FedPCL-CDR: A federated prototype-based contrastive learning framework for privacy-preserving cross-domain recommendation.

A multi-dimensional FinTech composite integrating infrastructure, access, usage, knowledge transfer, and governance-by-technology: The role of digital silk road policy in BRI economies

• Efficiently transfers zero-shot capabilities from CLIP to pre-trained vision encoders • Improves performance over previous SOTA work in domain • Training data diversity/coverage improves mapping quality and zero-shot performance • Increased data coverage achieved through the use of tailored loss functions • Training data is entirely unlabelled and unpaired image and text data Foundation models like CLIP demonstrate exceptional capabilities over a broad domain of knowledge, such as with zero-shot classification; however, they also require significant computational resources, narrowing their real-world utility. Recent studies have shown that mapping features from pre-trained vision encoders into CLIP’s latent space can transfer some of CLIP’s abilities to smaller vision encoders, offering a promising alternative. Yet, the performance of these vision encoders still falls short of CLIP’s native capabilities, particularly in low-data regimes. In this work, we argue that enhancing training data coverage/diversity significantly improves mapping efficacy. We achieve this using tailored loss functions rather than relying on data augmentation or increasing training samples. For instance, we exploit the inherent multimodal nature of CLIP’s latent space, by incorporating cycle-consistency loss as one of our loss functions. Moreover, the mapping is learned using entirely unlabelled and unpaired data, eliminating the need for manual labelling or data pairing in novel domains. From these findings, our resulting method (Zoom-shot) offers a viable path to flexible zero-shot models for resource-limited, data-scarce settings. We test Zoom-shot’s zero-shot performance across various pre-trained vision encoders on coarse- and fine-grained datasets and achieve superior performance compared to recent works. In our ablations, we find Zoom-shot allows for a trade-off between data and compute during training; allowing for a significant reduction in required training data. All code and models are available on GitHub.

MuDiS-GDA: Multiscale discriminative graph domain adaptation.

Past stays and future standards: How customer-mediated knowledge transfer enhances service quality

VQIT-GNN: A collaborative knowledge transfer for node-level structure imbalance

Advancing open-set object detection with SAM knowledge transfer and variational feature reconstruction

A vast amount of scientific knowledge is embedded in journal articles as unstructured text, creating challenges for efficiently extracting detailed insights. Traditionally, expert-authored reviews summarize research progress, but they often struggle to capture the intricate synthesis protocols in individual papers and provide limited quantitative comparisons of experimental techniques. Recent advancements in machine learning, particularly natural language processing (NLP), have enabled automated text mining and information extraction. However, in materials science, most approaches have focused on refining model architectures rather than addressing domain-specific challenges such as data annotation and the extraction of complex synthesis details. We present a machine learning framework for extracting synthesis protocols of 2D materials, including graphene and TMDs, from publications spanning 1980-2022. By combining named entity recognition (NER) and extractive question answering (EQA), we retrieve both categorical and numerical synthesis parameters. Generative models are further used to summarize and generate experimental recipes, enabling knowledge transfer across material systems. Our domain-specific, fine-tuned models offer improved precision and interpretability compared to general-purpose approaches. This scalable framework helps unlock hidden insights from literature, supporting data-driven synthesis optimization and accelerating materials discovery.

One-shot Federated Learning (OFL) has emerged as a promising paradigm, enabling global model training with minimal communication overhead. In OFL, the server model is usually distilled from an ensemble of pre-trained client models, while the ensemble also facilitates synthetic data generation for the knowledge distillation process. Prior works show that the performance of the final model is fundamentally tied to both the quality of the synthetic data and the ensemble. However, existing methods often optimize these two components separately, overlooking their interaction. To address this coupled optimization problem and provide a unified solution to the dual challenges of data and model heterogeneity inherent in OFL, we introduce Co-Boosting++, a novel OFL framework where synthetic data generation and ensemble construction mutually enhance each other in an iterative fashion. First, we fix the ensemble and generate hard samples in an adversarial manner. These samples are crucial for enhancing the robustness of knowledge transfer, as they challenge the model to generalize better, thereby improving quality of the synthetic data and subsequent distillation process. Second, leveraging these hard samples, we enhance the ensemble via a Mixture of Experts (MoE) mechanism. MoE allows dynamic adjustment of ensemble weights based on the generated hard samples, which enables the ensemble to better capture diverse and heterogeneous knowledge from client models. Furthermore, we extend Co-Boosting++ to support the simultaneous generation of multiple heterogeneous target models, enabling efficient adaptation to diverse device constraints. Extensive experiments on benchmark datasets demonstrate that Co-Boosting++ consistently outperforms state-of-the-art methods due to its coupled optimization of data and ensemble quality. Additionally, Co-Boosting++ is highly practical in real-world model market scenarios, requiring no local training modifications, additional transmissions, or restrictions on client model architectures. Our code is available at https://github.com/rong-dai/Co-Boosting-PP.

