Search Results (103)

Accurate and robust weed detection is a critical prerequisite for precision agriculture and site-specific weed management. However, real-world agricultural environments pose significant challenges to existing object detectors due to severe illumination variability, high inter-class similarity between crops and weeds, and the prevalence of small and occluded targets at early growth stages. To address these challenges, this paper proposes DSD-YOLOv11, a domain-adaptive and structurally refined detection framework tailored for complex field scenarios. Specifically, a physics-based data augmentation strategy is first introduced to simulate realistic illumination conditions and soil background variations, effectively broadening the training distribution without increasing model complexity. In addition, a lightweight Feature Enhancement Module (FEM) is selectively injected at the P3 detection layer, where high-resolution features are preserved. The FEM integrates a SpatialAttentionLite mechanism with a projection-based feature alignment strategy, enabling precise enhancement of fine-grained spatial cues while maintaining compatibility with pre-trained backbones. An epoch-aware alpha controller is further designed to ensure stable optimization by gradually activating the enhancement pathway during training. Extensive experiments on a real-world agricultural weed dataset demonstrate that the proposed method consistently outperforms baseline YOLOv11 models across multiple evaluation metrics. Notably, DSD-YOLOv11 achieves an absolute mAP@50 improvement of +12.73 percentage points over the native baseline without data augmentation (reaching 87.14%, where the physics-based augmentation contributes +7.94 percentage points and the FEM module contributes an additional +4.79 percentage points over the augmented YOLO11n baseline), while operating at 84.2 FPS on a desktop GPU (NVIDIA RTX 4090; NVIDIA Corporation, Santa Clara, CA, USA) and 7.2 FPS on an edge computing platform (NVIDIA Jetson Nano; NVIDIA Corporation, Santa Clara, CA, USA) with only marginal parameter increases. These results indicate that the proposed framework provides an effective and efficient solution for weed detection in unstructured agricultural environments. Full article

Human-Centered AI (HCAI) has emerged as a guiding paradigm for designing AI systems that align with human values, needs, and well-being, yet the field lacks consensus on what constitutes human-centeredness. This study addresses that gap through a four-phase sequential mixed-methods design: (1) thematic analysis of 81 HCAI definitions from academic, institutional, and industry sources, yielding 78 keywords; (2) frequency-based statistical categorization; (3) expert evaluation producing a final inventory of 26 attributes; and (4) a cross-sectional survey (N = 145), predominantly drawn from the Asia-Pacific region (77.2%, with Myanmar, Singapore, and Thailand most represented), in which practitioners, academics, and students rated each attribute on a 7-point Likert scale, complemented by a reflexive thematic analysis of open-ended responses. The 26-item scale demonstrated excellent internal consistency. Trust, values, benefits, needs, and usability were rated most highly, while affective and cognitive attributes—emotions, behaviours, and empathy—were consistently rated lower, a pattern the qualitative data suggest reflects perceived intractability rather than indifference. Inter-attribute correlations revealed interpretable substructures, including an experience/usability cluster, an emotion/empathy cluster, and a participatory engagement cluster, while human control operated as a conceptually independent dimension. Five qualitative themes provided interpretive context: user needs and augmentation as design drivers, ethical foundations and value alignment, trust as a relational outcome contingent on transparency, the complexity of human experience as a design challenge, and structural barriers including corporate incentives, regulatory gaps, and resource constraints. In this predominantly Southeast Asian sample, all three stakeholder groups showed substantial agreement on which attributes matter most and least. The primary divergence ran between academics and students: academics assigned higher importance to participatory and process-oriented attributes, while students emphasized tangible outcomes. Practitioners occupied an intermediate position, with a distinctive emphasis on ethical values. These findings offer an empirically grounded vocabulary for human-centeredness, positioned as an exploratory foundation for future psychometric refinement, with implications for HCAI design practice, education, and cross-stakeholder dialogue. Full article

