ABSTRACT Background Inquiry-based practical work is central to science education reform, yet implementation remains challenging in resource-constrained rural contexts. Purpose This study investigated rural primary school science teachers’ perceptions and classroom implementation of inquiry-based practical work, framed by the National Research Council’s essential features of inquiry and the four-level inquiry continuum. Sample Eleven primary school science teachers from a resource-constrained rural district. Design and method Qualitative case study using semi-structured interviews, classroom observations, and document analysis. Data were analysed using deductive thematic analysis. Results A consistent gap emerged between teachers’ positive attitudes toward inquiry and actual practices. While teachers endorsed hands-on, evidence-based learning, few facilitated student-generated inquiries or open-ended investigations. Practical work was predominantly confirmatory and structured, constrained by limited resources, large class sizes, and time pressure. Nevertheless, inquiry-oriented intent was observed through empirical observation and concept demonstration. Conclusion The study demonstrates how inquiry elements manifest in adapted forms within resource-limited contexts and provides contextually grounded guidelines to support authentic inquiry-based practical work.

Advances in monitoring and therapy from minimal residual disease to CAR-T cells in multiple myeloma and precursor disorders.

Simultaneous mitigation of reactive nitrogen and greenhouse gas emissions in livestock systems is a critical challenge for sustainable food production. Here we conduct a meta-analysis of over 3,000 empirical observations to assess the mitigation performance of integrated technologies within livestock farms. Mitigation options are categorized into low-cost and easily scalable technologies and precision-integrated technologies, with the latter performing better by reducing reactive nitrogen and non-CO2 greenhouse gas emissions by two-fifths and one-third, respectively. High-efficacy hotspots emerged in North America, Europe and East and Southeast Asia. Beyond current practices, a transition towards an integrated management framework centred on enclosed manure-treatment systems offers further reductions of over half for reactive nitrogen and two-thirds for non-CO2 greenhouse gas by 2050. Meeting net-zero targets will require sustained capture of almost three-quarters of CO2 from these systems. These findings highlight the critical role of integrated approaches for advancing sustainable and climate-resilient livestock production systems.

Artificial intelligence (AI) systems increasingly mediate how individuals learn, work and make decisions, raising foundational philosophical questions about the nature of knowledge, agency and autonomy. This article integrates philosophical analysis with illustrative empirical cases from Romania to examine how AI restructures human epistemic and practical activity. A central empirical observation, the engagement–performance paradox, reveals that AI-driven learning environments can produce dramatic increases in learner interaction while generating only marginal improvements in understanding. Interpreted through post-phenomenology, virtue epistemology and theories of autonomy, this paradox highlights the emergence of epistemic superficiality: a condition in which algorithmically mediated engagement replaces reflective, conceptually grounded learning. Complementary findings from AI-supported workplace contexts further illustrate how intelligent systems automate aspects of decision-making, thereby reshaping autonomy, responsibility and the phenomenology of action. Synthesizing these insights, the article argues that AI functions as a structuring force that co-authors human agency by reorganizing the conditions under which cognition and action occur. The study contributes to contemporary debates in the philosophy of technology, epistemology and AI ethics by proposing the concept of structured agency as a lens for understanding how AI-mediated environments transform the foundations of knowledge, autonomy and human flourishing.

Abstract This paper analyzes the enclitic =mi in Saraguro Kichwa (a severely endangered language spoken in Saraguro, Ecuador) in matrix declarative clauses in an approach that integrates the broader Quechuan language family. Based on original fieldwork that documents an otherwise undocumented variety of Kichwa, I make three novel empirical observations: (i) I provide evidence suggesting that =mi signals verum, (ii) I show that using =mi is possible to confirm the truth of the scope proposition when following up a sentence with a reportative (but not a direct) evidential, and (iii) I show that =mi surfaces in contrastive (corrective) uses. I make a proposal where =mi is analyzed as a focus marker, which is likened to focus-sensitive items such as only . I further broaden the discussion of =mi to the Quechuan family, showing that integration with prior discourse is the common feature across the family. The discussion bears on general debates of how to best analyze verum and contrast strategies cross-linguistically by introducing a novel strategy instantiated by =mi in that analyzing this element requires the integration of elements of both focus (alternative semantics) and discourse management (sensitivity to the question under discussion).