Having a support system that provides novice users with quick and practical assistance for navigating learning management systems (LMS) is essential. This research aims to develop an interactive, video-based online tutorial with engaging elements for LMS usage at the Afyon Kocetepe University and to examine users' experiences. Despite the availability of a help-support structure for navigating the LMS interface, instructors have faced difficulties in adapting to it. To address this, we created a new modular and flexible online tutorial enhanced with multimedia components to assist users in navigating the LMS.The study employed a formative research method, presenting the tutorial to real users for evaluation. The findings revealed that the average usability score of the tutorial was 76.8. Additionally, the average scores for knowledge retention and transfer testscores related to the interactive video content were 73.2. No significant differences were found in usability scores, retention, or transfer test scores based on gender or field of study. In conclusion, the online tutorial, featuring interactive video content, demonstrated usability scores well above average and had a positive impact on both conceptual and procedural learning. These findings will provide valuable insights for the literature and inform future research focused on designing and implementing similar systems.

Alongside the editorial control mechanisms of traditional media, such as fact-checking and source verification, medical journalism ensures the quality of medical knowledge transfer, similar to the peer-review process of professional journals. However, this role, which is central to ademocracy, is increasingly threatened by the rise of the internet and social media. Today, most news from science and medicine reaches citizens as "end consumers" of information directly, i.e., without intermediary quality control according to established journalistic standards. Deteriorating working conditions have forced many medical journalists to move to other occupational fields. The loss of quality control in medical knowledge transfer has become adanger to society. It has never been so difficult to distinguish between reliable information and misinformation or disinformation.

Aiming at the problem that the multi-scale feature interaction ability of the traditional deep learning-based line selection algorithm is insufficient, resulting in the decline of line selection accuracy, a multi-scale feature fusion line selection method based on transfer learning is proposed, abbreviated as TLM-Net. Firstly, to address the issue of the insufficient generalization ability of the line selection network in small-sample scenarios, a simulation data pre-training framework is constructed, and a robust feature representation basis is established through a cross-domain knowledge transfer mechanism. Secondly, aiming at the problem of insufficient extraction of feature information by traditional algorithms, a multi-scale feature fusion network (MFFN) is designed to integrate global context information and local detail features, achieving cross-level semantic complementarity and spatial alignment optimization. Then, to enhance the representation ability of weak fault feature information, an EKA mechanism integrating variable kernel convolution is designed. The background interference is reduced through adaptive multi-region feature focusing, and the edge recognition accuracy of the model for irregular targets is improved. Finally, the pre-trained model is transferred to the target domain by adopting the transfer learning strategy, and the network parameters are fine-tuned in combination with the on-site data to achieve cross-domain adaptation of the feature space. The experimental results show that the TLM-Net algorithm’s mAP@0.5 reaches 98.5%, the accuracy rate and recall rate reach 98.3% and 96.5%, respectively, and the accuracy is improved by 37.5% compared with the original model.