Corporate sustainability and the reliability of ESG reporting have gained relevance in the evaluation of listed companies, particularly in emerging capital markets, where reporting practices are still in their early stages of development. The purpose of this study is to analyze the relationship between the quality of ESG reporting, the risk of greenwashing estimated using a proxy derived from reported information, and the market value of companies listed on the Bucharest Stock Exchange. The research employs a mixed-methods design, involving content analysis of annual reports, sustainability reports, and sustainability statements for 25 companies over the 2020–2024 period. The scores corresponding to the Environmental, Social, and Governance dimensions, as well as the proxy for greenwashing risk, were developed using an ordinal scoring grid, which was validated through inter-rater assessment. During the course of the study, the empirical relationships were tested using pooled OLS specifications on short panel data, incorporating the natural logarithm of market capitalization, financial controls, year effects, and sector dummy variables. The results highlight the presence of an association between the quality of ESG reporting and market value, particularly for environmental and social dimensions, while the greenwashing risk proxy exhibits a limited statistical influence. The study contributes to the literature on ESG reporting in emerging markets and highlights the need for a cautious interpretation of indicators constructed based on corporate disclosures. Full article

Enterprise financial risk prediction is a fundamental task in financial risk management, yet its performance is often hindered by severe class imbalance, cross-enterprise heterogeneity, and the limited availability of labeled risky samples. These challenges are particularly pronounced in few-shot settings, where conventional machine learning and deep classification models tend to suffer from unstable representation learning, feature collapse, and weak decision boundaries. To address this issue, this study proposes a hierarchical metric learning framework for few-shot enterprise financial risk prediction on imbalanced tabular data. The framework integrates a state-space feature embedding network, an Adaptive Spectral Decomposition and Multi-Scale State Embedding module, and a Hierarchical Metric Manifold Alignment mechanism to enhance risk-sensitive representation learning, preserve geometric consistency across embedding levels, and improve prototype-based discrimination in the metric space. Experiments are conducted on three public datasets, namely American Bankruptcy, Corporate Financial Risk Assessment, and Enterprise Financial Network, under a unified 2-way 20-shot setting. The proposed method consistently achieves the best overall performance across Precision, Recall, Accuracy, F1-score, and AUC, with AUC values of 0.9526, 0.9687, and 0.9716 on the three datasets, respectively. Ablation studies and visual analyses further show that the proposed framework improves intra-class compactness, inter-class separability, and classification robustness under highly imbalanced conditions. These findings indicate that the proposed method provides an effective and robust machine learning solution for enterprise financial risk prediction and early warning in data-scarce financial scenarios. Full article

Perishable products are usually priced in real-time to volatile market environments, thereby optimizing inventory control, minimizing resource wastage, and maximizing corporate profitability. Based on the public dataset from the 2023 Higher Education Press Cup National College Students Mathematical Modeling Competition, this paper addresses the challenge of multi-product joint pricing for perishable fresh produce and proposes an attention-augmented proximal policy optimization algorithm (termed ATT-PPO), which embeds an attention mechanism into the proximal policy optimization (PPO) framework. The integrated attention mechanism confers three core advantages to the model: first, it dynamically captures inter-product interdependencies, enabling an accurate reflection of cross-price elasticity and demand correlations; second, it reduces feature redundancy and computational overhead in multi-product collaborative pricing strategies; third, it enhances both the interpretability and computational efficiency of the model. Experimental results demonstrate that in the scenario of multi-product pricing, the ATT-PPO algorithm achieves competitive performance compared to PPO, DDPG (Deep Deterministic Policy Gradient), SAC (Soft Actor-Critic), and TD3 (Twin Delayed Deep Deterministic Policy Gradient), with the key advantage lying in its ability to provide interpretable attention weights that reveal dynamic cross-product dependencies in pricing decisions. This study not only expands the applicability of DRL (Deep Reinforcement Learning) to practical economic problems in the fresh produce sector but also provides valuable theoretical insights that can be generalized to other short-lifecycle product domains, including fashion apparel and consumer electronics. Full article

Designing an aircraft requires multidisciplinary analysis and data processing abilities, which are often spread over various partners. Effective collaboration across organisational boundaries is difficult, but essential. As the aerospace industry becomes increasingly digitalised, ever larger volumes of data and models must be exchanged. Heterogeneous tools, data formats, and infrastructures make it difficult to exchange data and to trace it. We propose using semantic graphs for data exchange to ensure interoperability, while semantic links between data models facilitate multidisciplinary and cross-organisational collaboration. Furthermore, our approach captures comprehensive metadata that describes the creation and modification of each dataset, thereby establishing a fully traceable data provenance chain. We demonstrate its functionality via a design process for an electromechanical actuator (EMA) given requirements from a different stakeholder (simulated). Having the requirements and the EMA models translated in Resource Description Framework (RDF) graphs, we are able to create links between them. This then enables the EMA model to be automatically re-evaluated when requirements change, ensuring that it complies with them. For the data exchange, we use the DLR SemanticHub, which utilises a graph database. By providing traceability of the data results provided in different data formats and the data origins, we enable transparency and accountability across organisational boundaries, which is important for trusted collaboration and compliance in intercorporational data exchange. Full article

