67639 research outputs found
Sort by
OmniResponse: Online Multimodal Conversational Response Generation in Dyadic Interactions
No full textIn this paper, we introduce Online Multimodal Conversational Response Generation (OMCRG), a novel task that aims to online generate synchronized verbal and non-verbal listener feedback, conditioned on the speaker's multimodal input. OMCRG reflects natural dyadic interactions and poses new challenges in achieving synchronization between the generated audio and facial responses of the listener. To address these challenges, we innovatively introduce text as an intermediate modality to bridge the audio and facial responses. We hence propose OmniResponse, a Multimodal Large Language Model (MLLM) that autoregressively generates high-quality multi-modal listener responses. OmniResponse leverages a pretrained LLM enhanced with two novel components: Chrono-Text, which temporally anchors generated text tokens, and TempoVoice, a controllable online TTS module that produces speech synchronized with facial reactions. To support further OMCRG research, we present ResponseNet, a new dataset comprising 696 high-quality dyadic interactions featuring synchronized split-screen videos, multichannel audio, transcripts, and facial behavior annotations. Comprehensive evaluations conducted on ResponseNet demonstrate that OmniResponse significantly outperforms baseline models in terms of semantic speech content, audio-visual synchronization, and generation quality.This work is supported by the KAUST Center of Excellence for Generative AI under award number 5940. The computational resources are provided by IBEX, which is managed by the KAUST Supercomputing Core Laboratory
Design and Channel Modeling of Electromagnetically Reconfigurable Antennas
No full textIn this work, a novel design of electromagnetically reconfigurable antennas (ERAs) based on a fluid antenna system (FAS) is proposed, and the corresponding wireless channel model is established. Different from conventional antenna arrays with static elements, the electromagnetic characteristics of each array element in the proposed ERA can be flexibly reconfigured into various states, introducing electromagnetic degrees of freedom to enhance wireless system performance. Based on the proposed ERA design, the corresponding channel model is developed. Finally, full-wave simulations are conducted to validate the overall design concept. The results reveal that a gain enhancement of 2.5 dB is achieved at a beamforming direction
Investigation of Fluid Segregation and Phase Redistribution in Wellbore Pressure Buildup
No full textSummary
During wellbore pressure buildup tests, fluid segregation and phase redistribution commonly introduce oscillations and anomalies into pressure measurements that complicate Pressure Transient Analysis (PTA) interpretation. Fluid segregation and redistribution lead to three distinct pressure response types, with pronounced effects observed at high water cuts, where redistribution is most significant. Corrective techniques, such as rate normalization and type-curve matching, proved effective in mitigating the impact of redistribution on pressure analysis. However, these solutions fail to capture complex physics in the wellbore and cannot accurately represent phase redistribution effects at early times. This paper presents an integrated framework combining analytical solutions to overcome these challenges and improve reservoir characterization. We first review conventional type-curve techniques and identify their inability to capture early-time distortions. Building on the models of Fair (1981), Hegeman et al. (1993), and Li et al. (2014), we show that existing formulations cannot reproduce the varying timing of the pressure hump. To address these limitations, we derive a new analytical solution that incorporates a flexible time-delay term which allows the hump to be displaced in time and match field observations. We also introduce a cleanup strategy that removes phase redistribution effects directly from the measured pressure data rather than relying on pressure-adjustment fitting or matching in data affected by phase redistribution. A synthetic dimensionless example demonstrates that our approach accurately reproduces hump characteristics, recovers the ideal pressure response, and eliminates redistribution artifacts. The resulting framework equips engineers with practical tools for reliable data interpretation, improved reservoir parameter estimation, optimized well-testing procedures, and enhanced operational efficiency under multiphase flow conditions.<br
Single-Atom Catalysts for Light-Assisted Green Chemical Processes
No full textFacing serious environmental pressures from industrial CO2 emissions, adopting solar energy, an abundant, carbon-free resource, as a substitute for fossil-derived heat and electrons is a promising decarbonization pathway. The limited efficiency of reactant adsorption and activation continues to hinder photocatalytic performance. By offering atomically dispersed active sites, SACs present a feasible pathway to mitigate these bottlenecks.
