Search Results (367)

The accelerating integration of electric vehicles (EVs) into contemporary transportation infrastructure has underscored significant limitations in traditional charging paradigms, particularly in accommodating heterogeneous user requirements within dynamic operational environments. This study presents a differentiated optimization framework for EV charging strategies through the systematic classification of user types. A multidimensional decision-making environment is established for three representative user categories—residential, commercial, and industrial—by synthesizing time-variant electricity pricing models with dynamic carbon emission pricing mechanisms. A bi-level optimization architecture is subsequently formulated, leveraging deep reinforcement learning (DRL) to capture user-specific demand characteristics through customized reward functions and adaptive constraint structures. Validation is conducted within a high-fidelity simulation environment featuring 90 autonomous EV charging agents operating in a metropolitan parking facility. Empirical results indicate that the proposed typology-driven approach yields a 32.6% average cost reduction across user groups relative to baseline charging protocols, with statistically significant improvements in expenditure optimization (p < 0.01). Further interpretability analysis employing gradient-weighted class activation mapping (Grad-CAM) demonstrates that the model’s attention mechanisms are well aligned with theoretically anticipated demand prioritization patterns across the distinct user types, thereby confirming the decision-theoretic soundness of the framework. Full article

Direct preference optimization (DPO) has been successfully applied to align large language models (LLMs) with human preferences. In recent years, DPO has also been used to improve the generation quality of text-to-image diffusion models. However, existing techniques often rely on a single type of reward model. They are also prone to overfitting to inaccurate reward signals. As a result, model quality cannot be continuously improved. To address these limitations, we propose xDPO. This method introduces a novel regularization approach that implicitly defines reward functions for both preferred and non-preferred samples. This design greatly enhances the flexibility of reward modeling. The experimental results show that, after fine-tuning Stable Diffusion v1.5, xDPO achieves significant improvements in human preference evaluations compared to previous DPO methods. It also improves training efficiency by approximately 1.5 times. Meanwhile, xDPO maintains image–text alignment performance that is comparable to the original model. Full article

Background/Objectives: Contextual memory associated with methamphetamine (METH) use contributes to relapse and persistence of addiction. Angiotensin II type 1 receptor (AT1R) signaling has been implicated in drug reinforcement. LCZ696, a clinically used combination of sacubitril (a neprilysin inhibitor) and valsartan (an AT1R antagonist), may interfere with METH-associated memory through the modulation of dopaminergic pathways. Methods: Male C57BL/6J mice were tested in a conditioned place preference (CPP) paradigm to assess the effects of LCZ696, sacubitril (AHU377), and valsartan on METH-induced memory expression and reinstatement. Synaptic plasticity in the nucleus accumbens (NAc) was examined by assessing the levels of synaptophysin (Syp) and postsynaptic density protein 95 (Psd95), as well as dendritic spine density. Dopaminergic signaling in the ventral tegmental area (VTA) was evaluated via ELISA, Western blotting, and chromatin immunoprecipitation (ChIP), targeting cAMP response element-binding protein (Creb) binding to the tyrosine hydroxylase (Th) promoter. To further assess the role of Th, an adeno-associated virus (AAV9) carrying a CRISPR-Cas9-based sgRNA targeting Th (AAV9-Th-sgRNA) was microinjected into the VTA. Results: LCZ696 and valsartan significantly reduced METH-induced CPP and reinstatement. LCZ696 reversed METH-induced synaptic and dopaminergic alterations and suppressed Creb-mediated Th transcription. Th knockdown attenuated both CPP acquisition and relapse. Conclusions: LCZ696 disrupts METH-associated contextual memory by modulating dopaminergic signaling and Creb-dependent Th expression, supporting its potential as a treatment for METH use disorder. Full article

Sweet sensing is a fundamental sensory experience that plays a critical role not only in food preference, reward and dietary behaviour but also in glucose metabolism. Sweet taste receptors (STRs), composed of a heterodimer of taste receptor type 1 member 2 (T1R2) and member 3 (T1R3), are now recognised as being widely distributed throughout the body, including the gastrointestinal tract. Preclinical studies suggest these receptors are central to nutrient and glucose sensing, detecting energy availability and triggering metabolic and behavioural responses to maintain energy balance. Both internal and external factors tightly regulate their signalling pathways, and dysfunction within these systems may contribute to the development of metabolic disorders such as obesity and type 2 diabetes (T2D). Functional magnetic resonance imaging (fMRI) has provided valuable insights into the neural mechanisms underlying sweet sensing by mapping brain responses to both lingual/oral and gastrointestinal sweet stimuli. This review highlights key findings from fMRI studies and explores how these neural responses are modulated by metabolic state and individual characteristics such as body mass index, habitual intake and metabolic health. By integrating current evidence, this review advances our understanding of the complex interplay between sweet sensing, brain responses, and health and identifies key gaps and directions for future research in nutritional neuroscience. Full article