Amid the dual pressures of the global energy transition and green technology upgrading, corporate environmental responsibility increasingly depends on interactions among firms rather than on isolated firm-level resources. From a systems perspective, this study focuses on the inter-firm green innovation linkages within the new energy sector, where knowledge diffusion, technological learning, and governance signals are jointly shaped by network structure. Using quarterly panel data from 52 listed Chinese new energy firms from 2018Q1 to 2023Q2, we employ the Adaptive Elastic Net Generalized Method of Moments approach to reconstruct a green innovation network from the observed dynamics of the panel data, and examine how firms’ positions within the network affect their environmental responsibility. The results show that the network exhibits a clear core–periphery spillover structure. Inter-firm ties are more likely to form when firms are located in the same province and when target firms have higher green patent citation impact and more executives with environmental backgrounds. Higher network centrality is associated with better corporate environmental responsibility, especially among firms facing intense market competition, state-owned firms, and non-key environmental regulatory units. These findings suggest that green innovation networks can alleviate innovation imbalances and strengthen informal inter-firm governance mechanisms in emerging green industries. Full article

Global climate governance and corporate low-carbon transition have made carbon information disclosure important for assessing firms’ environmental governance and climate-risk responses. This study develops an industry-specific carbon information disclosure quality (CIDQ) framework for Chinese A-share listed petroleum and petrochemical firms, using 45 firm-year observations from 15 firms during 2022–2024. The framework includes 7 primary, 15 secondary, and 33 tertiary indicators. Disclosure texts were scored by the DeepSeek-V3.2 large language model (LLM) under predefined rule-based criteria, with temperature set to 0. Reliability was assessed against manual scoring of 15 reports, yielding an intraclass correlation coefficient (ICC) of 0.974. The full-sample mean score is 34.02, accounting for only 51.55% of the theoretical maximum of 66, indicating that the overall disclosure level remains relatively low. The annual mean score increased from 29.07 in 2022 to 37.60 in 2024, representing a cumulative rise of 8.53 points, or 29.34%. Substantial inter-firm differences are also observed: Sinopec recorded the highest three-year average score of 52.67, whereas Yunnan Yunwei recorded the lowest at 10.67. This study may provide a methodological reference for structured CIDQ evaluation and disclosure improvement in high-emission industries. Full article

Freshwater scarcity is emerging as a critical constraint on industrial clusters, production networks, and urban service systems, where water functions simultaneously as an essential production input and a shared regional resource. This study investigates how post-allocation water-quota imbalances in large inter-basin diversion systems can be addressed through adaptive secondary water rights trading. Focusing on China’s South-to-North Water Diversion Project (SNWDP), the research aims to explain under what institutional and efficiency conditions water rights trading can enhance corporate social responsibility, environmental management, and sustainable supply chain resilience. The study’s main innovation lies in the development of a corporate-oriented evolutionary game model that links water governance with corporate production, urban–industrial demand, and responsible supply chain management. Unlike conventional models, it incorporates bounded rationality, heterogeneous water-use efficiency, information asymmetry, transaction costs, primary allocation water pricing, and the risk of unrecovered basic water fees. Using a case inspired by the Zhengzhou–Nanyang transaction along the Middle Route of the SNWDP, the model simulates the strategic interaction between a water-rich node with surplus quota and a water-scarce node facing deficit demand. The findings show that a socially desirable Trade–Trade equilibrium emerges only when efficiency expectations and institutional conditions are favorable. Lower transaction costs and basic water prices, higher sunk-fee risk, and clearer efficiency differentials significantly increase trading willingness. The study demonstrates the practical value of transparent secondary water markets in improving allocative flexibility, reducing governance rigidity, and promoting more responsible and environmentally efficient regional water management. Full article

