Globally Research Articles

Fast-iTPN: Integrally Pre-Trained Transformer Pyramid Network With Token Migration.

We propose integrally pre-trained transformer pyramid network (iTPN), towards jointly optimizing the network backbone and the neck, so that transfer gap between representation models and downstream tasks is minimal. iTPN is born with two elaborated designs: 1) The first pre-trained feature pyramid upon vision transformer (ViT). 2) Multi-stage supervision to the feature pyramid using masked feature modeling (MFM). iTPN is updated to Fast-iTPN, reducing computational memory overhead and accelerating inference through two flexible designs. 1) Token migration: dropping redundant tokens of the backbone while replenishing them in the feature pyramid without attention operations. 2) Token gathering: reducing computation cost caused by global attention by introducing few gathering tokens. The base/large-level Fast-iTPN achieve 88.75%/89.5% top-1 accuracy on ImageNet-1 K. With 1× training schedule using DINO, the base/large-level Fast-iTPN achieves 58.4%/58.8% box AP on COCO object detection, and a 57.5%/58.7% mIoU on ADE20 K semantic segmentation using MaskDINO. Fast-iTPN can accelerate the inference procedure by up to 70%, with negligible performance loss, demonstrating the potential to be a powerful backbone for downstream vision tasks.

Graph Multi-Convolution and Attention Pooling for Graph Classification.

Many studies have achieved excellent performance in analyzing graph-structured data. However, learning graph-level representations for graph classification is still a challenging task. Existing graph classification methods usually pay less attention to the fusion of node features and ignore the effects of different-hop neighborhoods on nodes in the graph convolution process. Moreover, they discard some nodes directly during the graph pooling process, resulting in the loss of graph information. To tackle these issues, we propose a new Graph Multi-Convolution and Attention Pooling based graph classification method (GMCAP). Specifically, the designed Graph Multi-Convolution (GMConv) layer explicitly fuses node features learned from different perspectives. The proposed weight-based aggregation module combines the outputs of all GMConv layers, for adaptively exploiting the information over different-hop neighborhoods to generate informative node representations. Furthermore, the designed Local information and Global Attention based Pooling (LGAPool) utilizes the local information of a graph to select several important nodes and aggregates the information of unselected nodes to the selected ones by a global attention mechanism when reconstructing a pooled graph, thus effectively reducing the loss of graph information. Extensive experiments show that GMCAP outperforms the state-of-the-art methods on graph classification tasks, demonstrating that GMCAP can learn graph-level representations effectively.

Mask2Anomaly: Mask Transformer for Universal Open-Set Segmentation.

Segmenting unknown or anomalous object instances is a critical task in autonomous driving applications, and it is approached traditionally as a per-pixel classification problem. However, reasoning individually about each pixel without considering their contextual semantics results in high uncertainty around the objects' boundaries and numerous false positives. We propose a paradigm change by shifting from a per-pixel classification to a mask classification. Our mask-based method, Mask2Anomaly, demonstrates the feasibility of integrating a mask-classification architecture to jointly address anomaly segmentation, open-set semantic segmentation, and open-set panoptic segmentation. Mask2Anomaly includes several technical novelties that are designed to improve the detection of anomalies/unknown objects: i) a global masked attention module to focus individually on the foreground and background regions; ii) a mask contrastive learning that maximizes the margin between an anomaly and known classes; iii) a mask refinement solution to reduce false positives; and iv) a novel approach to mine unknown instances based on the mask- architecture properties. By comprehensive qualitative and qualitative evaluation, we show Mask2Anomaly achieves new state-of-the-art results across the benchmarks of anomaly segmentation, open-set semantic segmentation, and open-set panoptic segmentation.

The Risk of Crimean Congo Haemorrhagic Fever in India as a Growing Health Concern.

Every year witnesses an outbreak of some or the other zoonotic disease that causes the unparalled loss of human life. The year 2022 presented the outbreak of Crimean Congo haemorrhagic fever (CCHF), which brought unprecedented challenges to individuals as well as to the healthcare system all around the world, making it a serious health concern. Rising health concerns have highlighted the importance of managing and decreasing the further transmission of the CCHF virus. CCHF is one of tick-borne viral diseases, which spreads due to various reasons like changes in global warming, environmental influences, and other ecological factors. All these factors somehow impact the disease prevalence. This disease has a negative impact on both humans and livestock. The diverse climate and significant livestock population of India make it susceptible to the prevalence of CCHF. Therefore, it is the need of the hour to develop some strategies in order to tackle the challenges posed by CCHF. This article includes all the cases of CCHF that have occurred in India from the year 2011, along with the fatality rates associated with this disease. Also this study discusses the need to explore some specific drugs for the management and prevention of such diseases. In addition, the pathogenesis of the disease progression, along with some protective measures suggested by the government has been described for prevention of CCHF. Subsequently, this article attempted to draw attention towards the risk that may be posed by CCHF in the coming scenario, emphasizing the importance of taking proactive measures in anticipation of such risks.