Digital phantoms are virtual representations of the human body used in medical research to test equipment, train medical professionals and develop or validate algorithms. These models can be created from ‘real-world’ clinical data or from ‘synthetic data’. Phantoms derived from clinical data often serves as ‘ground truth’ reference values anchored in empirical observations. However, there is growing demand for synthetic digital phantoms and datasets that do not originate from real patients, raising critical questions about how reliable knowledge is produced from data detached from reality. This article aims to investigate these issues through a document analysis of peer-reviewed publications on the development and use of digital phantoms in medical physics. We examine how researchers construct ‘ground truth’ and the challenges they encounter when advancing truth claims through technical work. By attending to the bodies fabricated in phantom creation and to the data made to represent human form, we show how synthetic data – detached from real human subjects – are valued for enabling researchers to sidestep the complexities or ‘messiness’ of real-world patients and clinical data. Moreover, we show how synthetic phantoms and data are framed as tools that enhance control and flexibility, functioning as ‘known truths’: workable approximations that enables the construction of what are claimed to be more representative datasets and models. This article contributes to Science and Technology Studies and critical data studies by examining the nature and implications of digital representations and synthetic data in the development of machine-learning models in medicine, and the truth claims they support.

This article develops a research framework for understanding workers and their unions as strategic actors in the climate transition, emphasising the centrality of their agency. While climate change poses profound challenges such as job losses, work reorganisation, and health risks, workers are often described as passive recipients of state or employer-led decisions. We conceptualise union engagement along two dimensions: the strategies they pursue (‘what’) and the factors shaping these strategies (‘why’). Drawing on existing literature and empirical observations, we first identify five generic strategies: abstentionist, unconditionally supportive, conditionally supportive, oppositional, and transformative, mapped along two dimensions: agreement with climate policies and engagement in climate action. Then, we propose a framework integrating internal factors, including worker interests, union power resources, and strategic capabilities, with external influences such as structural power, institutional power, societal discourse, and employer strategies.

Deploying highly generalizable deepfake detection systems on resource-constrained edge devices poses a significant technical challenge for conventional end-to-end large models that rely heavily on computational resources. Extracting multi-source physical prior features is a viable approach under limited computational power; however, in few-shot scenarios, the dimensional mismatch of heterogeneous features is prone to causing downstream classifiers to overfit. To mitigate this bottleneck, this paper proposes a “static feature extraction–central normalization alignment–independent downstream decision” decoupled detection system for few-shot cross-domain tasks on edge devices. The front end of the system constructs an 856-dimensional comprehensive feature reservoir, and a lightweight residual normalization adapter gϕ is introduced as the central support module. This module explicitly compresses the intra-class variance of heterogeneous features, providing a smoothly aligned manifold base for downstream classifiers. Experimental results indicate that this decoupled architecture demonstrates consistent stability in few-shot (K=10) cross-domain evaluations. When encountering intra-family cross-domain shifts and cross-mechanism distribution shifts from diffusion models, the accuracy reaches 84.9% and 76.1%, respectively. Compared to representative end-to-end meta-learning baselines (e.g., MAML), the relative error rate is reduced by over 30%. Furthermore, after completing the asynchronous offline pre-processing (approximately 897 ms) at the front end, a single-image online classification query requires only 7.7 ms under a simulated single-core CPU constraint, satisfying the low-latency requirements for lightweight deployment on edge devices. Finally, combined with empirical observations, this paper discusses the performance boundaries of the architecture in cross-mechanism metric mismatch scenarios, providing a low-barrier, robust engineering defense scheme for resource-constrained environments.

Aberrant glycosylation is a recognized hallmark of cancer, establishing Golgi α-mannosidase II (GMII) as strategic therapeutic target. While the natural alkaloid swainsonine demonstrated potent anticancer activity, its clinical use is hampered by toxicity from off-target inhibition of the lysosomal α-mannosidase (LMan). This review surveys computational methodologies advancing inhibitor development from empirical observations to precision structural optimization. We examine the evolution from Molecular Docking to advanced Quantum Mechanics (QM) and Molecular Dynamics (MD), highlighting their combined role in modeling metalloenzyme flexibility and energetics. Analysis reveals that selectivity relies on exploiting peripheral structural divergences, organelle-specific pH gradients, and distinct substrate conformational itineraries. In this context, electronic structure calculations and pKa predictions prove critical for designing “electrostatic switches”, inhibitors binding neutrally at Golgi pH while incurring lysosomal repulsion. Structurally, targeting the non-conserved “anchor site”, mimicking specific transition-state ring distortions and utilizing conformationally restricted scaffolds represent the most effective strategies. Integrating dynamic sampling with rigorous energetic profiling is therefore crucial for developing the next generation of safe, selective GMII inhibitors.

