrotactile haptics. The past two years have seen rapid advances in haptic technology. Researchers have investigated skin-stretch and vibrotactile cues for conveying spatial information to blind users, wearable ring actuators that support precise pinching in AR, cross-modal audio-haptic cursors that enable eyes-free object selection, and wrist-worn feedback for teleoperated surgery that improves force awareness at the cost of longer task times. Studies comparing pull versus push vibrotactile metaphors found that pull cues yield faster gesture completion and lower cognitive load. These findings motivate revisiting OST-AR guidance with a fresh perspective on wrist-based haptics. We design a custom wristband with six vibromotors delivering directional and state cues, integrate it with a handheld tool and OST-AR, and assess its impact on cue recognition and depth guidance. Through a formative study and two experiments (N=21 and N=27), we show that participants accurately identify haptic patterns under cognitive load and that multimodal feedback improves spatial precision and usability compared with visual-only or haptic-only conditions.

Shi Meng, Bin Tian, Xiaotong Zhang

Efficient truck dispatching via Reinforcement Learning (RL) in open-pit mining is often hindered by reliance on complex reward engineering and value-based methods. This paper introduces Curriculum-inspired Adaptive Direct Policy Guidance, a novel curriculum learning strategy for policy-based RL to address these issues. We adapt Proximal Policy Optimization (PPO) for mine dispatching's uneven decision intervals using time deltas in Temporal Difference and Generalized Advantage Estimation, and employ a Shortest Processing Time teacher policy for guided exploration via policy regularization and adaptive guidance. Evaluations in OpenMines demonstrate our approach yields a 10% performance gain and faster convergence over standard PPO across sparse and dense reward settings, showcasing improved robustness to reward design. This direct policy guidance method provides a general and effective curriculum learning technique for RL-based truck dispatching, enabling future work on advanced architectures.

Xinyan Niu

Since the concept of new vocational farmers was first put forward in the No. 1 central document, the cultivation of new vocational farmers has become a social hotspot. The new vocational farmers have four characteristics: professionalism, modernity, information and humanity. The cultivation of new vocational farmers has the characteristics of diversified cultivation perspectives, strong plasticity of the main team, and rich training content, But there are also some difficulties in the cultivation process, such as incomplete and incomplete design of the new vocational farmer cultivation system; Difficulties in transforming traditional farmers into new professional farmers; As an important and malleable subject of new vocational farmers, college students have a weak willingness to become new vocational farmers. In order to promote the cultivation and development of new vocational farmers, it is necessary to achieve the integration and systematization of the new vocational farmer cultivation system; Guide traditional farmers to achieve their modernization transformation; At the same time, it is also necessary for the government, society, and schools to work together to create a good social public opinion and guidance system, so that college students can form correct career concepts, and willing college students can help rural economic development and promote rural revitalization.

Christina Boateng

This study investigates the persistent underrepresentation of girls in Science, Technology, Engineering Mathematics (STEM) -related Technical and Vocational Education and Training (TVET) programmes at the pre-tertiary level in the Cape Coast Metropolis. Recognizing the critical role of STEM careers in driving socio-economic development, the study identifies the factors contributing to this gender disparity and proposes strategies to enhance female participation in STEM –related TVET programmes. The study is informed by the Social Role Theory. The convergent mixed-method research approach and the cross-sectional research design were employed. Uutilising the census sampling, 109 girls enrolled in STEM- related TVET programmes in pre-tertiary TVET delivery institutions in the Metropolis were selected for the study. Primary data as well as secondary data as collected from the Institutions and utilised. The primary data was collected using a structured questionnaire and interviews. Data analysis was done using descriptive statistics. The findings revealed that despite ongoing efforts to promote gender equality, significant barriers continue to hinder girls' enrolment in STEM-related TVET programmes. These barriers include systemic challenges, entrenched societal stereotypes, insufficient institutional support, and absence of targeted interventions. The study concludes that a multi-faceted approach, involving a comprehensive action and strong support is needed at the school level to address the issue. It is therefore recommended that school leaders should employ strategies to counteract the persistent stereotypes, encourage support, and inspire more girls to pursue STEM-related TVET. These should include instituting mentorship programmes to connect female students with successful women in STEM-related TVET fields as role models to provide guidance, share personal experiences, and encourage female students to explore and participate in the field. Again, community awareness campaigns aimed at changing the perceptions of parents, guardians as well as the broader community about STEM field and professions for girls should be launched to promote the relevance and benefits of STEM –related TVET education for girls and society as a whole.

