A5076360288- Adversarial Robustness in Machine Learning
- Organic Electronics and Photovoltaics
- Advanced Malware Detection Techniques
- graph theory and CDMA systems
- Anomaly Detection Techniques and Applications
- Network Security and Intrusion Detection
- Organic Light-Emitting Diodes Research
- Conducting polymers and applications
- Software Engineering Research
- Topic Modeling
- Coding theory and cryptography
- Higher Education and Teaching Methods
- Domain Adaptation and Few-Shot Learning
- Software Testing and Debugging Techniques
- Perovskite Materials and Applications
- Energy Load and Power Forecasting
- Cryptography and Data Security
- Natural Language Processing Techniques
- Metaheuristic Optimization Algorithms Research
- Genomic variations and chromosomal abnormalities
- Image and Signal Denoising Methods
- Military Defense Systems Analysis
- Optimal Experimental Design Methods
- Face and Expression Recognition
- Chinese history and philosophy
Nanjing University of Posts and Telecommunications
2017-2025
KU Leuven
2021-2024
Shandong University of Science and Technology
2024
University of Colorado Colorado Springs
2023
Nanjing Institute of Technology
2023
North China Electric Power University
2023
Yulin Normal University
2022
Nanjing University
2022
Data Assurance and Communication Security
2022
University of Illinois Urbana-Champaign
2022
Abstract Hole‐transporting materials (HTMs) play a significant role in hole transport and extraction for perovskite solar cells (PeSCs). As an important type of HTMs, the spiro‐architecture‐based material is widely used as small organic HTM PeSCs with good photovoltaic performances. The skeletal modification spiro‐based HTMs critical way modifying energy level mobility. Thus, many spiro alternatives are developed to optimize spiro‐type HTMs. Herein, novel carbazole‐based single‐spiro‐HTM...
With the rapid development of text mining, many studies observe that generally contains a variety implicit information, and it is important to develop techniques for extracting such information. Named Entity Recognition (NER), first step information extraction, mainly identifies names persons, locations, organizations in text. Although existing neural-based NER approaches achieve great success language domains, most them normally ignore nested nature named entities. Recently, diverse focus...
Named entity recognition is a fundamental task in natural language processing. Recognizing entities biomedical text, known as the BioNER, particularly crucial for cutting-edge applications. However, BioNER poses greater challenges compared to traditional NER due (1) nested structures and (2) category correlations inherent entities. Recently, various models have been developed based on region classification or large models. Despite being successful, these still struggle balance handling...
Abstract White organic light‐emitting diodes (WOLEDs) composed of conventional fluorophores possess color purity, low efficiency roll‐off, and rare metal absence, but suffer from theoretical limits due to the lack triplet utilization. Due different diffusion distance for singlets triplets, multiple Förster resonance energy transfer (FRET) channels can be adequately built up. Herein, besides complementary component, a blue fluorescence layer, hosted by pure hydrocarbon material SF4‐TPE, is...
Machine Learning (ML) techniques can facilitate the automation of malicious software (malware for short) detection, but suffer from evasion attacks.Many studies counter such attacks in heuristic manners, lacking theoretical guarantees and defense effectiveness.In this paper, we propose a new adversarial training framework, termed Principled Adversarial Malware Detection (PAD), which offers convergence robust optimization methods.PAD lays on learnable convex measurement that quantifies...
A new TADF emitter was developed for blue OLEDs, and its thermal, photophysical electrochemical properties were investigated.
The fluorination on the π-bridge can obviously affect photophysical behavior of emitters and thus improve device performance.
Named Entity Recognition (NER) is a fundamental task in natural language processing. In order to identify entities with nested structure, many sophisticated methods have been recently developed based on either the traditional sequence labeling approaches or directed hypergraph structures. Despite being successful, these often fall short striking good balance between expression power for structure and model complexity. To address this issue, we present novel NER named HIT. Our proposed HIT...
Abstract With the development of electric market, load forecasting has been increasingly pursued by many scholars. Because is affected factors, it characterized volatility and uncertainty, cannot be forecasted accurately only a single model. In research, short‐term integrated model proposed to solve problem inaccurate The key point using forecast optimize decomposed sequence improve accuracy forecast. empirical wavelet decomposition (EWT) used decompose into stationary sequences avoid modal...
This article investigates the duty-ratio-based direct torque control strategies for dual three-phase permanent magnet synchronous machine (PMSM) drives. The classical switching-table-based (ST-DTC) drives usually suffers from significant harmonic currents and a large ripple. To reduce currents, 12 combined voltage vectors are generated employed as active vectors. Then, ST-DTC is introduced to Considering that typical exhibits drawbacks of parameter dependent certain degree failure rate in...
Abstract Energy consumption in the residential sector is increasing rapidly China. This study applies an integrated assessment model to investigate adverse impacts of household energy by various fuel types across rural and urban areas on age- sex-specific premature deaths associated with PM 2.5 pollution at provincial levels for 2015. We further estimate economic health co-benefits a switch from solid fuels electricity within households. find that consumed Chinese households nearly 1.6 times...
Let there is an . For or , has been determined by Hanani, and for the first author. In this paper, we investigate case A necessary condition It known that all with a possible exception We need to consider proved thereby,
Planar heterojunction (PHJ) organic photodetectors are potentially more stable than traditional bulk counterparts because of the absence uncontrolled phase separation in donor and acceptor binary blend system. This work reports a new class PHJ based on medium-band gap fullerene C60 low-band fused-ring non-fullerene ID-MeIC bilayer structure, which allows wide range spectral response tuning across UV-visible-near-infrared (UV-vis-NIR) region by tailoring individual layer thickness. The not...
Feature extraction is a key factor to detect pesticides using terahertz spectroscopy. Compared traditional methods, deep learning able obtain better insights into complex data features at high levels of abstraction. However, reports about the application in THz spectroscopy are rare. The main limitation analyse insufficient samples. In this study, we proposed WGAN-ResNet method, which combines two networks, Wasserstein generative adversarial network (WGAN) and residual neural (ResNet),...
<italic>Meta</italic>-alkyl-phenyl substitutions and fluorination approach selectively optimizes a decacyclic indacenodithiophene-based fused electron acceptor (<italic>p</italic>-IDTIDT-IC), improving PCE from 6.48% to 11.32%.