Search Results (7,128)

Recent breakthroughs in AI-generated content (AIGC) have transformed video creation, empowering systems to translate text, images, or audio into visually compelling stories. Yet reliable evaluation of these machine-crafted videos remains elusive because quality is governed not only by spatial fidelity within individual frames but also by temporal coherence across frames and precise semantic alignment with the intended message. The foundational role of sensor technologies is critical, as they determine the physical plausibility of AIGC outputs. In this perspective, we argue that multimodal large language models (MLLMs) are poised to become the cornerstone of next-generation video quality assessment (VQA). By jointly encoding cues from multiple modalities such as vision, language, sound, and even depth, the MLLM can leverage its powerful language understanding capabilities to assess the quality of scene composition, motion dynamics, and narrative consistency, overcoming the fragmentation of hand-engineered metrics and the poor generalization ability of CNN-based methods. Furthermore, we provide a comprehensive analysis of current methodologies for assessing AIGC video quality, including the evolution of generation models, dataset design, quality dimensions, and evaluation frameworks. We argue that advances in sensor fusion enable MLLMs to combine low-level physical constraints with high-level semantic interpretations, further enhancing the accuracy of visual quality assessment. Full article

Studying nonlinear oscillations in mechanical systems is fundamental to understanding complex dynamic behavior in engineering applications. While classical analytical methods remain valuable for systems with limited complexity, they become increasingly inadequate when nonlinearities are strong and geometrically induced, as in the case of large-amplitude oscillations. This paper presents a combined numerical and experimental investigation of a mechanical system composed of two coupled pendulums, exhibiting significant nonlinear behavior due to elastic deformation throughout their motion. A mathematical model of the system was developed using the MatLab/Simulink ver.6.1 environment, considering gravitational, inertial, and nonlinear elastic restoring forces. One of the major challenges in accurately modeling such systems is accurately representing damping, particularly in the absence of dedicated dampers. In this work, damping coefficients were experimentally identified through decrement measurements and incorporated into the simulation model to improve predictive accuracy. The simulation outputs, including angular displacements, velocities, accelerations, and phase trajectories over time, were validated against experimental results obtained via high-precision inertial sensors. The comparison shows a strong correlation between numerical and experimental data, with minimal relative errors in amplitude and frequency. This research represents the first stage of a broader study aimed at analyzing forced and parametrically excited oscillations. Beyond validating the model, the study contributes to the design of a robust experimental framework suitable for further exploration of nonlinear dynamics. The findings have practical implications for the development and control of mechanical systems subject to dynamic loads, with potential applications in automation, vibration analysis, and system diagnostics. Full article

To enhance the molten salt corrosion resistance of Ni200 alloy plasma arc welds, the welds were subjected to tensile deformation followed by heat treatment. The grain boundary character distribution (GBCD) was analyzed using electron backscatter diffraction (EBSD) in conjunction with orientation imaging microscopy (OIM). A constant-temperature corrosion test at 900 °C was conducted to evaluate the impact of GBCD on the corrosion resistance of the welds. Results demonstrated that after processing with 6% tensile deformation, and annealing at 950 °C for 30 min, the fraction of low-ΣCSL grain boundaries increased from 1.2% in the as-welded condition to 57.3%, and large grain clusters exhibiting Σ3ⁿ orientation relationships were formed. During the heat treatment, an increased number of recrystallization nucleation sites led to a reduction in average grain size from 323.35 μm to 171.38 μm. When exposed to a high-temperature environment of 75% Na₂SO₄-25% NaCl mixed molten salt, the corrosion behavior was characterized by intergranular attack, with oxidation and sulfidation reactions resulting in the formation of NiO and Ni₃S₂. The corrosion resistance of Grain boundary engineering (GBE)-treated samples was significantly superior to that of Non-GBE samples, with respective corrosion rates of 0.3397 mg/cm²·h and 0.8484 mg/cm²·h. These findings indicate that grain boundary engineering can effectively modulate the grain boundary character distribution in Ni200 alloy welds, thereby enhancing their resistance to molten salt corrosion. Full article