In the rapidly evolving landscape of artificial intelligence, the relationship between basic arts education and higher art education institutions faces unprecedented challenges and transformations. This research examines the disparities in enrollment standards between fundamental art education programs and higher art colleges in China during the AI era, drawing from extensive professional experience and empirical observations. The study specifically investigates the complex dynamics between traditional art education principles and the emerging influence of specialized art college entrance examination training institutions. Through systematic analysis, this research identifies critical gaps in the current educational framework, particularly focusing on the misalignment between basic art education objectives and the selection criteria employed by higher art institutions. The investigation reveals significant tensions in the supply-demand relationship between foundational art education and professional art colleges, highlighting how these disparities affect student development and educational outcomes. Furthermore, the research examines the structural challenges within China's contemporary art education system, including the impact of commercialized training institutions and their influence on student preparation methods. By critically evaluating these interconnected factors, this study proposes strategic approaches to address the existing educational disparities and suggests innovative solutions to bridge the gap between basic and higher art education standards, ultimately aiming to enhance the coherence and effectiveness of the entire art education pipeline in the AI era.

Simo Tang mitigates mitochondria-dependent apoptosis via PI3K/AKT pathway activation in Chronic atrophic gastritis.

Hydrogen-bonded organic frameworks (HOFs) have recently been highlighted as next-generation structural materials owing to their lightweight nature, mechanical flexibility, and chemical selectivity. However, despite extensive research efforts, the understanding of the structural behavior of nanometer-sized HOFs remains confined to empirical observations. Using molecular dynamics, we uncover how HOF lattices respond mechanically from energy gradients and deformation tests. This highlights that catenation acts as a key source for reduced atomic fluctuations, effective shear redistribution, and emerging auxetic deformation under in-plane loading. In particular, we demonstrate the robustness of HOF substrates modeled after biomolecular exoskeletons, proposing an engineering perspective on computational methodologies for advancing the structural design of porous organic materials.

Echolocating bats operate within a closed sensorimotor loop in which call emission, echo reception, sensory processing, and motor response are linked by finite propagation delays and bounded response times. Although synchrony between wingbeats and call timing is frequently observed, it remains unclear when such coordination is temporally feasible and when it must necessarily break down. Here, I develop a constraint-based framework formalising how temporal feasibility limits shape wingbeat-call coordination during active echolocation. Building on the responsivity framework, the analysis derives explicit conditions under which call emission remains phase-locked to a cyclic motor rhythm, and identifies regimes in which phase locking becomes progressively infeasible as acoustic delay shrinks and call rate rises during prey approach. Simulations across three motor-control configurations-fixed wingbeat frequency and excursion, dynamically adjusted frequency, and dynamically adjusted frequency and excursion-show that transitions from synchrony to asynchrony arise as necessary consequences of delayed feedback and bounded motor dynamics, rather than discrete changes in behavioural strategy. Increasing motor flexibility extends the synchrony-permissive range of call rates but does not eliminate the feasibility boundary. Simulation ensembles spanning biologically plausible parameter combinations confirm that regime transitions are robust and that asynchronous call phases exhibit structured clustering near the feasibility boundary. Empirical observations of transient decoupling during prey pursuit and the terminal buzz are consistent with these predicted transitions. The results identify temporal feasibility as a governing constraint on echolocation behaviour, clarify how apparent closed-loop coordination can arise without tight motor coupling, and generate testable predictions for when and why wingbeat-call synchrony must fail during prey capture.

