Core Challenges Research Articles

This research addresses the core challenge in Hindu Religious Education: students’ low cognitive and reflective engagement in studying the Vedic Scriptures, caused by teaching methods focused largely on memorization rather than deep understanding. The main argument advanced is that implementing a Problem-Based Learning (PBL) model directly stimulates deep learning in Vedic studies, shifting the educational focus from rote memorization to analytical thinking, reflective understanding, and the integration of spiritual values. Using a descriptive qualitative approach combining participatory observation, interviews with teachers and students, and analysis of learning documents. The study analyzes how PBL promotes the construction of knowledge and transforms learning experiences. Findings demonstrate that PBL facilitates analytical thinking, encourages spiritual reflection, strengthens the ability to contextualize Vedic teachings, and fosters meaningful collaboration. By highlighting the effectiveness of PBL, this research substantiates the need for pedagogical innovations in Hindu Religious Education to produce holistic, morally, and spiritually aware students, thereby meeting the demands of 21st-century learning.

With the deep integration of digital technology and the real economy, AI auditing has emerged as a core paradigm that breaks through the pain points of traditional auditing, such as “sampling limitations, post-event lag, and reliance on manual labor”. This paper systematically reviews the theoretical connotations of AI auditing, reveals its current practical status, deeply analyzes four core challenges: data quality, ethical compliance, talent adaptation, and institutional synergy, and proposes feasible development paths from four dimensions: technological optimization, institutional construction, talent cultivation, and industry synergy. The research indicates that AI auditing needs to be “based on data elements, driven by technological innovation, with institutional guarantees as the bottom line, and talent adaptation as the core”, and achieve an upgrade from “tool assistance” to “governance synergy” under the promotion of new productive forces.

With the rapid development of global maritime trade and the rising demand for real-time, accurate marine ship monitoring, satellite image-based ship detection has become crucial for marine management and national defense. However, it faces two core challenges: complex backgrounds in high-resolution marine remote sensing images, and great variations in ship sizes–especially difficult small ship extraction. To address these, this study proposes an enhanced method based on improved YOLOv13, using China’s Gaofen-2 (GF-2) satellite images. First, GF-2 image data is preprocessed, including radiometric correction to eliminate atmospheric effects, orthorectification to correct image distortion, and fusion of multispectral and panchromatic images to improve spatial resolution and enrich spectral information. Then, three key optimizations are made to the YOLOv13 model: 1) In the backbone network, the A2C2f module is modified by introducing a single-head attention mechanism. By parallelly fusing global and local feature information, it avoids multi-head redundancy and improves the recognition accuracy of small ship targets; 2) In both the backbone and neck networks, the DS_C3K2 module is modified by integrating a lightweight attention mechanism, which enhances the model’s feature extraction capability in complex backgrounds while reducing channel and spatial redundancy; 3) In the head network, a path-fused Global Feature Pyramid Network (GFPN) is introduced, which leverages skip-layer and cross-scale connections to strengthen cross-scale feature interaction, refine the representation of small ship features, and effectively address the issues of insufficient deep supervision and feature information loss in multi-scale ship detection. Additionally, the improved YOLOv13 model is pre-trained using the open-source DOTA dataset (rich in non-ship negative samples) to enhance its ability to distinguish between ship foreground and background clutter, and then applied to ship detection in segmented sub-images of GF-2 remote sensing images; finally, the detected sub-images are stitched to restore complete regional images. Experiments show that the accuracy rate reaches 96.9%, the recall rate reaches 91.4%, the mAP50 reaches 95.5%, and the mAP50-95 reaches 75.9%, all of which are higher than the mainstream target detection models. It provides a high-performance solution for complex marine ship detection and has important practical significance for both civilian and military fields.

Abstract Generative AI (GenAI) systems raise fundamental challenges for copyright law at both the input and output stages. On the input side, legal uncertainty surrounds the large-scale scraping of copyrighted data for model training, with divergent rules across jurisdictions and limited transparency on how data is sourced. On the output side, courts struggle to determine when AI-assisted creations are sufficiently human to merit protection, leading to inconsistent or unclear legal outcomes. This paper outlines the “AI copyright conundrum” and examines its impact on the incentives to create, the accessibility of high-quality datasets, and the sustainability of cultural production. We discuss policy options and open questions for research.