Background/Objectives: The RTK-RAS signaling cascade is a central axis in colorectal cancer (CRC) pathogenesis, governing cellular proliferation, survival, and therapeutic resistance. Somatic alterations in key pathway genes—including KRAS, NRAS, BRAF, and EGFR—are pivotal to clinical decision-making in precision oncology. However, the integration of these genomic events with clinical and demographic data remains hindered by fragmented resources and a lack of accessible analytical frameworks. To address this challenge, we developed AI-HOPE-RTK-RAS, a domain-specialized conversational artificial intelligence (AI) system designed to enable natural language-based, integrative analysis of RTK-RAS pathway alterations in CRC. Methods: AI-HOPE-RTK-RAS employs a modular architecture combining large language models (LLMs), a natural language-to-code translation engine, and a backend analytics pipeline operating on harmonized multi-dimensional datasets from cBioPortal. Unlike general-purpose AI platforms, this system is purpose-built for real-time exploration of RTK-RAS biology within CRC cohorts. The platform supports mutation frequency profiling, odds ratio testing, survival modeling, and stratified analyses across clinical, genomic, and demographic parameters. Validation included reproduction of known mutation trends and exploratory evaluation of co-alterations, therapy response, and ancestry-specific mutation patterns. Results: AI-HOPE-RTK-RAS enabled rapid, dialogue-driven interrogation of CRC datasets, confirming established patterns and revealing novel associations with translational relevance. Among early-onset CRC (EOCRC) patients, the prevalence of RTK-RAS alterations was significantly lower compared to late-onset disease (67.97% vs. 79.9%; OR = 0.534, p = 0.014), suggesting the involvement of alternative oncogenic drivers. In KRAS-mutant patients receiving Bevacizumab, early-stage disease (Stages I–III) was associated with superior overall survival relative to Stage IV (p = 0.0004). In contrast, BRAF-mutant tumors with microsatellite-stable (MSS) status displayed poorer prognosis despite higher chemotherapy exposure (OR = 7.226, p < 0.001; p = 0.0000). Among EOCRC patients treated with FOLFOX, RTK-RAS alterations were linked to worse outcomes (p = 0.0262). The system also identified ancestry-enriched noncanonical mutations—including CBL, MAPK3, and NF1—with NF1 mutations significantly associated with improved prognosis (p = 1 × 10⁻⁵). Conclusions: AI-HOPE-RTK-RAS exemplifies a new class of conversational AI platforms tailored to precision oncology, enabling integrative, real-time analysis of clinically and biologically complex questions. Its ability to uncover both canonical and ancestry-specific patterns in RTK-RAS dysregulation—especially in EOCRC and populations with disproportionate health burdens—underscores its utility in advancing equitable, personalized cancer care. This work demonstrates the translational potential of domain-optimized AI tools to accelerate biomarker discovery, support therapeutic stratification, and democratize access to multi-omic analysis. Full article

Blasting is an effective technique for fragmenting rock in open-pit mining operations. Blasting operations produce either boulders or fine fragments, both of which increase costs and pose environmental risks. As a result, predicting the mean fragmentation size (MFS) distribution of rock is critical for assessing blasting operations’ quality and mitigating risks. Due to the limitations of empirical and statistical models, several researchers are turning to artificial intelligence (AI)-based techniques to predict the MFS distribution of rock. Thus, this study uses three AI tree-based algorithms—extra trees (ET), gradient boosting (GB), and random forest (RF)—to predict the MFS distribution of rock. The prediction accuracy of the models is optimized utilizing the tree-structured Parzen estimators (TPEs) algorithm, which results in three models: TPE-ET, TPE-GB, and TPE-RF. The dataset used in this study was collected from the published literature and through the data augmentation of a large-scale dataset of 3740 blast samples. Among the evaluated models, the TPE-ET model exhibits the best performance with a coefficient of determination (R²), root mean squared error (RMSE), mean absolute error (MAE), and max error of 0.93, 0.04, 0.03, and 0.25 during the testing phase. Moreover, the block size (XB, m) and modulus of elasticity (E, GPa) parameters are identified as the most influential parameters for predicting the MFS distribution of rock. Lastly, an interactive web application has been developed to assist engineers with the timely prediction of MFS. The predictive model developed in this study is a reliable intelligent model because it combines high accuracy with a strong, explainable AI tool for predicting MFS. Full article