In photovoltaic remote sensing image segmentation tasks, fully supervised methods can achieve high accuracy. However, the high cost of pixel-level annotation significantly limits their scalability in large-scale scenarios. To overcome this annotation bottleneck, this paper proposes a zero-shot cross-modal segmentation framework based on the visual-language pre-trained foundation model (CLIP). This approach harnesses CLIP’s cross-modal knowledge transfer capabilities to achieve precise extraction of photovoltaic targets without requiring any downstream training. This paper first introduces the Layer-wise Augmented Residual Attention (LARA) mechanism to enhance fine-grained detail representation in the feature space. Subsequently, a Cross-modal Semantic Attribution Module (CMSA) is designed to generate precise activation maps by leveraging image-text alignment gradient information. Finally, the Confidence-Aware Refinement Strategy (CARS) replaces the conventional training-based denoising process, directly producing high-quality binary segmentation masks through adaptive thresholding. Comparative experiments were conducted to evaluate the proposed method against various baselines using several public datasets with varying resolutions in Jiangsu Province including Unmanned Aerial Vehicles imagery, Beijing-2, Gaofen-2, and a self-created Sentinel-2 imagery covering multiple countries. Notably, the proposed method achieved an IoU of 70.3% on the Gaofen-2 PV03 dataset with a spatial resolution of approximately 0.3 m and 50.8% on the self-created Sentinel-2 PV_Sentinel-2 dataset with a spatial resolution of 10 m. Experimental results demonstrate that our proposed approach maintains excellent cross-domain generalisation capabilities while reducing annotation costs, thereby providing an efficient and viable technical pathway for the automated monitoring of large-scale photovoltaic facilities.

Virtual communities of practice in health and social care: Early outcomes from a developmental evaluation in Quebec.

Repeat prostate biopsy prediction remains limited by small patient cohorts that constrain artificial intelligence application despite theoretical advantages in capturing complex clinical patterns. This study develops and validates a meta-learning optimized TabNet framework using readily available clinical parameters to overcome sample size constraints and enhance repeat biopsy (RB) prediction accuracy through knowledge transfer from larger initial biopsy (IB) cohorts, with particular applicability to resource-limited settings where mpMRI remains unavailable. Meta-learning enables rapid model adaptation by leveraging knowledge from related tasks with minimal training examples. This retrospective study analyzed 2,087 initial prostate biopsies and 139 subsequent RBs without mpMRI data. A two-stage training paradigm implemented Model-Agnostic Meta-Learning for pre-training on IB data, followed by fine-tuning on the RB cohort. Performance evaluation included discrimination analysis, calibration assessment, and decision curve analysis compared to original TabNet and conventional machine learning approaches, with classification performance benchmarked against established clinical risk calculators. Among 139 RB patients, cancer was detected in 40 cases (28.8%), including 31 clinically significant cancers (75.5%). On the independent testing set of 42 patients, meta-learning TabNet achieved superior discriminative performance (AUROC 0.872) compared to XGBoost (0.808), original TabNet (0.800), and conventional approaches. The model demonstrated optimal calibration (Brier score 0.068, ECE 0.100) and high specificity (90.0%) with only three false positives, substantially outperforming ERSPC and PCPT calculators. Meta-learning optimization successfully addresses sample size limitations in repeat prostate biopsy prediction without requiring advanced imaging. This provides an evidence-based decision support tool enhancing diagnostic accuracy while minimizing unnecessary procedures.

In agriculturally reliant areas, community cooperatives are crucial to rural development and the creation of inclusive, sustainable economic growth. Therefore, as part of the SURE DA2E (Sustainable and Resilient Development in Agriculture through Expansion and Enhancement) extension program, Isabela State University (ISU) established a capacity-building program for community cooperatives through its College of Business, Accountancy, and Public Administration (CBAPA). ISU and the Local Government of San Isidro, Isabela, collaborated on this program on August 19, 2025, to improve community cooperative governance, foster entrepreneurial and financial literacy, and support sustainable livelihood practices. This article highlights ISU's CBAPA's dedication to enabling communities to promote sustainable development through knowledge transfer, multi-sectoral collaboration, and the construction of resilient, community-based economic systems. It also describes the program's objectives, implementation, key activities, participants, and results.