Antibiotic resistance is a major threat to global public health. It is vital to understand the mechanism of antibiotic resistance development to prevent the emergence of new pan-resistant pathogenic bacteria and to develop new antibiotics. Measuring the differences in proteins among single bacterial cells can aid in identifying antibiotic resistance and antibiotic susceptibility due to their regulatory roles in bacterial physiology and homeostasis. Although single-cell proteomics has been successful in mammalian systems, attaining comparable performance in bacteria remains challenging due to the extremely limited proteome content of a single bacterial cell. This review discusses the role of proteomics analysis in determining antibiotic resistance and the various mass spectrometry-based strategies that have been successful in detecting protein biomarkers for antibiotic resistance from bulk proteomics analysis. It highlights the core challenges of bacterial single-cell proteomics in contrast to mammalian systems, explores emerging technologies, and the proteomes beyond the cells in studying antibiotic resistance development and antibiotic susceptibility testing.

Tar-rich coal (with a tar yield ≥ 7%), as a special coal-based oil and gas resource, is of great significance for ensuring national energy security and promoting the clean conversion of coal. The selection of suitable geological sites represents a core challenge for the safe and efficient application of its in situ pyrolysis technology. Focusing on the tar-rich coal seams in the Santanghu Basin, this study constructed a comprehensive geological evaluation system for site selection by integrating numerical simulation, data mining, and laboratory experiments. The Analytic Hierarchy Process (AHP) and a fuzzy comprehensive evaluation method were employed to achieve a quantitative assessment and identify favorable areas within the study region. The results indicate that resource scale, coal seam conditions, and the properties of the roof and floor strata are the key controlling factors. One optimally comprehensive Class I favorable area (Tiao IV block) was successfully identified. This block exhibits a large resource scale, favorable coal seam conditions, a high tar yield, excellent geological sealing, and superior engineering compatibility, making it the recommended priority target for pilot testing. The evaluation system developed in this study can provide a theoretical basis and technical reference for the geological site selection of in situ pyrolysis of tar-rich coal in similar mining areas and advance its industrialization.

The coordinated development of Production-Living-Ecological (PLE) spaces has emerged as a core challenge for regional sustainability amid rapid urbanization processes. This study examines the Guanzhong Plain Urban Agglomeration (2001–2021) using an integrated Markov-PLUS model coupled with Random Forest algorithms and 17 driving factors to construct 4 policy scenarios for future projections. The results reveal dramatic spatial restructuring: living space expanded 73.89% while production and ecological spaces contracted 7.47% and 8.94%. Evolution occurred through four distinct phases—rapid expansion, structural adjustment, quality improvement, and green transformation—each corresponding to national policy transitions with regional lags. Driving mechanism analysis identified environmental factors contributing 45–55% of variance, population density driving 24.2% of living space expansion, and elevation thresholds constraining urban growth above 1000 m. Multi-scenario simulations revealed fundamental trade-offs: urban development scenarios achieved 55.34% built-up expansion but sacrificed 15.4% ecological space, while ecological protection scenarios maintained 92% food production capacity with optimal connectivity (0.63) and maximum carbon storage (1287 Mt C). Model validation achieved exceptional accuracy (Kappa = 0.91, FoM = 0.24). This research emphasizes three strategic imperatives: (1) differentiated spatial governance (urban priority in cores, farmland protection in plains, ecological restoration in mountains); (2) temporal coordination mechanisms accounting for 3–5-year policy transmission lags; (3) adaptive management approaches addressing nonlinear evolution characteristics. This framework provides scientific foundations for balancing economic development, food security, and ecological protection in rapidly urbanizing regions.

In the current context, as robotic technology penetrates deeper into more complex scenarios, autonomous decision-making capability has become a core indicator for evaluating robot performance. The multimodal perception system, as the central hub for robots to acquire environmental information and understand task scenarios, this paper explores its applications in different fields and analyzes the existing challenges. Combining specific cases from the past five years, it systematically analyzes the application modes of multimodal perception systems in three typical robotic autonomous decision-making fields: agriculture, bionics, and medical care. It concludes that the current multimodal perception systems face core challenges at both technical and safety-ethical levels. Finally, it summarizes the research limitations of this paper in terms of extreme environments and lightweight design, and proposes future development directions. The findings reveal that advancing multimodal perception systems is essential for enhancing autonomous decision-making in robots, yet addressing the identified technical and safety-ethical challenges will be crucial for successful implementation across diverse applications.

