Abstract
This study examines the relationship between quality of financial information (QFI) and the probability of corporate decline in the food manufacturing sector within the European Union (EU) controlled by a set of internal (age, size, liquidity, return on assets, and debt) and external (gross domestic product and unemployment rate) determinants. The study employs a logit regression model applied to a balanced panel dataset of 335 large food manufacturing firms in the EU from 2011 to 2021. Quality of financial information is estimated using discretionary accruals, based on the Jones model (1991), while corporate decline is measured by fluctuations in sales. The findings indicate that low-quality financial information (high discretionary accruals) is positively associated with corporate decline, highlighting the role of financial transparency in business sustainability. The results suggest that earnings management practices can increase business vulnerability, reinforcing the importance of accurate financial reporting in mitigating corporate failure. The study underscores the need for enhanced regulatory oversight and financial reporting transparency in the food manufacturing industry. Policymakers and stakeholders should strengthen financial disclosure requirements to curb earnings management practices and ensure better resource allocation for long-term sustainability. This research contributes to the limited literature on quality of financial information and corporate decline, particularly in the food manufacturing sector, which is crucial for economic stability and public welfare. By integrating financial reporting quality into corporate failure analysis, this study provides new insights into the role of earnings management in business deterioration.

Abstract
Outdoor learning has been used in different European contexts to support personal and social development. However, the ways in which these initiatives are organised and implemented can vary according to the target group, the educational purpose, and the local context. The aim of this study was to comparatively analyze selected Outdoor Learning initiatives for young people and adults in Europe, focusing on their main characteristics and on the forms of development they promote. A qualitative exploratory review was carried out using websites of educational networks, NGOs, training centers, and institutional projects. Seven initiatives were selected from an initial set of 27 cases identified through the search process. The selected initiatives were Venture Trust, Venture Out, Forest School, KOA Aventura, Moc Warmii, Die Seminarschneider, and LIFE Trekking. The results show a diversity of program durations, pedagogical approaches, and participant profiles, but also some common elements, such as experiential learning, reflection, cooperation, and contact with nature. The analysis suggests that Outdoor Learning initiatives may support different forms of personal and social development, although their design and aims are strongly influenced by the context in which they are developed.

Abstract
Multi-label image classification datasets are often partially labeled where many labels are missing, posing a significant challenge to training accurate deep classifiers. Most existing approaches assume the missing labels as negatives and/or exploit image and category relationships to regularize training. Orthogonally, this paper studies blending samples in such incomplete datasets as new samples, extending the training data magnitude to increase generalization. First, the proposed LogicMix mixes multiple partially labeled samples to produce new samples, where their unknown labels are naturally mixed by OR's logical equivalences, without replacement with constants. Subsequently, a Decouple Partial-Asymmetric Loss is proposed to assign separate label-focusing policies to original and new samples, addressing the learning imbalance from the different positive-negative label imbalances between original and augmented samples. Finally, we propose a complete learning framework called 2WayAug-PL. LogicMix and conventional data augmentation collaborate to extend the diversity of new samples in both the sample-sample relation and human prior knowledge, while pseudo-labeling compensates for the lack of labels to provide more supervision signals. 27 partially labeled dataset scenarios derived from three benchmarking datasets with various learning difficulties are utilized for comprehensive experiments. LogicMix has shown remarkable effectiveness and generality in improving mAP against compared sample-mixing data augmentation methods. In particular, 2WayAug-PL achieves state-of-the-art average mAP of 84.3%, 50.1 %, and 93.8% on MS-COCO, VG-200, and Pascal VOC 2007, respectively. It further pushes the previous best performance achieved by different frameworks by 0.6% (CFT), 0.6% (CFT), and 0.1 % (SR). Moreover, 2WayAug-PL significantly outperforms all compared frameworks, as shown by statistical tests. Code is available at: https://github.com/maxium0526/logic_mix.

Abstract
This study examines how the quality of financial information (QFI), using financial reporting discretion, influences trade credit decisions for Portuguese medium-sized retail companies, particularly in contexts of bank financing constraints (2012 to 2022). Findings indicate that isuppliers respond more to optimistic financial signals-specifically, income-increasing discretionary accruals-than to intrinsic accounting quality. Instead of relying purely on formal data, suppliers prioritize relational and operational factors, such as trust, interpersonal relationships, debt history, payment terms and operational profitability. Additionally, company size, asset tangibility and profitability negatively influence trade credit. Firms with low asset tangibility tend to utilize trade credit more frequently because they lack sufficient collateral to secure bank financing. As tangibility increases, these firms may access bank credit at lower costs and consequently substitute trade credit. A similar pattern is observed for firm size, with smaller firms relying more heavily on supplier credit, whereas larger firms display lower dependence on this financing channel. Moreover, greater availability of internal resources reduces the need for supplier-based financing. Our results contradict part of the existing literature, which predicts positive relationships between trade credit, firm size, tangibility and profitability. The study highlights the importance of fostering trust between companies and suppliers, as well as enhancing financial transparency to secure more favorable financing conditions.

Abstract
In today’s digital age, email communication plays a crucial role in both personal and professional settings. However, the increasing volume of spam emails poses significant challenges, leading to wasted time, security risks, and potential financial losses. This work explores various Machine Learning models to classify emails as spam or non-spam, aiming to enhance the accuracy and efficiency of spam detection systems. The study involves data preprocessing, feature extraction, and the implementation of supervised learning techniques such as Naïve Bayes, Support Vector Machines, and Neural Networks. A comparative analysis is performed based on key performance metrics, including Accuracy, Precision, and Recall. Furthermore, the impact of feature selection and dataset balancing on classification results is examined. The findings provide valuable insights into the strengths and limitations of different models, contributing to the development of more reliable spam detection models. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.

Abstract
The video game industry has grown to become one of the largest in the market, surpassing even the film industry over a decade ago (Statista in Video game industry revenue worldwide 2000-2020). However, the development of games designed with visually impaired players in mind is still almost non-existent when compared to the sheer number of games released yearly. NonVisual Pong is our approach to addressing this challenge, providing blind players with a way to engage in competitive fun through gaming. We took the original Pong game from 1972 and fully adapted it to be played using only a controller-no visual display required. Following the development process, we tested our implementation with experts, discovering that, overall, our game was easy to pick up, required no overly complex setup, and successfully delivered the intended experience. Players enjoyed a balanced challenge and immersion, facilitated by audio cues and the controller's vibrations.