Purposes: This study examines the relationship between ESG performance, financial constraints, and corporate tax avoidance. It investigates whether ESG performance reduces tax avoidance by promoting socially responsible practices and whether financial constraints lead to increased tax avoidance as a strategy to generate internal funds. Specifically, it seeks to assess how ESG practices and financial constraints influence tax avoidance behavior among publicly listed firms in ASEAN countries. Various determinants of tax avoidance have been identified in the literature; however, ESG performance and financial constraints have received less attention in association with tax avoidance from a regional perspective in a group of countries with similar yet diverse institutional contexts, such as ASEAN. It offers new insights into these relationships in a multi-country, emerging market setting.
Methods: Using an unbalanced panel dataset of 242 publicly listed companies in ASEAN-5 over the 2019–2023 period with generalized least squares (GLS) regression.
Findings: ESG performance is negatively associated with tax avoidance, as evidenced by higher Effective Tax Rate (ETR). Additionally, firms facing financial constraints exhibit a positive relation- ship with tax avoidance, leading to lower ETR. Financially constrained companies are hindered in accessing external funds, thus pursuing aggressive tax planning strategies to generate additional internal funds.
Novelty: This study contributes to the literature by extending the understanding of tax avoidance behavior in emerging markets and at a regional level. It provides empirical evidence from the ASEAN market, which has been underrepresented in prior studies.

The Vocational Life Skills Program is an educational breakthrough designed to equip students with applied skills, respond to globalization, and strengthen skill-based education in the madrasah environment. This study aims to analyze the implementation of the vocational life skills program to enhance skills education at State Islamic Senior High School 2 Banyumas. The researcher used a descriptive qualitative method with a case study design. Data were collected through observation, interviews, and documentation, and then analyzed interactively through data reduction, presentation, and conclusion drawing. The results show that the Life Skills Program is intrakurricular and dialectical, allowing students to choose majors based on their interests. Learning comprises 70% practical experience through industrial visits, a 160-hour internship, and 30% theory. It involves eight majors emphasizing Islamic values, independence, and job readiness. This program develops technical and soft skills, fosters professional ethics, teamwork, and communication, preparing graduates to be highly competitive and accomplished. Challenges include a complex curriculum, limited time, inadequate facilities, and teacher competency. Solutions involve project-based learning, industry collaboration, digital simulations, and continuous teacher training to keep the program relevant and practical. Conclusion: Strengthening the vocational life skills program is a strategic effort to equip students with practical skills, develop holistic character, and respond effectively to the challenges of 21st-century life.

the digital and globalization era, entrepreneurs face complex challenges related to intellectual property rights (IPR) violations, such as piracy, brand counterfeiting, and idea theft. This phenomenon not only has an impact on the legal aspect, but also creates significant business ethics dilemmas. This study aims to analyze the relationship between the application of business ethics and IPR protection in the context of entrepreneurship, and to explore how the integration of the two can improve business competitiveness and sustainability. The methodology used is a qualitative literature study with a descriptive-analytical approach, reviewing various sources such as scientific journals, legal regulations related to IPR, and case studies of entrepreneurs who apply ethical principles and IPR protection. The results of the study indicate that the application of strong business ethics has a significant impact on the awareness and compliance of business actors in protecting intellectual property. Meanwhile, an understanding of IPR law can encourage innovation and create a fair and sustainable business climate. This is in accordance with the theory of the application of business ethics, IPR protection and legal protection for consumers have a significant influence on business, emphasizing the importance of integrating the three to support fair and sustainable business practices. This study confirms that the integration of business ethics and IPR protection is not only a formal legal aspect, but also an important strategy in building a positive image, consumer trust, and long-term business sustainability. Therefore, education and awareness of the importance of business ethics and IPR protection need to be increased among entrepreneurs to create a healthy, competitive and sustainable business ecosystem.