Nanoscratch is a key experimental method for evaluating material mechanical properties, studying material removal mechanisms, and revealing surface damage formation mechanisms at the nanoscale. In the semiconductor, microelectronics, and optoelectronics industries, nanoscratches on the surface of InP substrates are considered fatal defects that must be eliminated, as they can seriously affect the performance and reliability of devices. However, in the chemical mechanical polishing (CMP) process of InP, the formation of nanoscratches is difficult to avoid, and precise control is the core challenge to achieve atomic level smooth surfaces. To address this challenge, this study systematically investigated the effects of three typical scratch directions (Face-forward (FF); Edge-forward (EF); Side-face-forward (SFF)) of the Berkovich indenter on the single crystal InP(001) substrate during the nanofabrication process using molecular dynamics simulation (MD) and nanoscratch experiments. The research focuses on analyzing the evolution of surface morphology, scratch force response, and the material removal mechanism. The results indicate that the material removal mechanism of InP in the nanoscratch process is characterized by the coexistence of elastic and plastic deformation, and the damage is mainly manifested by the initiation and propagation of transverse microcracks and the accumulation of sheet-like chips. Compared with the EF direction, the FF direction can significantly reduce the lateral extension length of the groove by about 52%-55%; under the condition of small rake angle, the material removal rate is increased by about 8%-17% compared to the condition of large rake angle. The plastic transformation rate is highest in the EF direction with an average of 31%. The experimental results are highly consistent with the simulation results: the atomic displacement map simulated by MD and the SEM characterization of the experiment both show that the morphology and distribution characteristics of the chips are consistent under different scratch directions. The mechanism revealed in this study provides important theoretical basis and process guidance for effectively suppressing nanoscratch defects in the InP CMP process.

The aviation industry operates as a complex, dynamic system generating vast volumes of data from aircraft sensors, flight schedules, and external sources. Managing this data is critical for mitigating disruptive and costly events such as mechanical failures and flight delays. This paper presents a comprehensive application of predictive analytics and machine learning to enhance aviation safety and operational efficiency. We address two core challenges: predictive maintenance of aircraft engines and forecasting flight delays. For maintenance, we utilise NASA’s C-MAPSS simulation dataset to develop and compare models, including one-dimensional convolutional neural networks (1D CNNs) and long short-term memory networks (LSTMs), for classifying engine health status and predicting the Remaining Useful Life (RUL), achieving classification accuracy up to 97%. For operational efficiency, we analyse historical flight data to build regression models for predicting departure delays, identifying key contributing factors such as airline, origin airport, and scheduled time. Our methodology highlights the critical role of Exploratory Data Analysis (EDA), feature selection, and data preprocessing in managing high-volume, heterogeneous data sources. The results demonstrate the significant potential of integrating these predictive models into aviation Business Intelligence (BI) systems to transition from reactive to proactive decision-making. The study concludes by discussing the integration challenges within existing data architectures and the future potential of these approaches for optimising complex, networked transportation systems.

A core challenge in photocatalytic water splitting for hydrogen production, as a clean method of hydrogen fuel generation, lies in simultaneously achieving efficient hydrogen production and long‐term stability of catalysts. Herein, we report the successful synthesis of a novel bimetallic cluster‐based metal‐organic framework (MOF) exhibiting exceptional performance for visible‐light‐driven photocatalytic hydrogen production. This unique MOF catalyst demonstrates a remarkably high hydrogen evolution rate of 20,778.7 μmol g −1 h −1 under visible‐light irradiation, significantly surpassing many MOF photocatalysts. Crucially, the MOF exhibits outstanding stability, maintaining its high catalytic activity over multiple reaction cycles without significant degradation. The synergistic effects within the bimetallic clusters and sensitively light‐responsive linkers are regarded as the key contributors to both the enhanced efficiency and remarkable durability. This work presents a highly promising and stable MOF photocatalyst for efficient solar energy harnessing towards clean hydrogen fuel generation.

Commercial satellites play a pivotal role in maintaining civil communications and military operations. However, these privately operated space systems remain vulnerable, particularly when deployed in high-stakes public emergency scenarios where secure and continuous communication is critical. This paper examines the cyber risks associated with commercial satellite communication (SATCOM) networks, such as those operated by SpaceX and Amazon, when deployed during civil conflicts and national emergencies. We argue that the convergence of military reliance, profit-driven motives, and emerging AI-enabled cyber threats has created a critical need for a public–private cybersecurity paradigm. We analyse three core challenges: misaligned stakeholder interests, the rise of generative AI-enabled attacks, and transparency gaps in satellite protocol governance. Building on the National Institute of Standards and Technology (NIST) Cybersecurity Framework (CSF) and the 2024 NIST AI Risk Management Framework (AI RMF), we propose an integrated approach for securing commercial SATCOMs. Our framework adapts NIST core functions to satellite systems and aligns sector-specific guidance from NIST Internal Reports (IR)s to facilitate coordination among government, military, and commercial actors. We further evaluate existing U.S. practices, including the Cybersecurity and Infrastructure Security Agency’s Space System Working Group and the Space Force’s Infrastructure Asset Pre-Assessment Program, to assess how cross-sectoral collaboration can be standardized and institutionalized. We argue for pre-emptive regulation on AI model deployment, cryptographic protocol disclosure, and open standards for hybrid satellite networks. By synthesizing technical frameworks with policy case studies, this study makes three contributions: first, it articulates a novel application of the NIST CSF to commercial satellite cybersecurity; second, it provides a conceptual bridge between AI risk management and satellite network governance; third, it offers practical strategies for harmonizing public benefit with private infrastructure in space-based communication. This research supports the development of a resilient satellite cybersecurity ecosystem that safeguards public trust and international stability.