Polymer-based electrothermal composites (PECs) have been increasingly attracting attention in recent years owing to their flexibility, low density, and high electrothermal efficiency. However, although a large number of reviews have focused on flexible and transparent film heaters as well as polymer-based conductive composites, comprehensive reviews of polymer-based electrothermal composites remain limited. Herein, we provide a comprehensive review of recent advancements in polymer-based electrothermal materials. This review begins with an introduction to the electrothermal theoretical basis and the research progress of PECs incorporating various conductive fillers, such as graphene, carbon nanotubes (CNTs), carbon black (CB), MXenes, and metal nanowires. Furthermore, a critical discussion is provided to emphasize the factors influencing the electrothermal conversion efficiency of these composites. Meanwhile, the development of multi-functional electrothermal materials has been also summarized. Finally, the application progress, future prospects, limitations, and potential directions for PEC are discussed. This review aims to serve as a practical guide for engineers and researchers engaged in the development of polymer-based electrothermal composites. Full article

The transition to clean and renewable energy sources is critically dependent on efficient hydrogen production technologies. This review surveys recent advances in photocatalytic water splitting, focusing on MNb₂O₆ nanomaterials, which have emerged as promising photocatalysts due to their tunable band structures, chemical robustness, and tailored morphologies. The objectives of this work are to (i) encompass the current synthesis strategies for MNb₂O₆ compounds; (ii) assess their structural, electronic, and optical properties in relation to photocatalytic performance; and (iii) elucidate the mechanisms underpinning enhanced hydrogen evolution. Main data collection methods include a literature review of experimental studies reporting bandgap measurements, structural analyses, and hydrogen production metrics for various MNb₂O₆ compositions—especially those incorporating transition metals such as Mn, Cu, Ni, and Co. Novelty stems from systematically detailing the relationships between synthesis routes (hydrothermal, solvothermal, electrospinning, etc.), crystallographic features, conductivity type, and bandgap tuning in these materials, as well as by benchmarking their performance against more conventional photocatalyst systems. Key findings indicate that MnNb₂O₆, CuNb₂O₆, and certain engineered heterostructures (e.g., with g-C₃N₄ or TiO₂) display significant visible-light-driven hydrogen evolution, achieving hydrogen production rates up to 146 mmol h⁻¹ g⁻¹ in composite systems. The review spotlights trends in heterojunction design, defect engineering, co-catalyst integration, and the extension of light absorption into the visible range, all contributing to improved charge separation and catalytic longevity. However, significant challenges remain in realizing the full potential of the broader MNb₂O₆ family, particularly regarding efficiency, scalability, and long-term stability. The insights synthesized here serve as a guide for future experimental investigations and materials design, advancing the deployment of MNb₂O₆-based photocatalysts for large-scale, sustainable hydrogen production. Full article

Relapsed/refractory diffuse large B-cell lymphoma and other non-Hodgkin lymphomas represent significant clinical challenges, particularly in patients who have exhausted standard immunochemotherapy and cellular therapies. Bispecific antibodies and antibody–drug conjugates have emerged as promising treatments, offering targeted and more effective treatment options compared to current standards. Bispecific antibodies, including epcoritamab and glofitamab, third-line therapies for diffuse large B-cell lymphoma, are recombinant immunoglobulins engineered to recognize two distinct antigens or epitopes simultaneously. This capability enhances therapeutic precision by bridging immune effector cells and tumor cells and modulating multiple signaling pathways involved in the pathogenesis of non-Hodgkin lymphoma. In the context of new therapies, antibody–drug conjugates, such as loncastuximab tesirine, are therapies composed of monoclonal antibodies linked to cytotoxic agents, in which the antibody selectively binds to tumor-associated antigens, delivering the cytotoxic payload directly to cancer cells while minimizing off-target effects. They combine the specificity of antibodies with the potency of chemotherapy, offering enhanced efficacy and safety in hematological malignancies. Ongoing clinical trials are investigating other molecules like odronextamab and the use of bispecific antibodies in combination regimens and earlier lines of therapy. The aim of this review is to explore actual therapies in relapsed/refractory diffuse large B-cell lymphoma, focusing on bispecific antibodies and antibody–drug conjugates. Full article