Medical image segmentation plays a vital role in diagnosis and treatment planning by extracting clinically relevant information from imaging data. Conventional methods often struggle with variations in anatomical structure and imaging quality, leading to suboptimal segmentation. Recent advancements in Deep Learning, particularly Convolutional Neural Networks (CNNs) and Transformers, have improved segmentation accuracy; however, individual models such as U-Net, ResNet, and Transformer still face limitations in preserving spatial details, extracting deep features, and modeling long-range dependencies. This study proposes a hybrid Deep Learning model that integrates U-Net, ResNet, and Transformer to overcome these challenges and enhance segmentation performance. The proposed hybrid model was evaluated on several publicly available datasets, including BraTS, ISIC, and DRIVE, using Dice Similarity Coefficient (DSC) and Intersection over Union (IoU) as performance metrics. Experimental results indicate that the hybrid model achieved a DSC of 0.92 and an IoU of 0.86, outperforming U-Net (DSC: 0.82, IoU: 0.75), ResNet (DSC: 0.85, IoU: 0.78), and Transformer (DSC: 0.88, IoU: 0.80). Additionally, the model maintained an inference time of 55 ms per image, demonstrating its potential for real-time applications. This study highlights the benefits of combining CNN-based and Transformer-based architectures to capture both local details and global context, providing an effective and efficient solution for medical image segmentation.

The rapid advancement of deepfake technology has raised significant concerns regarding misinformation, privacy breaches, and digital fraud. Existing deepfake detection models, particularly those based on deep learning, often require high computational resources, making them unsuitable for real-time applications on mobile devices. This study aims to develop a lightweight deepfake detection model that enhances accuracy while maintaining computational efficiency. To achieve this, we propose a hybrid approach that integrates Fast Fourier Transform (FFT), MobileNet, and an Attention mechanism. The FFT component enables frequency-domain analysis to detect subtle deepfake artifacts, while MobileNet provides a lightweight convolutional backbone, and the Attention layer enhances feature extraction. The proposed model was evaluated on a benchmark deepfake dataset, and the results demonstrated its superior performance compared to the standard MobileNet model. Specifically, the model achieved an accuracy of 94.2%, an F1-score of 93.8%, and a computational efficiency improvement of 27.5% in comparison to conventional CNN-based approaches. These findings indicate that the integration of FFT and Attention mechanisms significantly enhances the model's capability to distinguish real and manipulated media while reducing computational overhead. The contribution of this study lies in presenting a deepfake detection model that balances accuracy and efficiency, making it suitable for deployment in mobile and resource-constrained environments. Future research should explore further optimization for energy efficiency, the adoption of lightweight Transformer architectures, and extensive testing on diverse datasets to improve robustness against real-world variations.

Sentiment analysis for low-resource languages, particularly Austronesian languages, remains challenging due to the limited availability of annotated datasets. Traditional approaches often struggle to achieve high accuracy, necessitating strategies like cross-lingual transfer and data augmentation. While multilingual models such as mBERT offer promising results, their performance heavily depends on fine-tuning techniques. This study aims to improve sentiment analysis for Austronesian languages by fine-tuning mBERT with augmented training data. The proposed method leverages cross-lingual transfer learning to enhance model robustness, addressing the scarcity of labeled data. Experiments were conducted using a dataset enriched with augmentation techniques such as back-translation and synonym replacement. The fine-tuned mBERT model achieved an accuracy of 92%, outperforming XLM-RoBERTa at 91.41%, while mT5 obtained the highest accuracy at 99.61%. Improvements in precision, recall, and F1-score further validated the model’s effectiveness in capturing subtle sentiment variations. These findings demonstrate that combining data augmentation and cross-lingual strategies significantly enhances sentiment classification for underrepresented languages. This study contributes to the development of scalable Natural Language Processing (NLP) models for Austronesian languages. Future research should explore larger and more diverse datasets, optimize real-time implementations, and extend the approach to tasks such as Named Entity Recognition (NER) and machine translation. The promising results underscore the importance of integrating robust transfer learning techniques with comprehensive data augmentation to overcome challenges in resource-limited NLP scenarios

Radar systems play a crucial role in target tracking within urban environments, where challenges such as clutter, multipath effects, and electromagnetic interference significantly impact detection accuracy. Traditional radar methods often struggle to adapt to dynamic urban conditions, leading to decreased reliability in real-time target tracking. This study aims to develop and evaluate an AI-driven adaptive radar system that enhances tracking accuracy in urban settings. The research employs a quantitative approach using simulations to model radar signal processing under various environmental conditions. The AI model, based on Convolutional Neural Networks (CNN), is trained to optimize radar performance by filtering out noise and dynamically adjusting detection parameters. The results indicate that the AI-based radar system achieves a tracking accuracy of 95.2%, significantly outperforming traditional radar systems, which only reach 80% accuracy. Additionally, the AI-enhanced radar reduces response time to 120 milliseconds, compared to 250 milliseconds in conventional systems, demonstrating improved real-time processing capabilities. The system also exhibits greater resilience to high-clutter environments, maintaining stable target detection despite signal interference. These findings highlight the potential of AI in enhancing radar functionality for applications such as surveillance, traffic monitoring, and security. Future research should focus on integrating AI-driven radar with real-world radar hardware, exploring multi-sensor fusion, and refining adaptive learning techniques to further optimize tracking performance in complex environments