In this dissertation, SACs were developed and applied to a series of representative industrial chemical transformations. EXAFS and aberration-corrected HAADF-STEM were employed to confirm the presence of isolated metal centers, while in situ EPR and in situ DRIFTS were performed to track the key intermediates and clarify how the optimized catalytic architecture influence the reaction mechanism. Photoexcited carriers was captured and concentrated at single-atom sites, strengthening reactant activation and thus lowering kinetic barriers, and the uniform, well-defined coordination environment resulted a single, predictable adsorption mode, yielding a consistent transformation pathway. Consequently, compared with the conventional industrial process, the optimized systems achieve high product formation rates and selectivity at reduced operating temperatures, substantially decreasing fossil-energy demand and associated CO2 emissions. In the final stage of research, outdoor trials under natural sunlight were performed and delivered stable production of H2 and pyruvic acid; the persistence of activity and selectivity under fluctuating irradiance and ambient conditions underscores the robustness, scalability, and real-world application potential of these solar-driven catalytic systems
The global drought-sensitive areas will expand in the future
No full textClimate change and human activities are intensifying drought conditions, significantly altering drought propagation processes. However, these changes remain insufficiently understood, particularly for typical drought partitions such as humid, drought-sensitive, and arid types of drought propagation partition. Therefore, this study examines the change on these partitions under three Shared Socioeconomic Pathways (SSP1-2.6, SSP3-7.0, and SSP5-8.5) to elucidate global drought trends. The results show that the global drought-sensitive area will expand by 1.89 × 106 km2, which is roughly the size of Sudan (1.88 × 106 km2). Notably, in South America alone, the drought-sensitive area is projected to increase by 1.27 × 106 km2, equivalent to the size of Niger. While in South America the humid partition will shrink by 1.45 × 106 km2, with nearly 60 % transitioning into drought-sensitive areas. Overall, human activities are expected to drive drought partitions in an unfavorable direction, with approximately 3.26 × 106 km2 will shift from humid-type partition to drought-sensitive areas or from drought-sensitive areas to arid-type partitions under the SSP5-8.5 scenario, similar to the size of India (3.29 × 106 km2). Among all the climate zones, the changes are most notable in the humid and semi-arid regions. Significant transformations in drought propagation patterns are evident in the Great Plains of the United States of America, southern and central Europe, and the Amazon basin in South America, where declining precipitation and intensifying land–atmosphere coupling are the main driving factors. These findings provide critical insights for drought prediction and management in the context of climate change.This study was supported by the National Key Research and Development Program of China (No. 2024YFF1306305), the Natural Science Foundation of China (No. 42171022), the BNU-FGS Global Environmental Change Program (No.2023-GC-ZYTS-06), and the China Scholarship Council (No.202306040129)
Investigation of Micro Gas Turbines Topped with Wave Rotors for a Hybrid Vehicle Range Extender Engine
No full textMicro gas turbines are gaining renewed interest as range-extender
engines in hybrid vehicles due to their superior power-to-weight ratio,
fuel flexibility, and robust steady-state performance. However, their
widespread adoption is hindered by modest efficiency and high
component costs, particularly from recuperators. This study
investigates the thermodynamic performance enhancement of two
commercial micro gas turbines, the Capstone C-30 and C-60, through
wave rotor integration as a topping device. Using Aspen Plus and
Aspen Custom Modeler, three configurations were analyzed: a
recuperated engine with a single wave rotor, and unrecuperated
engines with a single and two cascaded wave rotors, respectively. Key
performance metrics—including brake thermal efficiency, specific
fuel consumption, and specific work—were evaluated across a range
of wave rotor pressure ratios.
Results show that the wave rotor significantly improves power output
and pressure ratio while maintaining or improving thermal efficiency.
The cascaded wave rotor configuration delivered the highest gains,
with power output increased by up to 80–95% and BTE reaching
32.4%. However, this also introduced challenges such as high
combustor outlet temperatures (>1400 K), raising the potential for NOₓ
formation and increased system complexity. A detailed sensitivity
analysis confirmed that the thermodynamic benefits scale with
pressure ratio up to a threshold (WR ≈ 2.6–3.0), beyond which gains
diminish. The study concludes that wave rotor-enhanced MGTs,
particularly when coupled with catalytic combustors, present a
potentially promising solution for high-performance, low-emission
hybrid vehicle range extender applications.Funding for this project came from baseline funding provided by the King Abdullah University of Science and Technology and is
gratefully acknowledged
Comprehensive Mapping of EZ-Tn5 Transposon Insertion Sites in Pseudomonas argentinensis SA190 Using RATE-PCR
No full textTransposon mutagenesis is a powerful tool for investigating gene function in bacteria, particularly in newly discovered species. In this study, we applied the hyperactive EZ-Tn5 transposase system to Pseudomonas argentinensis SA190, an endophytic bacterium known for enhancing plant resilience under drought stress. By leveraging the random amplification of transposon ends (RATE)-PCR method, we successfully mapped the insertion sites of the transposon within the SA190 genome. This approach enabled the precise identification of disrupted genes, offering insights into their roles in bacterial function and interaction with host plants. Our comprehensive protocol, including competent cell preparation, transformation, and insertion site mapping, provides a reliable framework for future studies aiming to explore gene function through mutagenesis. Key features • The use of the hyperactive EZ-Tn5 transposase system ensures efficient and detectable random mutagenesis across the Pseudomonas argentinensis SA190 genome, facilitating comprehensive gene disruption studies. • The technique is employed to identify and map the transposon insertion sites, allowing for precise determination of gene function and its impact on bacterial phenotypes. • This method enables the exploration of a broad range of gene functions within SA190, particularly those involved in plant growth promotion and stress tolerance. • This method can be readily adapted to generate mutant libraries in other bacterial species, emphasizing its transferability.The authors would like to thank all members of Hirt lab, the CDA management team, and the Bioscience Core Labs in KAUST for the technical assistance and their help in many aspects of this work. The work was funded by KAUST fund BAS/1/1062-01-01 to H.H. as part of the DARWIN21 desert initiative (http://www.darwin21.org/). S.K. and B.E. are supported by the Deutsche Forschungsgemeinschaft (DFG) under Germany’s Excellence Strategy – EXC 2048/1 – project 390686111 and under Priority Programme "2125 Deconstruction and Reconstruction of Plant Microbiota (DECRyPT)," project 401836049
Exploring the Latent Information in Spatial Transcriptomics Data via Multi-View Graph Convolutional Network Based on Implicit Contrastive Learning.