Federated Learning (FL), a key paradigm in privacy-preserving and distributed machine learning (ML), enables collaborative model training across decentralized data sources without requiring raw data exchange. FL enables collaborative model training across decentralized data sources while preserving privacy. However, selecting appropriate clients remains a major challenge, especially in heterogeneous environments with diverse battery levels, privacy needs, and learning capacities. In this work, a centralized reward-based payoff strategy (RBPS) with cooperative intent is proposed for client selection. In RBPS, each client evaluates participation based on locally measured battery level, privacy requirement, and the model’s accuracy in the current round computing a payoff from these factors and electing to participate if the payoff exceeds a predefined threshold. Participating clients then receive the updated global model. By jointly optimizing model accuracy, privacy preservation, and battery-level constraints, RBPS realizes a multi-objective selection mechanism. Under realistic simulations of client heterogeneity, RBPS yields more robust and efficient training compared to existing methods, confirming its suitability for deployment in resource-constrained FL settings. Experimental analysis demonstrates that RBPS offers significant advantages over state-of-the-art (SOA) client selection methods, particularly those relying on a single selection criterion such as accuracy, battery, or privacy alone. These one-dimensional approaches often lead to trade-offs where improvements in one aspect come at the cost of another. In contrast, RBPS leverages client heterogeneity not as a limitation, but as a strategic asset to maintain and balance all critical characteristics simultaneously. Rather than optimizing performance for a single device type or constraint, RBPS benefits from the diversity of heterogeneous clients, enabling improved accuracy, energy preservation, and privacy protection all at once. This is achieved by dynamically adapting the selection strategy to the strengths of different client profiles. Unlike homogeneous environments, where only one capability tends to dominate, RBPS ensures that no key property is sacrificed. RBPS thus aligns more closely with real-world FL deployments, where mixed-device participation is common and balanced optimization is essential. Full article

Cognitive abilities in fish have been widely demonstrated using experimental protocols commonly adopted with mammals and birds. Only a few studies have tested fish in the simultaneous match-to-sample task (sMTS), and mixed evidence regarding their capacity to solve the task has been reported. Here we investigated whether guppies (Poecilia reticulata) could discriminate stimuli based on their sameness in the sMTS where fish presented with a sample stimulus had to choose which of two simultaneously presented comparison stimuli matched it. We also assessed how performance was influenced by the training set size and stimulus type. Three experiments were conducted using three different sets of stimuli: two colors (red and green), two geometric shapes (circle vs. triangle); and multiple shapes. Performance was analyzed using binomial tests, t-tests, and generalized linear mixed models. The results showed that guppies learned to select the rewarding stimulus in a relatively limited number of trials and were successful in all experiments. Although no effect of the training set size was observed, guppies were more accurate when multiple stimuli were used. These findings support previous evidence suggesting that multiple training stimuli may improve generalization abilities and set the basis for future studies that adopt a delayed version of the task. Full article

The increasing need for sustainable energy management in smart buildings calls for cost-effective solutions that balance energy efficiency and occupant comfort. This article presents a Large Language Model (LLM)-based recommendation system capable of generating proactive, context-aware suggestions from dynamic building conditions. The system was trained on a combination of real-world data and Sinergym simulations, capturing inputs such as weather conditions, forecasts, energy usage, electricity prices, and detailed zone parameters. Five models were fine-tuned and evaluated: GPT-2-Small, GPT-2-Medium, DeepSeek-R1-Distill-Qwen-1.5B, DeepSeek-R1-Distill-Qwen-7B, and GPT-4. To enhance evaluation precision, a novel metric, the Zone-Aware Semantic Reward (ZASR), was developed, combining Sentence-BERT with zone-level scoring and complemented by F1-Score metrics. While GPT-4 demonstrated strong performance with minimal data, its high inference cost limits scalability. In contrast, open-access models like DeepSeek-R1-Distill-Qwen-1.5B, DeepSeek-R1-Distill-Qwen-7B, and GPT-2-Medium required larger datasets but matched or exceeded GPT-4’s performance at significantly lower cost. The system demonstrated adaptability across diverse building types, supported by heterogeneous datasets and parameter normalization. Importantly, the system was also deployed in a real-world multi-zone residential building in Thessaloniki, Greece. During a two-week operational period under near-identical weather and occupancy conditions, the model-assisted recommendations contributed to an estimated 10% reduction in electricity consumption, showcasing the practical potential of LLM-based recommendations in live building environments. Full article