Motivation: In modern organizations where remote and hybrid work has become normalized, fostering trust without frequent face-to-face interaction is a critical management challenge. This study aims to explore how non-verbal digital dynamics associate with trust formation within hybrid product development teams from a teamwork science perspective, integrating Big Five traits and established trust scales. Methods: The empirical study observed twelve product development teams (N = 40) participating in a major innovation competition over an eight-month period. Dynamic behavioral data, including speaking time, nodding, smiling, and silence, were extracted from online workshop recordings using synchronized behavioral coding validated by high inter-rater reliability (Cohen’s Kappa k ≥ 0.78). These were integrated with Big Five personality traits, mutual trust scales, and idea value metrics into a Bayesian Network (BN) to model probabilistic dependencies. The structural model was validated using the Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) to ensure predictive robustness. Furthermore, we performed sensitivity analysis on the BN to quantify how specific shifts in non-verbal cues—particularly nodding and the functional categories of silence—disproportionately affect the “Mutual Trust” node. While this exploratory study utilizes a sample of “digital native” student teams, it provides a critical baseline for “high digital fluency” collaboration, which we contextualize against the “asymmetric cues” found in multi-generational corporate environments. Results: Sensitivity analysis identified specific probabilistic associations suggesting that effective role fulfillment is the strongest predictor of idea originality. Crucially, nodding was identified as a behavioral ‘digital reward’ that enhances psychological safety, facilitating divergent thinking. Smiling showed a strong association with feasibility and consensus-building during convergent phases. The model further identifies distinct behavioral ‘fingerprints’: high-trust sequences are characterized by frequent non-verbal backchanneling and deliberate “thinking silences,” whereas low-trust sequences exhibit a disproportionate increase in unproductive lapses (e.g., a 10% increase in lapses correlating with an 18% decrease in trust probability). Furthermore, a probabilistic pathway was identified where teams with highly open members and frequent non-verbal validation exhibit higher mutual support behaviors. Conclusions: This research offers empirical insights into how trust can be modeled in hybrid environments through specific combinations of behavioral and personality traits. Practically, this study proposes “Hybrid Team Protocols”—such as intentional backchanneling and the normalization of deliberative silence—as actionable Organizational Development (OD) interventions. These provide managers with data-driven guidelines to visualize and monitor the quality of digital collaboration while emphasizing the ethical necessity of transparent implementation to prevent “digital performance” and ensure psychological safety across diverse organizational structures. Full article

This study assesses the elasticity between integrated media performance and corporate reputation by examining the relationship between Media Output Score (MOS) and RepScore™ in the Portuguese telecommunications sector (Altice/MEO, NOS, and Vodafone) between 2021 and 2023. Adopting a longitudinal observational design, the analysis compares inter-annual variation in communication output with corresponding changes in stakeholder-based reputation. Media performance is operationalized through MOS as a composite indicator of visibility, favorability, readership, targeting, and social amplification, while corporate reputation is measured using third-party RepScore™ data. The findings indicate directional alignment between media output and corporate reputation; however, the magnitude of reputational adjustment appears substantially lower than the amplitude of media volatility. Across heterogeneous crisis contexts, including cybersecurity incidents and governance-related events, reputational scores exhibit incremental and comparatively stable evolution despite pronounced fluctuations in media performance. These results suggest that the relationship between media output and corporate reputation is characterized by constrained responsiveness at the annual level, consistent with a stock–flow interpretation in which communication signals operate as high-variance flows and reputation evolves as a path-dependent stock. By empirically illustrating this asymmetry, the study contributes to media influence research by identifying a structural boundary condition in the translation of media exposure into stakeholder evaluation. The findings further clarify the analytical distinction between output-level communication metrics and outcome-level reputational constructs in digital media environments. Full article

The mobile industry has experienced long-run changes in its knowledge structure, including identifiable transition points observable through embedding-based semantic analysis. Using abstracts from 86,674 mobile industry publications published between 2005 and 2024, we embed documents with SPECTER2, build year-specific embedding distributions, and derive knowledge regimes by combining change-point detection with inter-year distribution distances. We then extract regime-specific topics via clustering and reconstruct topic lineages by aligning topic similarities to classify inheritance, differentiation, convergence, and disappearance. The analysis delineates three regimes spanning 2005 to 2012, 2013 to 2019, and 2020 to 2024, with pronounced transitions around 2012 to 2013 and 2019 to 2020. Regime 1 centers on foundational technologies such as wireless communication, power, sensors, and reliability. Regime 2 expands toward platforms, apps, and data analytics alongside cross-domain convergence. Regime 3 is characterized by strengthened 5G operations and data-driven services, together with the independent rise in policy, governance, and regulation topics. Transitions reflect recombination built on inherited knowledge rather than abrupt replacement, and post-transition topics display distinct growth typologies by network position and growth pattern. By integrating embedding-based changepoint detection with topic lineage reconstruction, we provide a reproducible account of regime transitions and quantitative evidence to inform the timing of corporate R&D, standard and platform strategies, and policy and regulatory design. Full article

