Training Resources Research Articles

Editing Language Model-Based Knowledge Graph Embeddings

Recently decades have witnessed the empirical success of framing Knowledge Graph (KG) embeddings via language models. However, language model-based KG embeddings are usually deployed as static artifacts, making them difficult to modify post-deployment without re-training after deployment. To address this issue, we propose a new task of editing language model-based KG embeddings in this paper. This task is designed to facilitate rapid, data-efficient updates to KG embeddings without compromising the performance of other aspects. We build four new datasets: E-FB15k237, A-FB15k237, E-WN18RR, and A-WN18RR, and evaluate several knowledge editing baselines demonstrating the limited ability of previous models to handle the proposed challenging task. We further propose a simple yet strong baseline dubbed KGEditor, which utilizes additional parametric layers of the hypernetwork to edit/add facts. Our comprehensive experimental results reveal that KGEditor excels in updating specific facts without impacting the overall performance, even when faced with limited training resources. Code and datasets will be available at https://github.com/AnonymousForPapers/DeltaKG.

Learning from Failure: Improving Meeting Summarization without Good Samples

Existing methods aligning language models with various human needs are reliant heavily on high-quality and task-specific data. However, industrial deployment of task-specific language models often encounter challenges in the availability of appropriate training samples. Taking meeting summarization for instance, public datasets are scarce, and private corpora are also hard to obtain due to privacy issues or resource-demanding annotation. To improve meeting summarization in the absence of positively-rated (i.e., ``good'') samples, we propose Score Tuning, a cold start tuning framework that leverages bad samples of distinguishable degrees to incrementally enhance the performance of summary generation without an initial presence of good samples. Our method utilizes asynchronous and numerical human feedback that measure the quality of generated summaries. Formulating data into triplets of (transcript, summary, score), our approach instructs a pre-trained model to learn the association between summary qualities and human-rated scores and hence to generate better summaries corresponding to higher scores. The experiment results show that our method is effective in improving meeting summarization on both English and Chinese corpora while requiring less annotated data and training resources compared to existing alignment methods. Additionally, we also preliminarily explore the transferability of our approach in machine translation tasks and demonstrate its potential for future development and usage in other domains.

GOODAT: Towards Test-Time Graph Out-of-Distribution Detection

Graph neural networks (GNNs) have found widespread application in modeling graph data across diverse domains. While GNNs excel in scenarios where the testing data shares the distribution of their training counterparts (in distribution, ID), they often exhibit incorrect predictions when confronted with samples from an unfamiliar distribution (out-of-distribution, OOD). To identify and reject OOD samples with GNNs, recent studies have explored graph OOD detection, often focusing on training a specific model or modifying the data on top of a well-trained GNN. Despite their effectiveness, these methods come with heavy training resources and costs, as they need to optimize the GNN-based models on training data. Moreover, their reliance on modifying the original GNNs and accessing training data further restricts their universality. To this end, this paper introduces a method to detect Graph Out-of-Distribution At Test-time (namely GOODAT), a data-centric, unsupervised, and plug-and-play solution that operates independently of training data and modifications of GNN architecture. With a lightweight graph masker, GOODAT can learn informative subgraphs from test samples, enabling the capture of distinct graph patterns between OOD and ID samples. To optimize the graph masker, we meticulously design three unsupervised objective functions based on the graph information bottleneck principle, motivating the masker to capture compact yet informative subgraphs for OOD detection. Comprehensive evaluations confirm that our GOODAT method outperforms state-of-the-art benchmarks across a variety of real-world datasets.

The role of employer branding and internal marketing in talent attraction and retention: an applied study in a Portuguese metallurgical industry

PurposeThis research seeks to understand whether employer branding (EB) and internal marketing (IM) are fundamental to the challenge of attracting and retaining talent and how these strategies can help companies to overcome the difficulties related to the talent shortage, from the perspective of a SME Portuguese metallurgical industry.Design/methodology/approachThe research resorts to a case study of a qualitative nature, through a semi-structured interview with the head of the human resources (HR) training and development area of the Navarra Group, and quantitative, through surveys to its employees. Based on the literature, a conceptual model was constructed, whose application allowed us to perceive the relationships between the practices of EB and IM; satisfaction, motivation and commitment; attraction and retention.FindingsThe exploratory interview concluded that organizations consider EB and IM essential for an effective talent management strategy. The quantitative results demonstrate that IM and EB practices implemented in the organization contribute to the satisfaction, motivation and involvement of employees, which results in a decrease in the intention to leave. It is also noted that these practices promote an increase in the perception of organizational attractiveness, which represents a positive impact on its ability to attract.Research limitations/implicationsFrom a theoretical perspective, the research contributes to the development of knowledge about IM, EB and talent management, providing relevant data that can help define the best strategies for attraction and retention, from the point of view of IM and EB.Originality/valueThe research presents preliminary insights that can be an auxiliary tool for HR managers and professionals in the context of industrial SMEs.