A vast proportion of scientific data remains locked behind dynamic web interfaces, often called the deep web—inaccessible to conventional search engines and standard crawlers. This gap between data availability and machine usability hampers the goals of open science and automation. While registries like FAIRsharing offer structured metadata describing data standards, repositories, and policies aligned with the FAIR (Findable, Accessible, Interoperable, and Reusable) principles, they do not enable seamless, programmatic access to the underlying datasets. We present FAIRFind, a system designed to bridge this accessibility gap. FAIRFind autonomously discovers, interprets, and operationalizes access paths to biological databases on the deep web, regardless of their FAIR compliance. Central to our approach is the Deep Web Communication Protocol (DWCP), a resource description language that represents web forms, HyperText Markup Language (HTML) tables, and file-based data interfaces in a machine-actionable format. Leveraging large language models (LLMs), FAIRFind combines a specialized deep web crawler and web-form comprehension engine to transform passive web metadata into executable workflows. By indexing and embedding these workflows, FAIRFind enables natural language querying over diverse biological data sources and returns structured, source-resolved results. Evaluation across multiple open-source LLMs and database types demonstrates over 90% success in structured data extraction and high semantic retrieval accuracy. FAIRFind advances existing registries by turning linked resources from static references into actionable endpoints, laying a foundation for intelligent, autonomous data discovery across scientific domains. Full article

In practical engineering, buildings are predominantly subjected to combined forces, and reinforced concrete (RC) beams serve as the primary load-bearing components of buildings. However, there is a paucity of research on the torsional effects of RC beams, particularly concerning the torsional failure mechanisms of large-size beams. To address this gap, this paper establishes a meso-scale numerical analysis model for RC beams reinforced with Carbon Fiber Reinforced Polymer (CFRP) sheets under combined bending and torsion pressures. The research analyzes how the fiber ratio and torsion-bending ratio govern torsion-induced failure characteristics and size effects in CFRP-strengthened RC beams. The results indicate that an increase in the fiber ratio leads to accumulated damage distribution in the RC beam, a gradual decrease in CFRP sheet strain, and an increase in peak load and peak torque, albeit with diminishing amplitudes; as the torsion-bending ratio increases, crack distribution becomes more concentrated, the angle between cracks and the horizontal direction decreases, overall peak load decreases, peak torque increases, and CFRP sheet strain increases; and the nominal torsional capacity of CFRP-strengthened RC beams declines with increasing size, exhibiting a reduction of 24.1% to 35.6%, which distinctly demonstrates the torsional size effect under bending–torsion coupling conditions. A modified Torque Size Effect Law is formulated, characterizing in quantitative terms the dependence of the fiber ratio and the torsion-bending ratio. Full article

Bacteriophages (phages), the most abundant biological entities on Earth, have long served as both model systems and therapeutic tools. Recent advances in synthetic biology and genetic engineering have revolutionized the capacity to tailor phages with enhanced functionality beyond their natural capabilities. This review outlines the current landscape of synthetic and functional engineering of phages, encompassing both in-vivo and in-vitro strategies. We describe in-vivo approaches such as phage recombineering systems, CRISPR-Cas-assisted editing, and bacterial retron-based methods, as well as synthetic assembly platforms including yeast-based artificial chromosomes, Gibson, Golden Gate, and iPac assemblies. In addition, we explore in-vitro rebooting using TXTL (transcription–translation) systems, which offer a flexible alternative to cell-based rebooting but are less effective for large genomes or structurally complex phages. Special focus is given to the design of customized phages for targeted applications, including host range expansion via receptor-binding protein modifications, delivery of antimicrobial proteins or CRISPR payloads, and the construction of biocontained, non-replicative capsid systems for safe clinical use. Through illustrative examples, we highlight how these technologies enable the transformation of phages into programmable bactericidal agents, precision diagnostic tools, and drug delivery vehicles. Together, these advances establish a powerful foundation for next-generation antimicrobial platforms and synthetic microbiology. Full article

