A5036524439- X-ray Diffraction in Crystallography
- Crystallization and Solubility Studies
- Gas Sensing Nanomaterials and Sensors
- Topic Modeling
- Advanced Chemical Sensor Technologies
- Natural Language Processing Techniques
- Analytical Chemistry and Sensors
- Crystallography and molecular interactions
- ZnO doping and properties
- Advanced Sensor and Energy Harvesting Materials
- Luminescence and Fluorescent Materials
- Conducting polymers and applications
- Semiconductor materials and devices
- Polydiacetylene-based materials and applications
- Multimodal Machine Learning Applications
- Carbon Nanotubes in Composites
- Surface Modification and Superhydrophobicity
- Advanced Photocatalysis Techniques
- Perovskite Materials and Applications
- Copper-based nanomaterials and applications
- Natural Fiber Reinforced Composites
- Transition Metal Oxide Nanomaterials
- Polymer Nanocomposites and Properties
- Polymer composites and self-healing
- Ga2O3 and related materials
Hunan Institute of Science and Technology
2024
Xinjiang University
2017-2024
Xinjiang Technical Institute of Physics & Chemistry
2015-2024
University of Electronic Science and Technology of China
2024
Fujian Institute of Research on the Structure of Matter
2019-2024
Chinese Academy of Sciences
2012-2024
Guangzhou Medical University
2024
Guangzhou First People's Hospital
2024
University of Washington
2019-2023
Allen Institute for Artificial Intelligence
2022-2023
Dissolved gas detection is very important for an oil-immersed transformer fault. We examined the adsorption characteristics and sensitivity processes of monolayer NbSe2 doped with Ag, Pd, Pt on five typical gases (H2, CO, CO2, CH4, C2H2) in using first-principles calculations. While Pd–NbSe2 system only has great desorption properties H2 at high temperature, Ag/Pt–NbSe2 outstanding possessions ambient temperature. Therefore, micro-nanomaterials prepared by Ag/Pd/Pt–NbSe2 are expected to be...
Alisa Liu, Zhaofeng Wu, Julian Michael, Alane Suhr, Peter West, Alexander Koller, Swabha Swayamdipta, Noah Smith, Yejin Choi. Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing. 2023.
Picric acid (PA) is an organic substance widely used in industry and military, which poses a great threat to the environment security due its unstable, toxic, explosive properties. Trace detection of PA also challenging task because highly acidic anionic character. In this work, silver nanoparticles (AgNPs)-decorated porous silicon photonic crystals (PS PCs) were controllably prepared as surface-enhanced Raman scattering (SERS) substrates using immersion plating solution. Rhodamine 6G dye...
The effects of polydimethylsiloxane (PDMS) on phase separation, optical transmittance and surface properties including composition, morphology wettability waterborne polyurethane (WPU) containing PDMS were investigated. After the introduction into WPU backbone by polymerization, large difference in solubility parameter non-polar segment high-polar urethane segments promoted enrichment at air–polymer interface enhanced resulting rough structures. Accordingly, combination structures...
Inspired by the enhanced gas-sensing performance one-dimensional hierarchical structure, polyaniline/multi-walled carbon nanotubes (PANI/CNT) fibers were prepared. Interestingly, simple heating changed sensing characteristics of PANI from p-type to n-type and CNTs form p-n hetero junctions at core-shell interface PANI/CNT composites. The (p-PANI/CNT) (n-PANI/CNT) performed higher sensitivity NO2 NH3, respectively. response times p-PANI/CNT n-PANI/CNT 50 ppm NH3 are only 5.2 1.8 s,...
The room-temperature phosphorescence quantum efficiency of [BPy]<sub>6</sub>[Pb<sub>3</sub>Br<sub>12</sub>] has been improved by a maximum fourteen-fold through sensitization thiadiazole-based molecules.
Abstract Ex situ characterization techniques in molecular beam epitaxy (MBE) have inherent limitations, such as being prone to sample contamination and unstable surfaces during transfer from the MBE chamber. In recent years, need for improved accuracy reliability measurement has driven increasing adoption of techniques. These techniques, reflection high-energy electron diffraction, scanning tunneling microscopy, X-ray photoelectron spectroscopy, allow direct observation film growth processes...
Abstract The applications of self-assembled InAs/GaAs quantum dots (QDs) for lasers and single photon sources strongly rely on their density quality. Establishing the process parameters in molecular beam epitaxy (MBE) a specific QDs is multidimensional optimization challenge, usually addressed through time-consuming iterative trial-and-error. Here, we report real-time feedback control method to realize growth with arbitrary density, which fully automated intelligent. We develop machine...
In order to sensitively, selectively, and rapidly detect the constituents relevant improvised explosive devices (IEDs), sensing properties of ZnS nanocrystals (NCs) are regulated by tailoring doping level Mn 2+ . The responses sensors fabricated NCs with different Mn‐doping levels (Mn:ZnS) toward constituents, such as sulphur powder black powder, generally increases first then decreases increase concentration doped , reaches climate an atomic ratio 2.23% at room temperature. sensory array...
In automatic speech recognition (ASR), model pruning is a widely adopted technique that reduces size and latency to deploy neural network models on edge devices with resource constraints. However, multiple different sparsity levels usually need be separately trained deployed heterogeneous target hardware specifications for applications have various requirements. this paper, we present Dynamic Sparsity Neural Networks (DSNN) that, once trained, can instantly switch any predefined...
Inspired by the pleated structure of dog’s maxillary turbinate, porous CRT with is successfully prepared carbonization rose tea and applied to gas-sensitive detection for first time.
Herein, we present a new series of CuI-based hybrid materials with tunable structures and semiconducting properties. The CuI inorganic modules can be tailored into one-dimensional (1D) chain two-dimensional (2D) layer confined/stabilized in coordination frameworks potassium isonicotinic acid (HINA) its derivatives (HINA-R, R = OH, NO2, COOH). resulting exhibit interesting behaviors associated the dimensionality module; for instance, containing 2D-CuI module demonstrate significantly enhanced...
Recent advancements in large language models (LLMs) have demonstrated their impressive abilities various reasoning and decision-making tasks. However, the quality coherence of process can still benefit from enhanced introspection self-reflection. In this paper, we introduce Multiplex CoT (Chain Thought), a method that enables LLMs to simulate form self-review while reasoning, by initiating double Chain Thought (CoT) thinking. leverages power iterative where model generates an initial chain...
Addressing the resource waste caused by fixed computation paradigms in deep learning models under dynamic scenarios, this paper proposes a Transformer$^{-1}$ architecture based on principle of adaptivity. This achieves matching between input features and computational resources establishing joint optimization model for complexity computation. Our core contributions include: (1) designing two-layer control mechanism, composed predictor reinforcement policy network, enabling end-to-end paths;...
This paper proposes an innovative Multi-Modal Transformer framework (MMF-Trans) designed to significantly improve the prediction accuracy of Chinese stock market by integrating multi-source heterogeneous information including macroeconomy, micro-market, financial text, and event knowledge. The consists four core modules: (1) A four-channel parallel encoder that processes technical indicators, macro data, knowledge graph respectively for independent feature extraction multi-modal data; (2)...