A review of the current policies and guidance regarding Apgar scoring and the detection of jaundice and cyanosis concerning Black, Asian and ethnic minority neonates

BackgroundEthnic inequalities in maternal and neonatal health in the UK are well documented. Concerns exist regarding the use of skin colour in neonatal assessments. Healthcare professionals should be trained to recognise symptoms of diverse skin tones, and comprehensive, and inclusive guidance is necessary for the safe assessment of all infants. Disparities in healthcare provision have been emphasised during the COVID-19 pandemic, and additional research is needed to determine whether such policies adequately address ethnic minority neonates.MethodsA desktop search included searches of guidance produced for the United Kingdom (UK). Further searches of the Cochrane and World Health Organization (WHO) were used to identify any international guidance applicable in the UK context.ResultsSeveral policies and one training resource used descriptors ‘pink,’ ‘pale,’ ‘pallor,’ and ‘blue’ about neonatal skin and mucous membrane colour. No policies provided specific guidance on how these colour descriptors may appear in neonates with different skin pigmentation. Only the NICE guidance and HEE e-learning resource acknowledged the challenges of assessing jaundice in infants with diverse skin tones, while another guideline noted differences in the accuracy of bilirubin measurements for the assessment of jaundice. Three policies and one training resource advised against relying on visual observation of skin colour when diagnosing neonatal conditions. The training resource included images of ethnic minority neonates, although most images included white infants.ConclusionsInadequate consideration of ethnicity in UK policy and training perpetuates disparities, leading to inaccurate assessments. A review is needed for inclusivity in neonatal care, regardless of skin pigmentation.

Training health professional students to conduct SBIRT: A step forward to interprofessional collaboration to address substance use

Substance use (SU) remains a significant public health problem. Screening, brief intervention, and referral to treatment (SBIRT) is an evidence-based method to address patient SU. Training students from multiple disciplines can positively impact continuity of care for these patients.The SBIRT training was incorporated into the physician assistant, physical therapy, pharmacy and counseling curricula via online modules and live skills practice. Knowledge about screening for harmful SU and developing confidence to discuss SU using motivational interviewing (MI) to facilitate positive behavior changes were the outcomes.Training outcomes were measured for three consecutive years, assessed via pre- and post-training assessments and ratings of SBIRT interactions with patients. Knowledge gains and improved perceived competence to address SU were found. Successful curricular integration was achieved and has been maintained.Online training modules and resources are available through the Program Evaluation and Research Unit (PERU) at the University of Pittsburgh at https://www.peru.pitt.edu/ihsi/.

Evaluation of Parents' and Teachers' Views on Digital Parenting Skills

The digital world, emerging in the age of technology, presents both positive and negative aspects for every user. However, it is widely recognized that child users represent the most vulnerable group due to their ongoing cognitive and psychological development. This underscores the significance of parents acquiring digital parenting skills. This qualitative study assesses the perspectives of parents and teachers regarding digital parenting skills. Through semi-structured interview forms prepared by the researcher, interviews were conducted with 25 parents and 25 teachers whose children attend primary or secondary schools affiliated with the Ministry of National Education in the Central district of Bartın province. The data collected were analyzed using the content analysis method. The majority of both parents and teachers defined digital parenting as safeguarding children from the potential risks of the digital world. Protective measures included examples such as children not possessing their personal digital devices, imposing time limitations, parental monitoring of online activities, and open conversations with children about potential dangers. Fewer respondents acknowledged the digital world's capacity to enhance experiences and diversify learning. It was evident that parents and teachers had limited knowledge of digital parenting skills, with a notable emphasis on potential risks. The majority recognized their need for training and educational resources to enhance their capabilities. This study underscores the need for tailored training and media support designed to address adult learning needs, enabling them to serve as role models, navigate digital technologies effectively, establish connections with children, and facilitate effective communication.