I derive a novel stochastic equation for the evolution of the additive genetic variance-covariance matrix G in response to mutation, selection, drift, and fluctuating population size. Common wisdom holds that the effect of drift on G is simply to reduce each of its entries by a common proportional amount while preserving its orientation. In contrast, I find that drift causes significant and directional shifts in the orientation of G by driving genetic correlations to their extremes. Biologically, this is a consequence of linkage build-up introduced by drift. I compare these theoretical results to empirical observations based on experiments conducted by Phillips et al. (2001). Additionally, to derive the model of G-matrix evolution, I developed a novel synthetic framework for modelling ecological and evolutionary dynamics of populations carrying multivariate traits. This framework is optimized for deriving new models across a wide range of topics in population biology. Foundations of the framework are formalized by the theory of measure-valued processes, but application of the framework only requires multivariate calculus, and heuristics are presented in the main text for making additional calculations involving stochastic processes. Collectively, this work establishes a powerful framework enabling efficient formal analysis of integrated population processes across evolution and ecology, and its potential for making new discoveries is illustrated by novel findings on fundamental aspects of G-matrix evolution.

Aiming at the problems that movement evaluation in Chinese dance teaching relies on empirical observation, feedback lags and is difficult to quantify, this paper designs a performance movement optimization and teaching feedback system based on motion capture model. The system takes multi-view video as input, combines two-dimensional pose estimation, three-dimensional skeleton recovery, spatio-temporal feature modeling and deviation semantic mapping, and realizes the integrated processing of action recognition, quality assessment and correction suggestion generation. The experimental results show that the recognition accuracy of the proposed method on the self-built dataset reaches 94.82%, and Macro-F1 reaches 94.17%. After the system assisted training, the joint Angle error, trajectory deviation, rhythm deviation and center of gravity stability deviation are significantly decreased. This method can improve the refinement level of Chinese dance performance movement recognition, deviation diagnosis and classroom feedback, and provide an implemensible technical path for Chinese dance digital teaching.

Despite the constitutional guarantees provided by the Indonesian State to recognize and respect the existence of indigenous communities, the implementation of these rights remains problematic. In West Kalimantan, a province rich in both cultural diversity and natural resources, a sharp contradiction exists: while legal frameworks for recognition are expanding, indigenous peoples continue to face systematic marginalization, particularly regarding land tenure and ancestral domain. This study aims to analyze the legal-political paradox in which formal state recognition often serves as a mechanism of exclusion rather than empowerment. It seeks to uncover the underlying factors that cause legal instruments to fail to protect indigenous constitutional rights, against the backdrop of large-scale extractive industries and bureaucratic hurdles. This research employs a qualitative socio-legal approach. Data were gathered through a combination of normative legal research—analyzing constitutional mandates, national laws, and local regulations (Perda)—and empirical observations of land disputes and administrative recognition processes in West Kalimantan. The findings reveal that the “paradox of recognition” is driven by two primary factors: first, overly complex administrative requirements for legal status that transform a fundamental right into a “state-granted” privilege; and second, the dominance of developmentalist agendas that prioritize investment over indigenous sovereignty. Consequently, legal recognition in West Kalimantan often serves as a “formal mask” that stabilizes state authority while indigenous communities remain vulnerable to displacement and criminalization. The study concludes that true constitutional protection requires a paradigm shift from a state-centric recognition model to a rights-based approach that honors the self-identification of indigenous peoples.

Decentralized Federated Learning has emerged as an alternative to centralized architectures due to its faster training, privacy preservation, and reduced communication overhead. In decentralized communication, the server aggregation phase in Centralized Federated Learning shifts to the client side, which means that clients connect with each other in a peer-to-peer manner. However, compared to the centralized mode, data heterogeneity in Decentralized Federated Learning will cause larger variances between aggregated models, which leads to slow convergence in training and poor generalization performance in tests. To address these issues, we introduce Catalyst Acceleration and propose an acceleration Decentralized Federated Learning algorithm called DFedCata. It consists of two main components: the Moreau envelope function, which primarily addresses parameter inconsistencies among clients caused by data heterogeneity, and Nesterov's extrapolation step, which accelerates the aggregation phase. Theoretically, we prove the optimization error bound and generalization error bound of the algorithm, providing a further understanding of the nature of the algorithm and the theoretical perspectives on the hyperparameter choice. Empirically, we demonstrate the advantages of the proposed algorithm in both convergence speed, computational cost, and generalization performance on CIFAR10/100 and Tiny-ImageNet with various non-iid data distributions. Moreover, extensive experiments are conducted to validate the theoretical properties of DFedCata, showing strong consistency between theory and empirical observations. Our code is available at https://github.com/zzylyxx/DFedCata.