A goal in implementing health research is to change behavior and workplace culture. Realist methods are being used increasingly in healthcare research. While these methods provide context-based pragmatic recommendations, the data can be dense. Thus, making the transfer of knowledge into practice challenging for researchers and clinicians. This paper offers a novel approach to studying complex issues, communicating and motivating behavior change in healthcare using a realist approach to narrative analysis and producing synthesized narratives. A realist lens to research enables researchers to understand what works, with whom, and under what conditions. When used alone, realist methods can result in complex findings that can prove challenging to translate into practice. However, combining a realist approach with narrative analysis can enable a better understanding of the topic and promote practice change. This paper employs a case study on speaking up in healthcare to illustrate this novel method. This case study reports on a longitudinal interview study with a cohort of allied health new graduates to illustrate these methods and discuss the benefits and limitations of their application. We will provide researchers with clear steps to support replication. We argue that this method can aid the implementation of research findings across various contexts within and outside healthcare settings.

Unsupervised Domain Adaptation (UDA) focuses on transferring knowledge from a labeled source domain to an unlabeled target domain, addressing the challenge of domain shift. Significant domain shifts hinder effective knowledge transfer, leading to negative transfer and deteriorating model performance. Therefore, mitigating negative transfer is essential. This study revisits negative transfer through the lens of causally disentangled learning, emphasizing cross-domain discriminative disagreement on non-causal environmental features as a critical factor. Our theoretical analysis reveals that overreliance on non-causal environmental features as the environment evolves can cause discriminative disagreements (termed environmental disagreement), thereby resulting in negative transfer. To address this, we propose Reducing Environmental Disagreement (RED), which disentangles each sample into domain-invariant causal features and domain-specific non-causal environmental features via adversarially training domain-specific environmental feature extractors in the opposite domains. Subsequently, RED estimates and reduces environmental disagreement based on domain specific non-causal environmental features. Experimental results confirm that RED effectively mitigates negative transfer and achieves state-of-the-art performance.

Abstract In constrained multi-objective optimization problems (CMOPs), the discontinuity of the target space and the fragmentation of the feasible solution space caused by complex constraints make the optimization algorithm face irreconcilable conflicts between convergence, diversity, and feasibility. To this end, this paper proposes a dual population co-evolution algorithm based on a dynamic Manhattan-Harmony hybrid distance. The algorithm constructs a main and auxiliary population with complementary structures: the main population focuses on deep search in the feasible domain, the auxiliary population conducts global exploration in the infeasible area, and introduces an evolutionary stage perception mechanism for differentiated environmental selection. In particular, the proposed dynamic Manhattan-Harmony hybrid distance can effectively characterize the convergence and diversity characteristics of individuals and guide the auxiliary population to adopt adaptive selection strategies at different stages. In addition, the algorithm draws on the theory of biological potential energy diffusion and designs a dynamic resource allocation mechanism that combines three types of potential energy: goal orientation, constraint recovery, and structural diversity, to achieve adaptive scheduling of offspring resources. Furthermore, the constructed bidirectional knowledge transfer channel realizes information sharing and co-evolution between the main and auxiliary populations. Experimental results on 33 standard test functions and 12 real-world problems show that HDCMO outperforms many existing representative constrained multi-objective evolutionary algorithms in terms of convergence, feasibility, and distribution balance, and has significant performance advantages and adaptability.

Virtual reality technology has transformed animation education by providing immersive learning environments, yet the proliferation of VR-based teaching resources presents significant challenges in resource discovery and personalized learning design. This research proposes an intelligent recommendation system that integrates transfer learning and knowledge graph technologies to address cold-start and data sparsity challenges in VR animation education. The system comprises two core components: a hybrid recommendation engine combining transfer learning algorithms with knowledge graph reasoning to generate context-aware resource suggestions, and a creativity development path prediction model based on LSTM-attention mechanisms that analyzes learning behaviors and forecasts individualized development trajectories. A comprehensive knowledge graph for VR animation teaching was constructed to capture domain concepts, resource attributes, and pedagogical relationships. Experimental results demonstrate substantial improvements over baseline methods across multiple metrics, with the proposed system achieving optimal performance in precision, recall, F1 score, and NDCG. Deployment in authentic educational settings yielded measurable gains in student learning outcomes and creativity competency development, validating the system's practical effectiveness in enhancing personalized animation education through intelligent resource recommendation and proactive pedagogical interventions.