VisRep (Visualisation Report) is an AI-powered system for capturing and structuring the early stages of the visualisation design process. It addresses a critical gap in predesign: the lack of tools that can naturally record, organise, and transform raw ideation, spoken thoughts, sketches, and evolving concepts into polished, shareable outputs. Users engage in talk-aloud sessions through a terminal-style interface supported by intelligent transcription and eleven structured questions that frame intent, audience, and output goals. These inputs are then processed by a large language model (LLM) guided by markdown-based output templates for reports, posters, and slides. The system aligns free-form ideas with structured communication using prompt engineering to ensure clarity, coherence, and visual consistency. VisRep not only automates the generation of professional deliverables but also enhances reflective practice by bridging spontaneous ideation and structured documentation. This paper introduces VisRep’s methodology, interface design, and AI-driven workflow, demonstrating how it improves the fidelity and transparency of the visualisation design process across academic, professional, and creative domains. Full article

To date, one of the main problems associated with the engineering application of metallic materials is corrosion protection. To increase their durability and reduce damage, a variety of protection methods have been studied and applied. In recent decades, coating techniques have become increasingly important. Among these coatings, Layered Double Hydroxides (LDHs) have shown unique properties, such as ion exchange, high adhesion, and hydrophobicity, particularly useful for biomedical applications. In this review, after a detailed exposition of the LDHs’ synthesis processes, the most recent corrosion protection methods are illustrated. Intercalation of corrosion inhibitors and release kinetics of intercalates are presented. Although this work is mainly focused on laboratory-scale investigations and fundamental research, the problems inherent to large-scale industrial manufacturing and application are outlined and briefly discussed. Full article

LLMs are trained using many thousands of GPUs in well-known conventional models. It is necessary to address numerous issues in the training process, such as manual data collection organization, data parallel, model parallel, evaluation, testing, deployment, transferring large data streams, detecting errors, ongoing maintenance, and project management. A team of dozens of engineers is required to handle system problems in the training process. Therefore, it is time-consuming and expensive to build an efficient and fault-tolerant system based on Kubernetes. This paper develops SimpleScale for building LLMs based on FSDP and Slurm, which is a simple and efficient training system that includes the training agent, the efficient parallel strategy, the optimal step of checkpoint, and so on. Using the proposed system enables us to significantly accelerate the process of building the LLM without incurring substantial time spent on various manual issues. The proposed 1024-GPU cluster was tested on TinyLlama, which has 1.1 billion parameters and 300 billion tokens. For example, utilizing a 16-H100 GPU cluster accelerated the traditional TinyLlama training time costs from 89.05 days to 39.05 days. In the future, the project team plans to integrate Flash-Attention3, aiming for an MFU of more than 60% while maintaining the acceleration achieved in the present work. Full article

Driven by carbon neutrality and peak carbon policies, hydrogen energy, due to its zero-emission and renewable properties, is increasingly being used in hydrogen fuel cell vehicles (H-FCVs). However, the high cost and limited durability of H-FCVs hinder large-scale deployment. Hydrogen internal combustion engine hybrid electric vehicles (H-HEVs) are emerging as a viable alternative. Research on the techno-economics of H-HEVs remains limited, particularly in systematic comparisons with H-FCVs. This paper provides a comprehensive comparison of H-FCVs and H-HEVs in terms of total cost of ownership (TCO) and hydrogen consumption while proposing a multi-objective powertrain parameter optimization model. First, a quantitative model evaluates TCO from vehicle purchase to disposal. Second, a global dynamic programming method optimizes hydrogen consumption by incorporating cumulative energy costs into the TCO model. Finally, a genetic algorithm co-optimizes key design parameters to minimize TCO. Results show that with a battery capacity of 20.5 Ah and an H-FC peak power of 55 kW, H-FCV can achieve optimal fuel economy and hydrogen consumption. However, even with advanced technology, their TCO remains higher than that of H-HEVs. H-FCVs can only become cost-competitive if the unit power price of the fuel cell system is less than 4.6 times that of the hydrogen engine system, assuming negligible fuel cell degradation. In the short term, H-HEVs should be prioritized. Their adoption can also support the long-term development of H-FCVs through a complementary relationship. Full article