Cosme Zinsou Odjo, Cordula Kaoti, Alaba Oluwatoyin Oyewusi

The study reviews the role of quality assurance in Technical and Vocational Education and Training (TVET) systems in supporting economic development, enhancing labor market competitiveness, and meeting industry expectations. The authors use a desk research strategy, focusing on document review and qualitative content analysis to analyze credible reports on the role of quality assurance in TVET to promote competent workforce. As per findings, quality assurance in TVET institutions is crucial to ensuring that graduates possess the necessary skills and competencies for the workforce, reducing skills gaps and enhancing employability. It aligns TVET programs with industry standards, enabling graduates to contribute effectively to the workforce. Quality assurance mechanisms also enhance the credibility and reputation of TVET institutions, increasing their attractiveness to students and employers. It helps produce highly skilled individuals, contributing to the country’s economic development and global competitiveness. However, the TVET sector in West Africa faces several challenges, including inadequate infrastructures, curriculum relevance, teaching and learning quality, limited industry collaboration, and lack of quality assurance mechanisms. To address these issues, the authors suggest some innovative dimensions of quality standards to strengthen and improve quality assurance in TVET institutions. These quality standards include leadership, governance, management, physical environment, human resources, curriculum, student participation, access, equity, career guidance, financial management, program evaluation, research, innovation and development. All of these are interrelated, thereby spurring the overall quality and effectiveness of TVET programs and institutions.

Pengzhi Li, Qiang Nie, Ying Chen et al.

Despite significant advancements in image customization with diffusion models, current methods still have several limitations: 1) unintended changes in non-target areas when regenerating the entire image; 2) guidance solely by a reference image or text descriptions; and 3) time-consuming fine-tuning, which limits their practical application. In response, we introduce a tuning-free framework for simultaneous text-image-guided image customization, enabling precise editing of specific image regions within seconds. Our approach preserves the semantic features of the reference image subject while allowing modification of detailed attributes based on text descriptions. To achieve this, we propose an innovative attention blending strategy that blends self-attention features in the UNet decoder during the denoising process. To our knowledge, this is the first tuning-free method that concurrently utilizes text and image guidance for image customization in specific regions. Our approach outperforms previous methods in both human and quantitative evaluations, providing an efficient solution for various practical applications, such as image synthesis, design, and creative photography.

Zhan Wang, Lin-Ping Yuan, Liangwei Wang et al.

Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.

Kaifeng Zhang, Zhao-Heng Yin, Weirui Ye et al.

Defining reward functions for skill learning has been a long-standing challenge in robotics. Recently, vision-language models (VLMs) have shown promise in defining reward signals for teaching robots manipulation skills. However, existing work often provides reward guidance that is too coarse, leading to insufficient learning processes. In this paper, we address this issue by implementing more fine-grained reward guidance. We decompose tasks into simpler sub-tasks, using this decomposition to offer more informative reward guidance with VLMs. We also propose a VLM-based self imitation learning process to speed up learning. Empirical evidence demonstrates that our algorithm consistently outperforms baselines such as CLIP, LIV, and RoboCLIP. Specifically, our algorithm achieves a $5.4 \times$ higher average success rates compared to the best baseline, RoboCLIP, across a series of manipulation tasks.

Henglei Lv, Jiayu Xiao, Liang Li et al.

Diffusion-based text-to-image personalization have achieved great success in generating subjects specified by users among various contexts. Even though, existing finetuning-based methods still suffer from model overfitting, which greatly harms the generative diversity, especially when given subject images are few. To this end, we propose Pick-and-Draw, a training-free semantic guidance approach to boost identity consistency and generative diversity for personalization methods. Our approach consists of two components: appearance picking guidance and layout drawing guidance. As for the former, we construct an appearance palette with visual features from the reference image, where we pick local patterns for generating the specified subject with consistent identity. As for layout drawing, we outline the subject's contour by referring to a generative template from the vanilla diffusion model, and inherit the strong image prior to synthesize diverse contexts according to different text conditions. The proposed approach can be applied to any personalized diffusion models and requires as few as a single reference image. Qualitative and quantitative experiments show that Pick-and-Draw consistently improves identity consistency and generative diversity, pushing the trade-off between subject fidelity and image-text fidelity to a new Pareto frontier.