The ethical wisdom of AI developers

AbstractThis paper explores ethical wisdom in the artificial intelligence (AI) developer community. Despite robust literature about the need for virtue ethics approaches in AI development, little research has directly engaged with the developer community about their progress in this regard. We have thus conducted semi-structured interviews with a worldwide cohort of 40 developers, which focused on their awareness of ethics issues, how they navigate ethical challenges, and the barriers they encounter in developing ethical wisdom. We find developers are largely aware of the ethical territories they must navigate and the moral dilemmas they personally encounter, but they face limited and inconsistent resources for ethical guidance or training. Furthermore, there are significant barriers inhibiting the development of ethical wisdom in the AI developer community, including the industry’s fixation on innovation, the narrow scope of technical practice, limited provisions for reflection and dialogue, and incentive structures that prioritize profits and prestige. The paper concludes by emphasizing the need to address the gap in domain-specific ethical skill and provides recommendations for organizations, educators, and the AI developer community.

HỢP TÁC GIỮA NHÀ TRƯỜNG VỚI DOANH NGHIỆP NHẰM HỖ TRỢ THIẾT THỰC CHO CÔNG TÁC SINH VIÊN, GÓP PHẦN NÂNG CAO CHẤT LƯỢNG ĐÀO TẠO NGUỒN NHÂN LỰC ĐÁP ỨNG NHU CẦU XÃ HỘI Ở TRƯỜNG ĐẠI HỌC VINH

It can be asserted that human resource training should be closely aligned with societal needs, the labor market, and corporate involvement. Consequently, collaboration and partnerships between universities and businesses are essential, as they contribute to enhancing the quality of education, generating direct benefits for society, higher education institutions, companies, and students. This alliance plays a crucial role in fostering a knowledge-based society. Such a relationship not only addresses labor challenges for businesses but also enables universities to enhance the quality of human resource training and transition towards more effective and efficient organizational structures and management. This approach adds value to learners' experiences. Additionally, students have the opportunity to learn, practice, and gain hands-on experience in real-life work environments at production facilities and businesses during their studies. This exposure helps them apply classroom knowledge in practical settings and develop professional skills early on. In conclusion, fostering cooperation between universities and businesses is an effective strategy for promoting mutual development and addressing common interests

Efficient Modulation Recognition by Predictive Correction and Double GRUs via a Lightweight Framework

Automatic modulation recognition (AMR) provides excellent performance advantages over conventional algorithms and plays a key role in modern communication. However, a general challenge is that the channel errors greatly deteriorate the classification capability, and the computational complexity is extremely high. To reduce the offset error of the signal and ensure fewer parameters to save training resources, we demonstrated an efficient modulation recognition scheme combined predictive correction with double Gate Recurrent Unit (GRU), thus realizing a lightweight neural framework. Predictive correction reduces channel errors, and double GRUs are better at capturing long-term dependencies. The results show that when the signal-to-noise ratio is around 18 dB, the highest recognition accuracy can be achieved, and the computational complexity is significantly reduced. The proposed scheme exhibits a tradeoff between the accuracy and the computational complexity, providing an attractive method for modulation recognition.

The Role of the Social Services and Rehabilitation Center for Persons with Social Welfare Problems (PMKS) of the East Java Provincial Social Service in Psychotic Services

This research study is to analyze the role of social service and rehabilitation centers for Persons with Social Welfare Problems (PMKS) of the East Java Provincial Social Service in serving people with psychotics. The approach used in this research is a qualitative descriptive approach, by conducting interviews, documentation and observation. The results of the research obtained show that: First, the Facilitative Role in services is to meet the basic needs of PMKS by meeting the needs of facilities and infrastructure as well as clothing and food needs for PMKS. Second, the role of education in services at the PMKS Center is currently very helpful for PMKS where various programs and training have been provided to improve the welfare of people with social welfare problems (PMKS). Third, the Representational Role in PMKS center services has collaborated with various institutions and agencies to obtain external resources for training to develop the potential of each different person. Fourth, the technical role in the PMKS center is very helpful in assessing needs for developing individual, group and community potential.

LATIHAN DAN PENGEMBANGAN SUMBER DAYA MANUSIA

Training and development of human resources is needed to improve performance is an important need for agencies and companies, so that every organization will strive to support these efforts for its employees. This study aims to present a literature review on the role of training and human resource development in improving performance. The research method used is the Systematic Literature Review (SLR) method. The results of this study show that the important role of training and human resource development in improving employee performance is that training aims to improve current capabilities, while development is aimed at preparing for higher roles in the future. The definition of training includes increased individual productivity, while development emphasizes preparation for higher responsibilities. Regular implementation of training and development can improve employee performance and provide a competitive advantage for the company. This effort is also more efficient than hiring new employees, as it builds internal skills and increases employee confidence. Based on the research that has been done, it can be concluded that training and development aims to improve employee competence and performance