Joint optimization of recyclable inventory routing problem under uncertainties in an incentive-based recycling system

Due to the value of resource recovery and the development of a circular economy, waste recycling has gathered global attention. Recently, many emerging cities designed new systems like an incentive-based recycling system (IBRS). In such systems, recyclables are collected through community recycling nodes by offering incentives, then transported to street recycling stations and sorted before being finally recycled. The increased recycling nodes and the incentives enhance the convenience and residents’ enthusiasm for waste recycling, but also intensify the uncertainty of recycling quantities and the complexity of the recycling operation management. Poor recycling operation management may result in increased recycling costs or greater loss of recyclables, which discourages residents from participating in recycling. Based on an existing IBRS, this study investigates the joint optimization problem of the recyclable inventory management at each community recycling node and the vehicle routing from the recycling nodes to the recycling station. A two-stage dual-objective multi-period stochastic programming model is established to minimize the loss of recyclables and logistics costs, which is further reformulated using the weighting method and transportation cost approximation parameters. To solve the reformulated model, a three-phase iterative algorithm is designed by combining the progressive hedging algorithm and route splitting algorithm based on the Lin-Kernighan heuristic. A case study is conducted using data from Shanghai’s IBRS. The proposed joint decision model is superior to separate decisions and the three-phase iterative algorithm can reduce the average total cost by up to 42.12% compared to the genetic algorithm and the Iteration-Move-Search method in the literature. Additionally, a sensitivity analysis is conducted to provide managerial insights.

Nanoplastics indirectly compromise lettuce growth in hydroponic systems via microbial extracellular vesicles derived from Curvibacter fontanus

Recent studies confirm that nanoplastics (NP) cause severe microbial imbalances in various ecosystems, significantly affecting microbial diversity and abundance. Hydroponic systems vital for lettuce production are increasingly threatened by NP contamination in irrigation water and this issue is gaining global attention. This study investigates microbial species in hydroponic irrigation water altered by NP exposure and their impact on lettuce growth. While NP (108-1010 particles/L) did not directly harm or accumulate in lettuce, significant changes in water parameters and microbial communities were observed, particularly an increase in Curvibacter fontanus abundance. Inoculation of sterile irrigation water with NP and C. fontanus led to lettuce mortality, suggesting C. fontanus as a critical mediator. Furthermore, extracellular vesicles (EVs) isolated from C. fontanus, treated with NP, were shown to suppress leaf development, growth, antioxidant defenses, and lettuce survival. This study concludes that NP-induced microbial shifts, particularly involving C. fontanus EVs, indirectly harm hydroponic lettuce production.

Integrity-centered framework for determining protected areas boundary: An application in the China's national park

National parks are critical components of protected areas and are drawing increased global attention. Clear and rational boundaries may provide a scientific foundation for the sustainable protection and management of these areas. However, criteria for feasible national park delimitation and techniques for resolving conflicts between socioeconomic development and environmental protection are still poorly defined, particularly regarding integrity during park establishment. Consequently, a systematic analytical framework is essential for advancing research and practice. This study developed a systematic framework to determine the spatial boundaries, using the Kunlun Mountain region in China as a case for empirical analysis. Key issues in national park development—ecology and management—were systematically integrated into this framework. Most importantly, spatial connectivity, a critical concern in national park planning, was thoroughly considered for improving integrity optimization, an aspect often overlooked in previous delimitation studies. The results demonstrated that our framework effectively identified potential national park boundaries, achieving a balance between development and conservation. Furthermore, core protected and general control areas were delineated to support differentiated management approaches. This framework offers a scientific foundation for policymakers and managers to pursue sustainable protection and management of national parks.