Art can have a positive impact on health and well-being and be efficacious in health research and dissemination processes. However, creative, arts-based approaches to research and knowledge translation sometimes have a precarious toehold in the spheres of both health and art. In this paper, we report on failures, misunderstandings, difficulties and ethical tensions associated with work undertaken in, and between, art and health. Using collaborative autoethnography, we draw on professional experiences and relevant literature to present four core challenges that can be encountered, and should be considered, when working in the art and health space: (1) who is art for? (2) gatekeeping, (3) ethical tensions and (4) taste and quality. We share these challenges to make visible the often-tacit expectations, ways of working and hierarchies of knowledge that underpin arts and health work. In our discussion, we offer suggestions for overcoming these challenges, in the hope they will be useful to others working in arts and health. We raise and explore questions-about knowledge, value, art and ethics-that might not have definitive answers, but that are productive to interrogate before undertaking an arts and health endeavour.

ABSTRACT Introduction Requirements engineering (RE) faces challenges due to the handling of increasingly complex software systems. These challenges can be addressed using generative artificial intelligence (GenAI). Given that GenAI‐based RE has not been systematically analyzed in detail, this review examines the related research, focusing on trends, methodologies, challenges, and future work directions. Methods A systematic methodology for paper selection, data extraction, and feature analysis is used to comprehensively review 238 articles published from 2019 to 2025 and available from major academic databases. Results Although generative pretrained transformer models dominate current applications (67.3% of studies), the research focus remains unevenly distributed across RE phases, with analysis (30.0%) and elicitation (22.1%) receiving the most attention and management (6.8%) remaining underexplored. Three core challenges—reproducibility (66.8%), hallucinations (63.4%), and interpretability (57.1%)—form a tightly interlinked triad affecting trust and consistency, and strong correlations ( co‐occurrence) indicate that these challenges must be addressed holistically. Industrial adoption remains nascent, with > 90% of studies corresponding to early‐stage development and only 1.3% reaching production‐level integration. Evaluation practices show maturity gaps, limited tool/dataset availability, and fragmented benchmarking approaches. Conclusions Despite the transformative potential of GenAI‐based RE, several barriers hinder its practical adoption. The strong correlations among core challenges demand specialized architectures targeting interdependencies rather than isolated solutions. The limited real‐world deployment reflects systemic bottlenecks in generalizability, data quality, and scalable evaluation methods. Successful adoption requires coordinated development across technical robustness, methodological maturity, and governance integration. A multiphase research roadmap emphasizing evaluation infrastructure strengthening, governance‐aware development, and industrial‐scale standardization is proposed.

BackgroundAs the core functional carrier of life activities, the quality of protein representation directly affects the accuracy of downstream functional prediction. In recent years, multimodal deep learning methods have significantly improved the effectiveness of protein representation learning by virtue of their advantages in fusing sequence, structure, and chemical characteristics. However, current research still faces two core challenges: first, the guiding mechanism for structural information during multi-modal feature interaction has not been fully explored; second, existing fusion strategies mostly use static weight allocation mechanisms, which is difficult to adapt to sequence-structural features. The dynamic correlation between features leads to limited accuracy in identifying key functional residues.ResultsWe proposed ProGraphTrans, a multimodal dynamic collaborative framework for protein representation learning. ProGraphTrans builds a dynamic attention multimodal fusion mechanism and captures local sequential patterns through a multi-scale convolutional neural network.ConclusionsExperimental results on four protein downstream tasks show that ProGraphTrans not only outperforms other methods in various indicators but also demonstrates excellent interpretability, demonstrating its advantages and effectiveness as a protein representation method.