Kun Wang, Zheng Chen, Fangmin Lu et al.

This paper addresses an optimal guidance problem concerning the vertical landing of a lunar lander with the objective of minimizing fuel consumption. The vertical landing imposes a final attitude constraint, which is treated as a final control constraint. To handle this constraint, we propose a nonnegative small regularization term to augment the original cost functional. This ensures the satisfaction of the final control constraint in accordance with Pontryagin's Minimum Principle. By leveraging the necessary conditions for optimality, we establish a parameterized system that facilitates the generation of numerous optimal trajectories, which contain the nonlinear mapping from the flight state to the optimal guidance command. Subsequently, a neural network is trained to approximate such mapping. Finally, numerical examples are presented to validate the proposed method.

Haoyu Li, Hao Wu, Badong Chen

Reconstructing visual stimuli from functional Magnetic Resonance Imaging fMRI enables fine-grained retrieval of brain activity. However, the accurate reconstruction of diverse details, including structure, background, texture, color, and more, remains challenging. The stable diffusion models inevitably result in the variability of reconstructed images, even under identical conditions. To address this challenge, we first uncover the neuroscientific perspective of diffusion methods, which primarily involve top-down creation using pre-trained knowledge from extensive image datasets, but tend to lack detail-driven bottom-up perception, leading to a loss of faithful details. In this paper, we propose NeuralDiffuser, which incorporates primary visual feature guidance to provide detailed cues in the form of gradients. This extension of the bottom-up process for diffusion models achieves both semantic coherence and detail fidelity when reconstructing visual stimuli. Furthermore, we have developed a novel guidance strategy for reconstruction tasks that ensures the consistency of repeated outputs with original images rather than with various outputs. Extensive experimental results on the Natural Senses Dataset (NSD) qualitatively and quantitatively demonstrate the advancement of NeuralDiffuser by comparing it against baseline and state-of-the-art methods horizontally, as well as conducting longitudinal ablation studies.

Roy Venketsamy, Zijing Hu, Nashua Naicker

Background: Sexual orientation discrimination impacts teachers’ well-being in school environments. The Prevention of Discrimination and Promotion of Equality Act of 2002 prohibits discrimination based on gender and sexual orientation, granting lesbian, gay, bisexual, transgender and queer (LGBTQ) individuals many of the same rights as their cis-gender, heterosexual peers. Despite this, many LGBTQ teachers still face problematic, uncomfortable, and tension-filled experiences in schools. Traditionally, the Foundation Phase (FP) of education was gendered, but this is changing. However, there is a perception that male FP teachers are primarily queer with a “gay agenda.” The authors used the Class Psychological Stressor Strain theory to explore the negative experiences of queer-identifying teachers. Objectives: This article examines the effects of Foundation Phase teachers’ sexual orientation on their work environment experiences. Methods: This qualitative study, grounded in an interpretivist paradigm, explored the lived experiences of three queer-identifying teachers in the Gauteng Province through individual face-to-face interviews. Results: The findings reveal that schools are not adequately addressing discrimination. Teachers often face marginalisation and discrimination. Conclusion: This article explored a sensitive issue faced by teachers whose sexual orientation differs from society’s norms. It highlighted the discrimination and harassment LGBTQ teachers face in the Foundation Phase in public schools in South Africa. Contribution: By unveiling these issues, the article aims to reduce the stigma, bias, and prejudice commonly faced by LGBTQ Foundation Phase teachers in the teaching profession.