Exploring clinicians' insertion experience with a new peripheral intravenous catheter in the emergency department

BackgroundHospitals frequently introduce new medical devices. However, the process of clinicians adapting to these new vascular access devices has not been well explored. The study aims to explore clinicians’ experience with the insertion of a new guidewire peripheral intravenous catheter (PIVC) introduced in the emergency department (ED) setting. MethodsThe study was conducted at two EDs in Queensland, Australia, utilising a qualitative explorative approach. Interviews were conducted with guidewire PIVC inserters, including ED doctors and nurses, and field notes were recorded by research nurses during insertions. Data analysis was performed using inductive content analysis, from which themes emerged. ResultsThe study compiled interviews from 10 participants and field notes from 191 observation episodes. Five key themes emerged, including diverse experience, barriers related to the learning process, factors influencing insertion success, and recommendations to enhance clinicians’ acceptance. These themes suggest that the key to successful adoption by clinicians lies in designing user-friendly devices that align with familiar insertion techniques, facilitating a smooth transfer of learning. ConclusionClinician adaptation to new devices is vital for optimal patient care. Emergency nurses and doctors prefer simplicity, safety, and familiarity when it comes to new devices. Providing comprehensive device training with diverse training resources, hands-on sessions, and continuous expert support, is likely to enhance clinician acceptance and the successful adoption of new devices in ED settings.

Power imbalances and equity in the day-to-day functioning of a north plus multi-south higher education institutions partnership: a case study

BackgroundPartnerships between Higher Education Institutions (HEIs) in the global north and south have commonly been used as a vehicle to drive global health research and initiatives. Among these initiatives, include health system strengthening, research capacity building, and human resource training in developing countries. However, the partnership functioning of many global north-south partnerships still carry legacies of colonialism through unrecognized behavior patterns, attitudes, and belief systems in how they function. Even with research literature calling for a shift from equality to equity in the functioning of these partnerships, many still struggle with issues of complex and unspoken power dynamics. To understand the successes and challenges of north-south partnerships, this paper explored partnership development and functioning of a northern and multi-southern HEIs partnership focused on nutrition education and research.MethodsA qualitative research approach was used; data were collected through in-depth interviews (IDIs) with questions developed from the Bergen Model of Collective Functioning (BMCF). Thirteen IDIs were conducted with partners from all institutions including stakeholders.FindingsThe partnership was built on the foundation of experiences and lessons of a previous partnership. Partners used these experiences and lessons to devise strategies to improve partnership inputs, communication, leadership, roles and structures, and maintenance and communication tasks. However, these strategies had an impact on partnership functioning giving rise to issues of inequitable power dynamics. The northern partner had two roles: one as an equal partner and another as distributor of project funds; this caused a conflict in roles for this partner. The partners distinguished themselves according to partner resources – two partners were named implementing partners and two named supportive partners. Roles and partner resources were the greatest contributors to power imbalances and caused delays in project activities.ConclusionUsing the BMCF to examine partnership dynamics illuminated that power imbalances caused a hierarchical stance in the partnership with northern partners having overall control and power of decision-making in the partnership. This could impact the effectiveness and sustainability of project in the southern institutions going forward.

Hybrid Quantum Vision Transformers for Event Classification in High Energy Physics

Models based on vision transformer architectures are considered state-of-the-art when it comes to image classification tasks. However, they require extensive computational resources both for training and deployment. The problem is exacerbated as the amount and complexity of the data increases. Quantum-based vision transformer models could potentially alleviate this issue by reducing the training and operating time while maintaining the same predictive power. Although current quantum computers are not yet able to perform high-dimensional tasks, they do offer one of the most efficient solutions for the future. In this work, we construct several variations of a quantum hybrid vision transformer for a classification problem in high-energy physics (distinguishing photons and electrons in the electromagnetic calorimeter). We test them against classical vision transformer architectures. Our findings indicate that the hybrid models can achieve comparable performance to their classical analogs with a similar number of parameters.

The importance of social networks in neurosurgery training in low/middle income countries.