Background/Objectives: Glucagon-like peptide-1 receptor agonists (GLP-1RAs), which were originally developed for managing type 2 diabetes by enhancing insulin secretion and reducing appetite, have emerged as promising candidates in alcohol use disorder (AUD). These medications offer a dual mechanism of action that aligns with the multifaceted nature of addiction by targeting both peripheral metabolic and central reward pathways. This review focused on the current clinical trials and real-world evidence regarding the effects of GLP-1RAs as novel therapeutics for AUD. We also discussed early but encouraging results from clinical trials in AUD, observational and real-world evidence, safety profiles, psychiatric considerations, and future directions leading beyond GLP-1RAs. Methods: A comprehensive English-language literature search was conducted per PRISMA guidelines across PubMed, Medline, Google Scholar, Web of Science, and trial registries. Using targeted keywords, we identified relevant clinical and observational studies on GLP-1RAs for alcohol use disorder, excluding off-topic or non-English works and assessing all studies for eligibility. Results: Out of 1080 records identified, seven studies met the inclusion criteria. The findings from recent clinical trials, large-scale observational studies, and real-world evidence suggest that GLP-1RAs may significantly reduce alcohol consumption, cravings, and alcohol-related hospitalizations. Their central effect on reward processing, coupled with a generally favorable safety profile, supports their potential therapeutic role beyond metabolic disorders. Conclusions: Emerging evidence positions GLP-1RAs as a promising new pharmacologic approach for managing AUD. Ongoing and future research should prioritize larger, longer-duration randomized controlled trials that include diverse populations, with specific attention to treatment motivation, co-occurring psychiatric conditions, and long-term outcomes. Full article

In practice, the new generation of knowledge-based employees often exhibits a “lying flat” attitude. This reflects the failure of organizational incentive mechanisms. In order to improve the incentive system and encourage employees to be proactive, the study explores and compares the synergistic effects of different rewards tools on various forms of proactive behavior in the new generation of knowledge employees. After conducting fsQCA and NCA analyses on paired data from 93 leaders and 210 employees based on the ERG theory, the findings indicate that no single reward tool is a necessary condition for triggering high proactive behavior. Instead, different reward tools need to work in synergy to produce effective motivation. Three patterns drive employees to exhibit high individual task proactivity. They are the “Dual-Drive Salary Security and Moderate Labor Dominant” pattern, the “Moderate Labor Dominant” pattern, and the “Salary Security Dominant” pattern. Two patterns drive employees to demonstrate high team member proactivity, namely the “Employee Care Dominant High-Investment” pattern and the “Pay Fairness Dominant High-Investment” pattern. Additionally, good work experience (i.e., colleague relationships) in the workplace has a significant impact on both types of proactive behavior. The research conclusions will provide insights and references for enterprise managers to design more targeted compensation incentive policies and unleash the vitality of the new generation of knowledgeable employees. Full article

Background: Previous syntheses on the neural effects of alcohol have been restricted to tasks assessing craving, cognitive control, and reward processing. Despite extensive research, a comprehensive synthesis of functional magnetic resonance imaging (fMRI) findings on alcohol use disorder (AUD) remains lacking. This study aimed to identify consistent brain activation alterations across all cognitive and emotional tasks administered to individuals with AUD while distinguishing between short-term and long-term abstinence and using activation likelihood estimation meta-analysis. Sub-analyses on task types were performed. Methods: A systematic review identified 67 fMRI studies on participants with an AUD. Results: The meta-analysis revealed significant alterations in brain activity, including both hypo- and hyperactivation in the left putamen across all AUD participants. These alterations were observed more frequently during decision-making and reward tasks. Short-term abstinent individuals exhibited hypoactivation in the right middle frontal gyrus (MFG), corresponding to the dorsolateral prefrontal cortex. In contrast, long-term abstinent individuals displayed hypoactivation in the right superior frontal gyrus (SFG) and dorsal anterior cingulate cortex (dACC). This meta-analysis highlights critical neural alterations in AUD, particularly in regions associated with reward processing (putamen), executive functions (MFG and SFG), and attentional salience (dACC). Putamen changes were predominantly observed during short-term abstinence and in decision-making, as well as reward processing tasks. dACC and SFG hypoactivation were specific to long-term abstinence, while MFG hypoactivation was specific to short-term abstinence. Conclusions: These findings support prior research indicating a motivational imbalance and persistent executive dysfunctions in AUD. Standardizing consumption metrics and expanding task diversity in future research is essential to further refine our understanding of the neural effects of AUD. Full article