ABSTRACT To address the core challenges of information asymmetry, privacy leakage, and low storage efficiency in rice supply chains, this study proposes an enhanced traceability system that integrates blockchain, adaptive encryption, and lightweight zero‐knowledge proofs. The system features a dynamic role‐based encryption model, where encryption levels are determined by both data sensitivity and role‐based weights. This model was designed and validated through surveys involving 50 stakeholders. By adopting an on‐chain and off‐chain collaborative storage architecture that leverages Merkle trees and IPFS, the system achieves a 67% reduction in storage overhead. Furthermore, an optimized Groth16‐based ZKP protocol ensures rapid verification in under 180 ms on ARM‐based devices. Experimental results demonstrate that, at a scale of 100,000 records, the system attains a transaction processing capacity of 328 TPS and an information entropy of 3.87, representing a 51% improvement over single‐layer encryption schemes. The monthly deployment cost remains affordable for smallholder farmers, ranging from $2 to $5. The system also supports interoperability with external traceability frameworks through cross‐chain channels and adaptation to the GS1 EPCIS standard, facilitating trusted collaboration in transnational rice supply chains. By effectively balancing data integrity and privacy protection, this solution significantly enhances system scalability and offers a novel pathway for the digital transformation of agricultural supply chains.

All-solid-state batteries offer a significant advancement in energydensity and safety compared to conventional lithium-ion battery technologies. However, their development is critically impeded by the complex instability of solid-solid interfaces. These buried junctions, governed by intricate chemical, electrochemical, and mechanical degradation processes, present substantial scientific challenges that cannot be effectively addressed by a single research methodology. This review emphasizes the crucial role of a synergistic paradigm that integrates computational simulations with advanced experimental characterization to overcome these significant interface issues. A conceptual framework is proposed that categorizes this synergistic approach into four hierarchical levels: Foundational, which validates essential material properties; Dynamic, which captures real-time interface evolution using operando techniques; Multi-Scale, which links atomic-scale changes to macroscopic failures; and Intelligent, which harnesses artificial intelligence and machine learning to accelerate discovery and enhance data analysis. Through detailed case studies, thevital role of this integated approach is demonstrated in elucidating ionic transport mechanisms, predicting interfacial reaction pathways, deconstructing the multiphysics of lithium-dendrite growth, and understanding chemo-mechanical failures in composite electrodes. It is concluded that this synergistic methodology is essential for transitioning from descriptive analysis to the predictive, rational engineering of stable, high-performance interfaces necessary for the next generation of energy storage systems.

From pixels to pathology: Restoration diffusion for diagnostic-consistent virtual IHC.

Lightweight Single-Image Super-Resolution (SISR) faces the core challenge of balancing computational efficiency with reconstruction quality, particularly in preserving both high-frequency details and global structures under constrained resources. To address this, we propose the Periodically Enhanced Cascade Network (PECNet). Our main contributions are as follows: 1. Its core component, a novel Multi-scale Adaptive Feature Aggregation (MAFA) module, which employs three functionally complementary branches that work synergistically: one dedicated to extracting local high-frequency details, another to efficiently modeling long-range dependencies and a third to capturing structured contextual information within windows. 2. To seamlessly integrate these branches and enable cross-window information interaction, we introduce the Periodic Boundary Padding Shift (PBPS) mechanism. This mechanism serves as a symmetric preprocessing step that achieves implicit window shifting without introducing any additional computational overhead. Extensive benchmarking shows PECNet achieves better reconstruction quality without a complexity increase. Taking the representative shift-window-based lightweight model, NGswin, as an example, for ×4 SR on the Manga109 dataset, PECNet achieves an average PSNR 0.25 dB higher, while its computational cost (in FLOPs) constitutes merely 40% of NGswin’s.

Accurate simulation of compressible turbulent flows and shock–vortex interactions remains a core challenge in computational fluid dynamics, especially when resolving fine-scale vortical structures alongside strong discontinuities. This paper introduces a hybrid weighted essentially non-oscillatory (WENO) scheme aimed at balancing the demands of shock-capturing and turbulence resolution in compressible flows. The method delivers improved accuracy in capturing both classical flow discontinuities and complex vortical structures typical of turbulent flows. The proposed scheme is tested across core case studies, including one-dimensional shock–entropy wave interactions, two-dimensional double-vortex pairing, and three-dimensional Taylor–Green vortex transition to turbulence. Results show that this hybrid scheme provides sharper resolution of discontinuities and better captures fine-scale turbulent structures. In the double-vortex pairing case, the method reduces numerical dissipation by nearly 20%, compared to earlier versions of WENO schemes, enabling a more precise depiction of vortex dynamics and mixing. For the Taylor–Green vortex, the scheme detects more turbulent structures than the 11th-order method, improving predictions of kinetic energy dissipation and enstrophy evolution. These advancements are vital for applications in science and engineering involving compressible turbulence and shock–boundary layer interactions, where accurately resolving both discontinuities and vortical features is essential.