A5100742205- Natural Language Processing Techniques
- Topic Modeling
- Advanced Sensor and Control Systems
- Iterative Learning Control Systems
- Advanced Algorithms and Applications
- Speech Recognition and Synthesis
- Aluminum Alloys Composites Properties
- Advancements in Battery Materials
- Advanced Battery Materials and Technologies
- Advanced Computational Techniques and Applications
- Power Systems and Renewable Energy
- Simulation and Modeling Applications
- Biofuel production and bioconversion
- Multimodal Machine Learning Applications
- Cooperative Communication and Network Coding
- Geochemistry and Elemental Analysis
- Enzyme Production and Characterization
- Integrated Energy Systems Optimization
- Mobile Crowdsensing and Crowdsourcing
- Microbial bioremediation and biosurfactants
- Translation Studies and Practices
- Teleoperation and Haptic Systems
- Polysaccharides and Plant Cell Walls
- Industrial Technology and Control Systems
- Advanced ceramic materials synthesis
First Affiliated Hospital of Xinjiang Medical University
2025
Xinjiang Medical University
2025
Northeast Forestry University
2023-2025
Ministry of Natural Resources
2024
Ministry of Agriculture and Rural Affairs
2024
Southwest Jiaotong University
2022-2024
Xi'an Jiaotong University
2024
China Land Surveying and Planning Institute
2024
Shaanxi Provincial Land Engineering Construction Group
2024
Shandong University of Science and Technology
2024
Continuing progress in machine learning (ML) has led to significant advancements agricultural tasks. Due its strong ability extract high-dimensional features from fruit images, deep (DL) is widely used detection and automatic harvesting. Convolutional neural networks (CNN) particular have demonstrated the attain accuracy speed levels comparable those of humans some harvesting fields. This paper presents a comprehensive overview review recognition based on DL for 2018 up now. We focus current...
Azvudine and nirmatrelvir-ritonavir (Paxlovid) were widely used to treat patients with COVID-19 in China during the Omicron wave. However, efficacy safety of azvudine versus Paxlovid are poorly established. This study included 40,876 hospitalized from eleven hospitals Henan Xinjiang Provinces, China. Clinical outcomes compared between two drugs via Kaplan-Meier analysis Cox regression models. Additionally, vitro vivo experiments evaluate antitumor effects both drugs. Single-cell RNA...
Predicting traffic conditions has been recently explored as a way to relieve congestion. Several pioneering approaches have proposed based on observations of the target location well its adjacent regions, but they obtain somewhat limited accuracy due lack mining road topology. To address effect attenuation problem, we propose take account surrounding locations. We an end-to-end framework called DeepTransport, in which Convolutional Neural Networks (CNN) and Recurrent (RNN) are utilized...
China is the largest sweet potato producer and exporter in world. Sweet residues (SPRs) separated after extracting starch account for more than 10 % of total dry matter potatoes. In China, 2 million tons SPRs cannot be utilized, unutilized are perishable result environmental pollution. Thus, an environmentally friendly highly efficient process bioethanol production from should developed. The swelling behaviour cellulose causes high-gravity to recalcitrant enzymatic hydrolysis. Cellulase...
Chinese spelling correction (CSC) is an important yet challenging task.Existing state-ofthe-art methods either only use a pre-trained language model or incorporate phonological information as external knowledge.In this paper, we propose novel end-to-end CSC that integrates phonetic features into by leveraging the powerful pre-training and fine-tuning method.Instead of conventionally masking words with special token in training model, replace their sound-alike words.We further adaptive...
Balancing accuracy and latency is a great challenge for simultaneous translation. To achieve high accuracy, the model usually needs to wait more streaming text before translation, which results in increased latency. However, keeping low would probably hurt accuracy. Therefore, it essential segment ASR output into appropriate units Inspired by human interpreters, we propose novel adaptive segmentation policy The learns source considering possible translations produced translation model,...
Abstract The roof water inrush disaster induced by coal mining is becoming a vital bottleneck restricting mine safety. To accurately predict the position of seam sandstone aquifer and prevent control it, this paper takes Lilou Coal Mine in Juye Coalfield as an example. Based on comprehensive analysis hydrogeological data, drilling geophysiological examines richness aquifer, insulation geological structure, fracture development characteristics aquifer. Starting from general side, nine...