The persistent challenges of polysemy and ambiguity continue to hinder the semantic accuracy of Neural Machine Translation (NMT), particularly in language pairs with distinct syntactic structures. While transformer-based models such as BERT and GPT have achieved notable progress in capturing contextual word meanings, they still fall short in understanding explicit semantic roles. This study aims to address this limitation by integrating Semantic Role Labeling (SRL) into a Transformer-based NMT framework to enhance semantic comprehension and reduce translation errors. Using a parallel corpus of 100,000 English-Indonesian and English-Japanese sentence pairs, the proposed SRL-enhanced NMT model was trained and evaluated against a baseline Transformer NMT. The integration of SRL enabled the model to annotate semantic roles, such as agent, patient, and instrument, which were fused with encoder representations through semantic-aware attention mechanisms. Experimental results demonstrate that the SRL-integrated model significantly outperformed the standard NMT model, improving BLEU scores by 6.2 points (from 32.5 to 38.7), METEOR scores by 6.3 points (from 58.5 to 64.8), and reducing the TER by 5.8 points (from 45.1 to 39.3). These results were statistically validated using a paired t-test (p < 0.05). Furthermore, qualitative analyses confirmed SRL's effectiveness in resolving lexical ambiguities and syntactic uncertainties. Although SRL integration increased inference time by 12%, the performance trade-off was deemed acceptable for applications requiring higher semantic fidelity. The novelty of this research lies in the architectural fusion of SRL with transformer-based attention layers in NMT, a domain seldom explored in prior studies. Moreover, the model demonstrates robust performance across linguistically divergent language pairs, suggesting its broader applicability. This work contributes to the advancement of semantically aware translation systems and paves the way for future research in unsupervised SRL integration and multilingual scalability.

In the era of Industry 4.0, optimizing fluid flow systems in smart factories is essential to improve energy efficiency and operational stability. Traditional Computational Fluid Dynamics (CFD) simulations provide accurate fluid flow analysis but require extensive computational resources and long processing times, making real-time applications challenging. To address this limitation, this study aims to develop an AI-based predictive model for CFD simulations, utilizing Convolutional Neural Networks (CNN) and Extreme Gradient Boosting (XGBoost) to accelerate the estimation of fluid flow characteristics in industrial environments. The research methodology involves generating CFD simulation datasets, preprocessing data, and training AI models to predict key fluid parameters such as pressure, velocity, and temperature. The evaluation results show that CNN achieves a Mean Squared Error (MSE) of 0.0025 and a Root Mean Squared Error (RMSE) of 0.05, outperforming XGBoost, which records an MSE of 0.0030 and an RMSE of 0.055. Moreover, CNN predicts fluid dynamics in just 15.2 seconds, while XGBoost achieves results in 10.5 seconds, compared to the 1200.5 seconds required by traditional CFD simulations. These findings highlight the potential of AI in reducing computation time by over 98%, making real-time fluid flow analysis feasible in industrial settings. This study contributes to the advancement of AI-integrated CFD modeling, demonstrating that AI can significantly enhance the efficiency of fluid dynamics analysis without compromising accuracy. Future research should focus on expanding AI models to handle more complex flow conditions and integrating AI with smart factory design tools for real-time optimization

This research is motivated by the problem of communication in economic transactions in the Aru Islands District, which is recruited by a multi-ethnic community with a variety of regional languages. Although Indonesian has great potential as a national communication tool, its use in local economic activities is still limited, thus hampering the smooth interaction between traders and buyers. This communication inefficiency has an impact on the negotiation process and local economic competitiveness. This study aims to instill the role of Indonesian in facilitating economic transactions, identify barriers to use, and develop strategies to improve Indonesian language skills in the community. The approach used is descriptive qualitative with a phenomenological method. Data were collected through observation, in-depth interviews, and document studies, with informants selected using purposive sampling techniques to obtain data relevant to the local economic context. Expected results include an understanding of the contribution of Indonesian in increasing the effectiveness of transactions, revealing communication challenges, and strategies for improving contextual and applicable language literacy. These findings can provide practical contributions to local governments, educational institutions, and business actors in developing language training programs that support economic activities. From a theoretical perspective, this research enriches the study of the relationship between language and economic development, especially in multilingual areas such as the Aru Islands.