A. A. Fedchenko, L. I. Borisova

The article is devoted to the analysis of Russian-Tajik cooperation in the field of education and summarizing the results of research by Russian and foreign scientists to form a position confirming the importance of integrating the activities of educational organizations, which is a prerequisite for increasing human potential as the basic component of the socio-economic development of society. A set of measures focused on the development of the Russian-Tajik (Slavic) University was proposed, a vector of Russian-Tajik cooperation in the field of education was formed.Aim. Based on the study of many years of experience in interaction in the field of education between Russia and the Republic of Tajikistan and the features of the new economic reality, identify the most adequate forms of international cooperation, justifying the vector of their transformation.Tasks. Study the origins of the formation and analyze the reality of scientific and educational interaction between Russia and Tajikistan. Summarize and evaluate the experience of cooperation between the Russian-Tajik (Slavic) University and Russian universities. Justify the most adequate forms of international cooperation in the field of education and identify the vector of their further development in the Republic of Tajikistan.Methods. The study applied: analysis and synthesis, induction and deduction, comparisons and grouping, evolutionary and historical approaches.Results. The study showed that Russian-Tajik cooperation in the field of education has deep roots based on a long-term legislative framework, which makes it possible to really assess the existing relations between Russia and the Republic of Tajikistan and identify the prospects for their development. The generalization and adaptation of the existing experience made it possible to form reasoned positions that contribute to improving the scientific and educational level and improving the professional competencies of the trained staff, which is aimed at the socio-economic development of the two friendly countries. The study made it possible to form the following elements of scientific novelty and practical significance. Theoretical novelty consists in justifying the need to integrate the activities of educational organizations and allocating the organizational structure of an educational organization by type of organization based on knowledge as preferable in an unstable socio-economic environment. Practical significance lies in the fact that a set of measures focused on the development of RTSU is proposed and a vector for the development of educational organizations is formed in the context of their international integration.Conclusions. Scientific and educational cooperation: between Russia and Tajikistan is mutually beneficial for each of the parties. Assessment of the current situation and justification of proposals for the further development of relations between these countries made it possible to identify the vector of their transformations. For Russia and Tajikistan, the connection of education with the changes taking place in each of the countries is important. Since these changes are dynamic, continuous vocational education should play a significant role. For more effective implementation of continuing vocational education, work on career guidance is necessary, which is well combined with a dual education system that provides for close cooperation with companies, as well as the development of additional vocational education that actively responds to changes in the labor market. Positive results can be achieved only in the context of the active development of modern computer technologies, providing for interactivity, visualization, the use of case stages, mini-research, design work, as well as the possibility of hybrid learning. With this approach, the quality of comprehensive interdisciplinary research and development increases. It is recommended to form an organizational structure of an educational organization according to the type of organization based on knowledge, which is preferable in an unstable socioeconomic environment. The basis for the implementation of proposals for the further development of scientific and educational relations between Russia and Tajikistan is social partnership, as an instrument for regulating social and labor relations.

Khadijeh Torki, Maryam Baratali, Zohreh Saadatmand et al.

Objective: The objective of this investigation was to conduct academic and career counseling through the lens of a brain training approach. Methods: The present study employs a qualitative research paradigm, specifically utilizing content analysis and a systematic comparative categorization framework. The scope of this research encompasses a comprehensive array of literature and resources pertinent to the domain of curriculum studies, alongside texts and empirical investigations situated within the realm of brain-based education. In light of the extensive assortment of both printed and digital materials, the time frame of 2010 to 2019 has been employed for analysis. The data underwent a tripartite analytical process characterized by open, axial, and selective coding, which were subsequently articulated in a model corresponding to the specified research inquiries. Results: The findings from the analysis indicate that the curriculum counseling framework, informed by the brain training approach, comprises: motivation derived from the brain training methodology - physiological psychology - family counseling utilizing the brain training perspective - academic guidance - career counseling grounded in the brain training approach - assessments of intelligence and aptitude within the context of self-awareness and academic advising - personal development strategies informed by the brain training model - positive psychology as conceptualized through the brain training lens - social psychology framed by the principles of brain training - personality psychology as influenced by the brain training methodology - cognitive psychology informed by the brain training paradigm - and metacognitive psychology based on the brain training approach, alongside considerations of mental health challenges. Conclusions: Overall, the evidence obtained from this study substantiates the significance of a neurocognitive educational methodology in the domains of academic and vocational counseling.