A systematic comparison of intercultural and indigenous cultural dance education from a global perspective (2010–2024)

Cultural dance fosters social cohesion, emotional well-being, creative thinking, and cultural identity by conveying cultural values and social meanings. However, systematic comparative research on cross-cultural and indigenous cultural dance education remains limited. This study, adopting a global perspective, examines the current applications, core issues, and educational strategies in this field. Through a systematic literature review and grounded theory approach, the research scrutinizes studies conducted from 2010 to 2024. The results show that intercultural research primarily focuses on multicultural exchange, cultural integration of immigrant groups, and cultural acceptance among socioeconomically disadvantaged populations, highlighting the influence of intercultural adaptability and globalization. In contrast, indigenous cultural research emphasizes local culture and educational systems, with a particular focus on local cultural identity and educational reform. While both types of research reach a consensus on the importance of cultural transmission, diversity in teaching strategies, and the critical role of educators, significant differences remain in cultural identity and the design of educational content. Future research could concentrate on virtual dance education, innovation in indigenous cultures, the exploration of dance education’s role in mental health, and the deep integration of artificial intelligence into dance pedagogy. Furthermore, consideration of the potential impact of globalization and technological advancement on cultural identity and educational models will foster theoretical innovation and practical development in dance education.

Exploring the Effects of Attention Mechanisms on Road Crack Detection and Classification

Pavement cracks are a prevalent form of roadway distress that pose significant safety hazards, necessitating prompt detection and repair. Due to the extensive road network, traditional image processing-based crack detection methods exhibit limitations in recognition accuracy and speed. To address these challenges, this paper proposes an efficient pavement crack detection model based on YOLOv8, termed YOLOv8-ACD (YOLOv8 - Attention for Cracks Detecting), which integrates a global attention mechanism. YOLOv8-ACD enhances detection efficiency and accuracy by focusing on crack identification while filtering out most irrelevant information. We evaluated YOLOv8-ACD on the RDD2022 dataset, and experimental results demonstrate significant improvements in key performance metrics, such as F1-Score and mean Average Precision (mAP), compared to the original YOLOv8 and other mainstream models. The real-time processing capability of this model makes it suitable for practical road inspection and maintenance, effectively reducing the workload of maintenance personnel and enhancing roadway safety.

The Relationship Between Physical Literacy, Strategies and Approaches in Physical Education

Physical literacy is an idea or concept that has relatively recently developed in the international world, including Indonesia, when compared to other literacy concepts. This new knowledge has an important role when we carry out our teaching duties as physical education, sports and health teachers at school. Schools, through extracurricular physical education, sports and health activities, curricular projects to strengthen the profile of Pancasila students, as well as extracurricular sports activities have an important role in promoting physical literacy. This role should make schools a forum for creating learning that is appropriate to the level of development and characteristics of students. The aim of this research is to determine the relationship between physical literacy and strategies and approaches in physical education. This research uses a literature review method, namely by reviewing the results of previous research based on theories that can be accounted for to produce new, more up-to-date theories. Based on the cause and effect interpretation of 10 articles related to physical literacy with approaches to physical education, a very firm conclusion was drawn from this research, namely that physical literacy carried out in physical education increases physical fitness, potential and academic achievement.

Amateur Adult, House Husband, or Very Late Teenager? Lay Theories of Emerging Adulthood Among Rural- and Urban-Reared Thai Youth

Amateur Adult, House Husband, or Very Late Teenager? Lay Theories of Emerging Adulthood Among Rural- and Urban-Reared Thai Youth

AFENet: An Attention-Focused Feature Enhancement Network for the Efficient Semantic Segmentation of Remote Sensing Images

The semantic segmentation of high-resolution remote sensing images (HRRSIs) faces persistent challenges in handling complex architectural structures and shadow occlusions, limiting the effectiveness of existing deep learning approaches. To address these limitations, we propose an attention-focused feature enhancement network (AFENet) with a novel encoder–decoder architecture. The encoder architecture combines ResNet50 with a parallel multistage feature enhancement group (PMFEG), enabling robust feature extraction through optimized channel reduction, scale expansion, and channel reassignment operations. Building upon this foundation, we develop a global multi-scale attention mechanism (GMAM) in the decoder that effectively synthesizes spatial information across multiple scales by learning comprehensive global–local relationships. The architecture is further enhanced by an efficient feature-weighted fusion module (FWFM) that systematically integrates remote spatial features with local semantic information to improve segmentation accuracy. Experimental results across diverse scenarios demonstrate that AFENet achieves superior performance in building structure detection, exhibiting enhanced segmentation connectivity and completeness compared to state-of-the-art methods.