The complexity of digital performance indicators in social commerce environments poses significant challenges for small and medium enterprises (SMEs) in formulating coherent and actionable marketing strategies. This study develops and evaluates a forward chaining based expert system to support structured, data driven, and interpretable marketing decision-making. A design science research methodology was employed, encompassing problem identification, artifact development, and evaluation. Knowledge was elicited through literature synthesis, expert consultation, and empirical observation, and subsequently formalized into IF–THEN production rules within a structured knowledge base. The system applies a forward chaining inference mechanism to process key indicators, including followers, engagement rate, promotion frequency, and conversion rate, in order to generate prioritized strategic recommendations. Evaluation was conducted using scenario-based testing and expert validation to assess accuracy, consistency, and contextual appropriateness. The results demonstrate complete alignment between system outputs and expert judgment across all evaluation scenarios, indicating high reliability and logical consistency of the rule-based reasoning process. The system also produces context-sensitive and interpretable recommendations aligned with varying levels of business performance. This study contributes by advancing rule-based decision support systems in social commerce and providing an explainable and practically applicable tool to enhance marketing decision quality among SMEs.

Background: In contemporary information environments, labels such as “conspiracy theory,” “misinformation,” and “extremism” increasingly shape whether contested claims receive evidentiary evaluation or are excluded before inquiry begins. Yet Christian responses often oscillate between institutional naivety, which equates authorized narratives with truth, and conspiratorial credulity, which treats suppression as proof. Aim: This article develops a Christian social epistemology for evaluating stigmatized, suppressed, and institutionally discredited knowledge claims with courage, restraint, and fidelity to biblical truth. Method: Using a secondary-data conceptual design, the study integrates social epistemology, sociology of knowledge, epistemic injustice, conspiracy theory studies, propaganda and misinformation research, critical realism, and biblical theology. It does not adjudicate any specific controversial claim. Instead, it constructs a normative framework for responsible inquiry. Findings: The article argues that “conspiracy theory” functions not only as an epistemic category but also as a social and institutional label that can discipline discourse, protect legitimate standards, or prematurely foreclose investigation. It proposes a ten-level layered discernment model moving from empirical observation, epistemic classification, institutional mechanism, historical genealogy, discursive framing, power analysis, rival explanations, and disconfirmation testing to biblical-theological discernment and ethical response. The framework distinguishes verified evidence, documented patterns, plausible inference, contested interpretation, and theological discernment. Contribution: The article contributes to theology, social epistemology, and public discourse by offering an academically rigorous and biblically governed alternative to both uncritical institutional trust and reckless suspicion. It contends that Christian scholars, churches, and public intellectuals must test contested claims through evidence, humility, methodological transparency, and moral accountability, refusing both fear-driven speculation and passive conformity to managed narratives.

The task of dynamic graph link prediction is to forecast the evolution of complex systems. Empirical observations reveal that interactions within these systems exhibit an Entangled Spatio-Temporal Pattern, which manifests through three interrelated phenomena, namely Latent High-Order Bridges, Multi-Frequency Temporal Dynamics, and Spatio-Temporal Entanglement, with stronger structural ties facilitating tolerance for longer temporal gaps. However, limited by computationally prohibitive multi-hop sampling or inefficient long-sequence modeling, existing methods struggle to capture this complex pattern. Inspired by State-Space Models (SSMs) like Mamba for efficient long-range modeling yet aiming to address their native agnosticism to structural and multi-frequency dynamics, we propose a framework named DyGHydra, which couples a tailored Continuous-Time Hierarchical Mamba (CT-HMamba) backbone with a multi-hop structural encoder. The framework first employs the multi-hop structural encoder to reveal latent high-order interactions, extracting interaction-level cross-hop features. Subsequently, the CT-HMamba backbone utilizes these features to address multi-frequency dynamics through a hierarchical architecture, decomposing interaction history to simultaneously model high-frequency bursts and long-term trends. To capture the spatio-temporal entanglement, CT-HMamba further tailors its core state-space mechanism to be co-driven by physical time and structural context. Specifically, physical time governs the state transition decay to reflect temporal forgetting, while structural context modulates the input-output projections to prioritize topologically significant events. Extensive experiments on eleven real-world datasets show that DyGHydra achieves state-of-the-art performance across most settings for both transductive and inductive link prediction, validating its effectiveness in modeling complex temporal dynamics with superior efficiency.