Parents provide most of the support needed for children with cerebral palsy (CP) to increase the child’s participation and independence. Understanding the experiences of parents caring for children with CP is essential for developing effective family programs and services. The current knowledge about parents’ experiences in CP is based on studies in Western countries, with little known about this phenomenon in Arab countries like Saudi Arabia. This study aimed to understand the unique experiences and support needs of Saudi parents caring for children with CP from a social-ecological perspective. We conducted a qualitative, exploratory, descriptive study involving 12 semi-structured interviews with mothers and fathers of children with different types of CP. We analyzed the data using a reflexive thematic approach, following six distinct phases. Participants’ narratives revealed a complex caregiving journey marked by both challenges and rewards. Support from Saudi nuclear and extended family members was considered important; however, many parents expressed a need for additional physical and financial assistance from their families. Parents reported feeling stressed and experiencing challenges in accessing and navigating educational and healthcare services. Our findings highlight that Islamic values play a crucial role in the experiences of Saudi parents. These values foster a sense of collectivism, highlighting the importance of family support and community involvement, which can affect the Saudi caregiving environment. Parents remain an essential yet often invisible part of the Saudi caregiving system. Without adequate support, parents are at risk of experiencing social, financial, academic, physical, and mental health challenges, which may affect their overall family well-being. Future work may need to consider spiritual and gender roles when developing programs or services to support Saudi parents of children with CP. Full article

Children acquire much of their knowledge through trusting others’ testimony, particularly that of teachers. They not only tend to trust their teachers but also imitate behaviors that teachers reward. However, it remains unclear if they show selective trust in those who provide such rewards. This study, therefore, examined how teachers’ rewards to other children and the types of these rewards influence the selective trust of preschoolers. In Study 1, 162 preschoolers from junior, middle, and senior classes watched videos of a teacher giving verbal and material rewards, while another provided neutral feedback. Then, children chose which teacher to trust in a novel object-naming task. The results showed that all preschoolers preferred to trust teachers who offered rewards compared to those who did not. Moreover, junior-class children displayed the highest level of selective trust among the preschoolers. In Study 2, 176 preschoolers judged which teacher to trust, one offering material rewards and the other verbal praise. The results showed senior-class girls preferred teachers with material rewards more than senior-class boys and middle-class girls. These findings indicate that preschoolers can assess teachers’ trustworthiness based on rewards and are more sensitive to material rewards than to verbal praise when accepting information from teachers. Full article

Crowdshipping establishes a short-term connection between shippers and individual carriers, bridging the service requirements in last-mile logistics. From the perspective of a carrier operating multiple vehicles, this study considers the challenge of maximizing profits by optimizing bid strategies for delivery prices and transportation conditions in the context of bid-based crowdshipping services. We considered two types of bid strategies: a price bid that adjusts the RFQ freight charge and a multi-attribute bid that scores both price and service quality. We formulated the problem as a Markov decision process (MDP) to represent uncertain and sequential decision-making procedures. Furthermore, given the complexity of the newly proposed problem, which involves multiple vehicles, route optimizations, and multiple attributes of bids, we employed a reinforcement learning (RL) approach that learns an optimal bid strategy. Finally, numerical experiments are conducted to illustrate the superiority of the bid strategy learned by RL and to analyze the behavior of the bid strategy. A numerical analysis shows that the bid strategies learned by RL provide more rewards and lower costs than other benchmark strategies. In addition, a comparison of price-based and multi-attribute strategies reveals that the choice of appropriate strategies is situation-dependent. Full article

Effective teacher training is critical for improving student outcomes. Behavior Skills Training (BST), which consists of instruction, modeling, role play, and feedback, has been demonstrated to be effective for training teachers, staff, and peers to deliver a wide range of interventions to individuals with disabilities. This study examined the effects of using BST to train a teacher to implement a Universal Protocol treatment package. The Universal Protocol package included procedures for building rapport, identifying the potential triggers of challenging behavior, and making data-based decisions on the type and intensity of services required. A multiple-baseline design across school settings (e.g., reward, work, transition) demonstrated a functional relation of BST with the correct implementation of the Universal Protocol. Student problem behaviors appeared to be sensitive to teacher implementation of the Universal Protocol. Limitations and future directions are presented. Full article

Background: Sodium-glucose cotransporter-2 inhibitors (SGLT2i) are increasingly used in patients with type 2 diabetes, chronic kidney disease, and heart failure. However, their safety and efficacy in patients with congenital or surgically altered urogenital anatomy remains underexplored. Methods: We conducted a narrative review of current evidence regarding the use of SGLT2i in patients with urinary tract malformations, urinary diversions, and functional voiding disorders. Key risks, clinical considerations, and management strategies were synthesized from the existing literature and case reports. Results: Patients with benign prostatic hyperplasia, vesicoureteral reflux, neurogenic bladder, nephrostomies, and ileal conduits may face increased risks of urinary tract infections, fungal colonization, and therapy-related complications due to persistent glycosuria and altered urinary flow. Nevertheless, these patients may still benefit from SGLT2i’s systemic renal and cardiovascular effects. Individualized risk assessment, close monitoring, and multidisciplinary management are essential. Conclusions: Patients with urological abnormalities represent a high-risk but potentially high-reward population for SGLT2i therapy. A cautious, tailored approach is necessary, and future dedicated research is urgently needed to better guide clinical practice. Full article