Beyond smell and sensationalism: remixing durian for and by Asians and Asian Canadians in Canada

ABSTRACT Increased imports of durian to Canada has reiterated what Wenying Xu identifies as the tension between dominant food sophistication and ‘crude’ immigrant practices, where denigrating immigrant foodways was integral to colonial assimilationist projects (2007). While fresh durian in Canada appears in sensationalist online media that associate durian with disgust, it is also made palatable and popularized by mainstream non-Asian food bloggers and chefs. Focusing on online food discourse, this article examines the fraught position of durian in dominant Canadian foodways as associated with disgust when imported for Asian consumers, and with intrigue and exoticism when marketed in ‘elevated’ dishes created or consumed by white Canadians. It proposes that durian has nevertheless developed into a diasporic cultural connector and site of community building within Asian and Asian Canadian creators and consumers that resists this Orientalization, linking durian’s emergence in Canada to a larger history of Asian food ingenuity and adaptation to global marketplaces. The ’remixing’ of durian by Asian chefs who combine Asian and Asian Canadian cuisines, and diasporic connections created over discussing durian in online spaces, contest xenophobic and racist sentiments focused on Asian foodways, positioning online responses to durian as an emerging yet significant site of community and resistance.

Deep Reinforcement Learning Recommendation System Algorithm Based on Multi-Level Attention Mechanisms

Traditional recommendation systems, which rely on static user profiles and historical interaction data, frequently face difficulties in adapting to the rapid changes in user preferences that are typical of dynamic environments. In contrast, recommendation algorithms based on deep reinforcement learning are capable of dynamically adjusting their strategies to accommodate real-time fluctuations in user preferences. However, current deep reinforcement learning recommendation algorithms encounter several challenges, including the oversight of item features associated with high long-term rewards that reflect users’ enduring interests, as well as a lack of significant relevance between user attributes and item characteristics. This leads to an inadequate extraction of personalized information. To address these issues, this study presents a novel recommendation system known as the Multi-Level Hierarchical Attention Mechanism Deep Reinforcement Recommendation (MHDRR), which is fundamentally grounded in a multi-layer attention mechanism. This mechanism consists of a local attention layer, a global attention layer, and a Transformer layer, allowing for a detailed analysis of individual attributes and interactions within short-term preferred items, while also exploring users’ long-term interests. This methodology promotes a comprehensive understanding of users’ immediate and enduring preferences, thereby improving the overall effectiveness of the system over time. Experimental results obtained from three publicly available datasets validate the effectiveness of the proposed model.

Text Mining for Ford Car Review Data – Based on Sentiment Analysis, LDA Topic Model, Semantic Web Analysis

As an established automotive brand with its origins in the early twentieth century, the Ford brand has developed a distinctive global influence over the course of decades. In order to gain insight into customer perceptions of the brand, this study employs a data set comprising customer evaluations of the Ford automobile between 2002 and 2018. Firstly, word frequency statistics are calculated in order to demonstrate the key points that are predominantly mentioned by customers. Subsequently, a sentiment analysis is conducted to categorise the keywords into two distinct groups: the positive and the negative. These groups represent the positive and constructive feedback provided by customers, respectively. Subsequently, latent Dirichlet allocation based on sentiment analysis was employed to identify five themes comprising positive and negative words. These included car appearance, user experience, car performance, comfort, and brand recognition. Furthermore, in order to gain insight into the semantic relationships between these reviews, a semantic network analysis was employed to demonstrate the interactions between the key points. In conclusion, the above analysis provides an overview of the Ford brand based on user reviews. The findings of this study may prove valuable in informing future improvements to Ford's automobile sales and manufacturing processes.

Beyond online and offline activism: #SaveSheikhJarrah and contemporary resistance in Palestine

ABSTRACT This article focuses on the interconnections between online and offline activism in Palestine, with a detailed analysis of the case of Sheikh Jarrah (east Jerusalem). We examine current strategies of Palestinian digital resistance that have highlighted the relevance of interlinking struggles both at the grassroots level and on the internet. Within the settler colonial framework, the hashtag #SaveSheikhJarrah stands as one of the most emblematic contemporary struggles against the Israeli military occupation, particularly against the ongoing evictions taking place in areas such as Sheikh Jarrah in east Jerusalem and the West Bank. In recent years, this hashtag has drawn unprecedented global attention to the Palestinian cause. This article is based on extensive digital ethnography and semi-structured interviews with young Palestinian activists conducted from 2021 to 2023. Through an exploration of activists’ narratives surrounding the #SaveSheikhJarrah movement, we critically examine the interplay of diverse forms of resistance in the age of social media. Our lens of analysis links grassroots and online struggles, offering a fresh perspective on digital activism in the current discourse on culture and communication within Palestinian studies.