Neurosurgery is evolving with new techniques and technologies, relies heavily on high-quality education and training. Social networks like Twitter, Facebook, Instagram and LinkedIn have become integral to this training. These platforms enable sharing of surgical experiences, fostering global knowledge-sharing and collaboration among neurosurgeons. Virtual conferences and courses are accessible, enhancing learning regardless of location. While these networks offer real-time communication and collaborative opportunities, they also pose challenges like the spread of misinformation and potential distractions. According to the PICO format, the target population (P) for the purpose of this paper are medical students, neurosurgical residents and consultants on the role of social media (I) in neurosurgery among Low-Middle income countries (C) with the main outcome to understand the collaborative domain of learning. This cross-sectional survey, conducted in June-July 2023, involved 210 medical students, neurosurgery residents, fellows, and practicing neurosurgeons from low and middle-income countries. A structured questionnaire assessed social network usage for neurosurgery training, covering demographic details, usage frequency, and purposes like education, collaboration, and communication. Participants rated these platforms' effectiveness in training on a 1-5 scale. Data collection employed emails, social media groups, and direct messaging, assuring respondent anonymity. The survey aimed to understand and improve social networks' use in neurosurgery, focusing on professional development, challenges, and future potential in training. In a survey of 210 participants from low and middle-income countries, 85.5% were male, 14.5% female, with diverse roles: 42.9% neurosurgery residents, 40% practicing neurosurgeons, 14.6% medical students, and 2.4% other healthcare professionals. Experience ranged from 0 to 35 years, with Mexico, Nigeria, and Kenya being the top participating countries. Most respondents rated neurosurgery training resources in their countries as poor or very poor. 88.7% used social media professionally, predominantly WhatsApp and YouTube. Content focused on surgical videos, research papers, and webinars. Concerns included information quality and data privacy. Interactive case discussions, webinars, and lectures were preferred resources, and most see a future role for social media in neurosurgery training. Our study underscores the crucial role of social media in neurosurgery training and practice in low and middle-income countries (LMICs). Key resources include surgical videos, research papers, and webinars. While social media offers a cost-effective, global knowledge-sharing platform, challenges like limited internet access, digital literacy, and misinformation risks remain significant in these regions.

Designing Effective Pedagogical Strategies for Fostering Meta-Competence Using Smart Resources in Language Training

Designing Effective Pedagogical Strategies for Fostering Meta-Competence Using Smart Resources in Language Training

Personality characteristics of Dutch nurse anesthetists and surgical nurses when compared to the normative Dutch population, a quantitative survey study

BackgroundWorking in the perioperative context is complex and challenging. The continual evaluation in this environment underscores the need for adaptability to technological advancements, and requires substantial allocation of resources for training and education. This study aimed to explore personality characteristics of nurse anesthetists and surgical nurses that are instrumental for sustainable employability in technologically advanced environment. MethodsExploratory, cross-sectional survey study including nurse anesthetists and surgical nurses, both certified and in training, and a sample of the normative Dutch population. Personality characteristics were identified with the Big Five Inventory, which consisted of 60 items answered on a five-point Likert scale (strongly disagree to strongly agree). ResultsSpecific personality traits were found for nurse anesthetists and surgical nurses when compared to the normative Dutch population. Traits of both nurse anesthetists and surgical nurses differed significantly on all domains of the Big Five Inventory, with the largest differences found within the dimension negative emotionally. ConclusionsThis study highlights the role of specific personality traits in maintaining employability within the rapidly evolving and technologically advanced landscape of healthcare. It emphasizes the relationship between individual traits and professional excellence, being crucial educational strategies for overall improvement in healthcare.

4-H Afterschool Resource Guide: Starting 4-H Clubs in After-School Programs

This resource guide is designed to be used by Extension/4-H professionals and after-school providers to deliver well-formed, positive youth development programs in after-school sites. It includes information on initiating the development of 4-H clubs at after-school sites, volunteer training resources and exercises, how to sustain the clubs, and connections to the community and beyond. The original educational material was edited and revised for this edition by the National Association of 4-H Youth Development Professionals Afterschool Working Group, led by Vanessa Spero. Published by the 4-H Youth Development Department, UF/IFAS Extension, March 2024.

Community-Driven Civic Science: Relationship Building to Prioritize Public Needs

At a time of unprecedented emphasis and investment in civic science, policymakers must grapple with whether existing civic science practices deliver on intended public benefits. Civic science, if poorly managed, can exploit communities due to inherent power imbalances between researchers and the public. For community stakeholders to guide or lead civic science—an approach we call community-driven civic science—we must invest in the relational infrastructure and relational capacity for scientists to build authentic relationships with communities. Relational infrastructure investments should include modifications to criteria for academic advancement—like tenure, expansion of funding sources, and grant reporting structures redesigned to focus on ongoing evaluation of community stakeholders’ value gained. Relational capacity building should include financial stipends for community members, training resources, and staff professional development. These investments will cultivate a new generation of civic scientists and scientifically engaged communities to collaborate towards using science for public benefit.