Against the background of carbon neutrality, the government’s carbon regulations have had a profound impact on the distributed energy sharing behavior of industrial parks. To deeply explore the interactive relationship between distributed energy sharing in industrial parks and government regulation, this paper constructs a three-party evolutionary game model composed of the government, core enterprises and supporting enterprises; endogenizes government behavior; and integrates inter-enterprise contractual mechanisms into the evolutionary framework. By establishing a revenue payment matrix and a replication dynamic equation, the stability conditions and system evolution paths of the strategy choices of each subject are analyzed, and numerical simulations are conducted. The results show that there are multiple evolutionary stable equilibria in the system, among which the equilibrium where core enterprises actively share, supporting enterprises actively share, and the government actively regulates carbon is the ideal state. Cost-sharing contracts and cooperative penalty contracts play a significant role in promoting the participation of supporting enterprises in sharing and curbing “free-riding” behavior, respectively. The changes in government subsidy rates and carbon tax rates have a crucial impact on the evolution of corporate strategies. Quantitatively, the carbon tax rate exhibits a threshold effect; enterprises shift to positive energy sharing when the tax rate exceeds 0.8, while a subsidy rate above 0.4 leads the government to withdraw from regulation. This indicates that a reasonable design of carbon regulations can help achieve coordinated energy emission reduction between the government and enterprises. The findings provide theoretical support for optimizing carbon regulations and designing cooperation strategies. Full article

The paper investigates the relationship between ESG transparency/performance and financial performance in tourism, with a focus on profitability (ROA), capital structure (D/E), and cost of capital (WACC). The empirical analysis uses a 2019–2024 panel for 10 listed tourism companies—Booking Holdings, Expedia Group, Airbnb, Marriott International, Hilton Worldwide, Hyatt Hotels, InterContinental Hotels Group, Wyndham Hotels & Resorts, TUI Group, and Carnival Corporation—covering distinct sub-sectors (OTA/Platform, Hotels, Tour Operator, Cruise). The study is based on a quantitative methodology that includes descriptive analyses and the application of advanced econometric models. Methodologically, the paper applies panel econometric models with fixed effects (firm and year), sectoral controls and robustness tests (ESG × Sector interactions, alternative size specifications). The results indicate, on average, a positive association between ESG and profitability (ROA) scores, as well as a negative relationship with WACC (indicating a lower cost of capital for firms with higher ESG), after controlling for size, country and sector. The effects are heterogeneous across sub-sectors, with the ESG–performance relationship more pronounced in hotels (where capital intensity and operational exposure are higher) and less pronounced for OTA platforms, but remain directional and statistically significant in most specifications. Overall, ESG compliance and performance emerge not only as reporting obligations, but also as strategic tools associated with sustainable competitive advantage in tourism. Therefore, the CSRD is not just a reporting obligation, but also a strategic tool that boosts financial performance and managerial innovation. The study provides directions for future research on the use of artificial intelligence in the evaluation of ESG reporting and the expansion of the analysis to other economic branches. Full article

In the context of high-quality urbanization, a key challenge for urban agglomerations is the structural mismatch between economic linkages and rapidly expanding information interactions, which may constrain the performance of coupled systems under shocks. Taking the Beijing–Tianjin–Hebei (BTH) urban agglomeration as a case, we construct an inter-city economic network from cross-city corporate investment ties and an information network from online attention flows, and further derive an economic–information coupled network using a coupling-coordination framework. Using social network analysis and resilience assessment (hierarchy, assortativity, clustering, and disruption simulations), we compare network structures in 2013 and 2023 and evaluate how the structural gap shapes coupled resilience. Results show that (i) economic ties strengthen steadily but moderately, whereas the information network expands faster and becomes more inclusive, widening the structural gap between “virtual” and “material” flows; (ii) despite a persistently high correlation between the two layers, coordination declines, indicating increasing local divergence in link organization; and (iii) resilience improves overall, but differentiation remains: the information network gains robustness through decentralization and redundancy, while the economic network is more sensitive to targeted removals of core nodes, and the coupled network exhibits intermediate performance. These findings suggest that enhancing BTH resilience requires strengthening cross-jurisdictional redundant links and reducing excessive dependence on core corridors to better translate information interactions into balanced economic connectivity. Full article