Saving Russia. Neorealist Theorists and the Interpretation of the Kremlin’s International Policy according to the Actor-Network Concept

The Russian-Ukrainian war of 2014–2024 and the destructive role of Russia negatively affects regional and global aspects of the life of peoples and states. It has created a threat of serious chaos in international relations. Russia’s aggressive policy has caused a crisis in world politics that even the great powers of today cannot overcome. The return to a new Cold War and the outbreak of a major war in Europe threatens the entire modern international system with dangerous turbulence. The purpose of the study is to analyze the concept of effectiveness and the role of influential international actors in the modern Russian-Ukrainian war in the field of actor-network theory. The research methodology is based on the principles of science, objectivity, historicism and the basic conceptual and theoretical provisions of the study of world politics and international relations. Actor-network theory was introduced by Michel Callon and Bruno Latour in the second half of the 1980s. In accordance with this theory, relations between participants are determined by the network itself and are marked by the intensity and orientation of the network of interactions. In the sense of international relations and world politics, this means that they are under the strong or even decisive influence of powerful international actors (big powers and flexible interstate coalitions). Russia is seen as one of the most influential international players or “core of localities” that interact. Accordingly, other influential international actors are very cautious about the complete break of relations with Russia; they do not consider the possibility of its complete defeat in the aggressive war against Ukraine. Such a position was reflected and substantiated in various theoretical approaches, examples of which are the ideas of neorealism representatives such as John Mearsheimer, Farid Zakaria and Henry Kissinger.

Education as capital? A critical discourse analysis of the investment discourse in international research on Chinese rural education

ABSTRACT This article examines the rise of human capital theory and the investment discourse in education with a focus on Chinese rural education. We carry out Critical Discourse Analysis on 38 research articles published in highly cited English-language social science journals between 1978 and 2022. We demonstrate that Chinese rural education boomed as a topic of international research in the 21st century due to intensified academic internationalisation in China, which is intertwined with China’s pursuit of global influence and rising nationalism. The articles in our sample respond more to mainstream discourses dominated by the Global North than to the concerns of local communities, thereby reinforcing the dominance of western social theories and discourses like the human capital theory and the investment discourse. Amidst increasingly antagonistic global politics, we call for the critical ‘double negation’ (this issue) of both coloniality and nationalism to open up possibilities of plural knowledges.

Otogenic brain abscess and concomitant acute COVID 19 infection - case report and review of the literature.

COVID-19 has attracted global attention primarily because of the severe acute respiratory symptoms associated with it. However, nearly one third of the patients also present with neurologic symptoms. This report describes a case of a previously healthy woman with acute COVID-19 infection, who developed acute facial nerve palsy and rapid progression to coma due to otogenic brain abscess. A 63-year-old woman with acute COVID-19 infection exhibited acute facial nerve paresis, high fever and purulent secretion from her left ear within 48 hours after COVID-19 onset. Cranial CT scan confirmed acute mastoiditis, precipitating an urgent mastoidectomy. A postoperative contrast-enhanced MRI on the same day revealed a subdural empyema, which prompted an urgent craniotomy and decompression. Intraoperative microbiological swabs confirmed a Streptococcus pyogenes infection, however RT-PCR was negative for SARS-CoV-2. After immediate i.v. antibiotic treatment, extubation was achieved four days post-operation, and the patient was discharged without neurologic deficits 19 days after postoperatively. This finding adds a layer of insight into the specific nature of the infection, suggesting a potential absence of SARS-CoV-2 involvement in otogenic subdural empyema. However, the impact of SARS-CoV-2 in otogenic brain abscess cannot be excluded to date and should be further prospectively investigated. The complete recovery of neurological status, emphasizes the importance of prompt and interdisciplinary interventions in managing rare and severe complications associated with COVID-19.

Investigating The Nexus of Child Marriage and Economic Growth in Indonesia

The issue of ending child marriage attracts global attention, especially in developing countries. However, the link between married children and economic growth remains in question. The purpose of this article is to utilize regional macroeconomic indicators to quantify the association between child marriages and the Indonesian economy. The analysis shows a negative relationship between marriage at a young age and the economy using panel regression and simultaneous equations model. The results also demonstrate that marriage at an early age might decrease economic growth, as it has an undesirable effect on health and education. Therefore, this paper delivers a suggestion related to the policy strategies, primarily educational policies and health, that support the Indonesian Government in ending child marriage.