No full textLatest developments in spatial transcriptomics enable thoroughly profiling of gene expression while preserving tissue microenvironment. Connecting gene expression with spatial arrangement is key for precise spatial domain identification, enhancing the comprehension of tissue microenvironments and biological processes. However, accurately analyzing spatial domains with similar gene expression and histological features is still challenging. This study introduces STMIGCL, a novel framework that leverages a multi-view graph convolutional network and implicit contrastive learning. First, it creates neighbor graphs from gene expression and spatial coordinates, and then combines these with gene expression through multi-view learning to learn low-dimensional representations. To further refine the obtained low-dimensional representations, a graph contrastive learning method with contrastive enhancement in the latent space is employed, aiming to better capture critical information in the data and improve the accuracy and discriminative power of the embeddings. Finally, an attention mechanism is used to adaptively integrate different views, capturing the importance of spots in various views to obtain the final spot representation. Experimental data confirms that STMIGCL significantly enhances spatial domain recognition precision and outperforms all baseline methods in tasks such as trajectory inference and Spatially Variable Genes (SVGs) recognition.This work was supported by the National Natural Science Foundation of China (No. 62172248), the Natural Science Foundation of Shandong Province of China (No. ZR2021MF098), and the King Abdullah University of Science and Technology (KAUST) Office of Sponsored Research (OSR) under award numbers FCC/1/1976\u201044\u201001, FCC/1/1976\u2010 45\u201001, URF/1/4379\u201001\u201001, and REI/1/4742\u201001\u201001
Magnetic Memory Driven by Orbital Current
No full textSpin-orbitronics, based on both spin and orbital angular momentum, presents a promising pathway for energy-efficient memory and logic devices. Recent studies have demonstrated the emergence of orbital currents in light transition metals such as Ti, Cr, and Zr, broadening the scope of spin-orbit torque (SOT). In particular, the orbital Hall effect, which arises independently of spin-obit coupling, has shown potential for enhancing torque efficiency in spintronic devices. However, the direct integration of orbital current into magnetic random-access memory (MRAM) remains unexplored. In this work, we design a light metal/heavy metal/ferromagnet multilayer structure and experimentally demonstrate magnetization switching by orbital current. Furthermore, we have realized a robust SOT-MRAM cell by incorporating a reference layer that is pinned by a synthetic antiferromagnetic structure. We observed a tunnel magnetoresistance of 66%, evident in both magnetic field and current-driven switching processes. Our findings underscore the potential for employing orbital current in designing next-generation spintronic devices.This work is supported by the King Abdullah University of Science and Technology, Office of Sponsored Research (OSR), under award Nos. ORA-CRG10-2021-4665, andORA-CRG11-2022-5031. A.C. acknowledges support by the National Key Research and Development Program of China (No. 2024YFA1408503) and Sichuan Province Science and Technology Support Program (No. 2025YFHZ0147). Z.Q.Q. acknowledge
support from US Department of Energy, Office of Science, Office of Basic Energy Sciences, Materials Sciences and Engineering Division under Contract No. DE-AC02- 05CH11231 (van der Waals heterostructures program, KCWF16), Future Materials Discovery Program through the National Research Foundation of Korea (No. 2015M3D1A1070467), and Science Research Center Program through the National Research Foundation of Korea (No. 2015R1A5A1009962)
Graphene-Interfaced NiCu-Layered Double Hydroxide Electrocatalyst for Hydrogen Production via Water Splitting
No full textThe widespread use of robust catalysts for water-splitting remains limited in practical use due to the instability of conductive supports in harsh electrolytes. This instability leads to the loss of electrocatalytic activity, eventually rendering the electrocatalyst ineffective. Therefore, there is an urgent need to develop more robust and stable electrocatalysts that can withstand harsh conditions and ensure long-term durability and sustainability. We address this challenge by introducing a highly stable hybrid electrode that combines graphene-coated nickel foam with nickel and copper-layered double hydroxide (NiCu-LDH) for water splitting. This unique hybrid electrode has enabled us to achieve 420 h of overall water-splitting performance at the applied potential of 2 V. To optimize the catalyst for both the hydrogen evolution reaction (HER) and oxygen evolution reaction (OER), we systematically investigated different Ni/Cu ratios in the NiCu-LDH. The exceptional stability of these NiCu-LDH hybrid electrodes, particularly the interfacial graphene layer, makes them highly promising for long-lasting and effective water-splitting applications. The synergistic interaction between the NiCu-LDH and graphene layers, and its unique two-dimensional structure with robust carbon–carbon bonds confer exceptional structural integrity, empowering the electrode to withstand deformation and maintain its catalytic performance over extended periods