This paper presents BSTC (Baidu Speech Translation Corpus), a large-scale Chinese-English speech translation dataset. dataset is constructed based on collection of licensed videos talks or lectures, including about 68 hours Mandarin data, their manual transcripts and translations into English, as well automated by an automatic recognition (ASR) model. We have further asked three experienced interpreters to simultaneously interpret the testing in mock conference setting. corpus expected...
ABSTRACT During blueberry picking operations, changes in the harvester’s body posture (BHBP) caused by undulating farmland surfaces significantly affect operational stability of harvesting device. To mitigate these effects, an active suspension (AS) control system was developed. First, considering coherence and time lag four-wheel tracks, filtered white noise method chosen to generate road excitations. Subsequently, virtual prototype model harvester (BH) built ADAMS with nonlinear properties...
The rare earth element lanthanum (La) as a heavy metal can cause stress damage to plants at high concentrations. Metallothioneins (MTs) are low molecular weight proteins rich in cysteine and play significant roles detoxification of ion plants. In this study, we identified novel function CnMT2 from Chrysanthemum naktongense. Two MT homologous genes, namely CnMT3, were cloned the wild plant species C. naktongense, which exhibits resistance (La). qRT-PCR analysis revealed that highest...
Initial Coulombic efficiency (ICE) is critical for determining the energy density of lithium-ion batteries (LIBs) used practical applications; however, it typically disregarded in anode research. We SiOx and graphite composite anodes commercial our preliminary research to achieve a balance between ICE, capacity, cycling life. ICE reached 88%; needs further improvement applications. Prelithiation process that involves introduction extra lithium ions into LIBs during their manufacturing...
In this paper, we present DuTongChuan, a novel context-aware translation model for simultaneous interpreting. This allows to constantly read streaming text from the Automatic Speech Recognition (ASR) and simultaneously determine boundaries of Information Units (IUs) one after another. The detected IU is then translated into fluent with two simple yet effective decoding strategies: partial decoding. practice, by controlling granularity IUs size context, can get good trade-off between latency...
End-to-end simultaneous speech-to-text translation aims to directly perform from streaming source speech target text with high quality and low latency. A typical (ST) system consists of a model policy module, which determines when wait translate. Thus the is crucial balance Conventional methods usually adopt fixed policies, e.g. segmenting length generating translation. However, this method ignores contextual information suffers quality. This paper proposes an adaptive segmentation for...
Simultaneous Translation is a great challenge in which translation starts before the source sentence finished. Most studies take transcription as input and focus on balancing quality latency for each sentence. However, most ASR systems can not provide accurate boundaries realtime. Thus it key problem to segment sentences word streaming translation. In this paper, we propose novel method boundary detection that takes multi-class classification task under end-to-end pre-training framework....
Predicting traffic conditions has been recently explored as a way to relieve congestion. Several pioneering approaches have proposed based on observations of the target location well its adjacent regions, but they obtain somewhat limited accuracy due lack mining road topology. To address effect attenuation problem, we suggest taking into account surrounding locations(wider than range). We propose an end-to-end framework called DeepTransport, in which Convolutional Neural Networks (CNN) and...
Xylan and cellulose are the two major constituents of numerous types lignocellulose. The bifunctional enzyme that exhibits xylanase/cellulase activity has attracted a great deal attention in biofuel production. Previously, thermostable GH10 family (XynA) from Bacillus sp. KW1 was found to degrade both xylan cellulose. To improve on basis structure, we first determined crystal structure XynA at 2.3 Å. Via molecular docking assays, revealed Gln250 His252 were indispensable bifunctionality,...
Purpose: Rapid detection and diagnosis of diseases facilitate timely effective treatment cardiovascular (CVD). The establishment a one-step rapid method provides new for the initial screening disease risk assessment patients with in primary medical units. Methods: Hollow gold nanoparticles (HGNPs) were synthesized using cobalt template followed by use as signal amplification probes ultra-sensitive quantitative serum C-reactive protein (CRP). To induce localized surface plasmon resonance...