A5028878544- Network Security and Intrusion Detection
- Spam and Phishing Detection
- Internet Traffic Analysis and Secure E-voting
- Advanced Malware Detection Techniques
- Cloud Computing and Resource Management
- Online Learning and Analytics
- Creativity in Education and Neuroscience
- Advanced Graph Neural Networks
- Scheduling and Optimization Algorithms
- Air Quality and Health Impacts
- Sleep and related disorders
- Semiconductor materials and devices
- Sleep and Work-Related Fatigue
- Artistic and Creative Research
- Advanced Steganography and Watermarking Techniques
- Distributed and Parallel Computing Systems
- Advanced Bandit Algorithms Research
- Electron and X-Ray Spectroscopy Techniques
- Advancements in Battery Materials
- Anomaly Detection Techniques and Applications
- Machine Learning and Data Classification
University of Science and Technology of China
2022-2024
China Tobacco
2024
Smart community networks bring great comfort and convenience for people, but also increase security risks of exposing system vulnerabilities private data to network intruders. This problem has become more prominent as the ever-increasing zero-day attacks which may escape existing intrusion detection (IDS) through unknown vulnerabilities. In this article, keep up with continuous change attacks, we conceive an evolvable IDS (EIDS), where model is incrementally updated turn newfound "unknown"...
Layered sodium transition-metal oxides generally encounter severe capacity decay and inferior rate performance during cycling, especially at a high state of charge. Herein, defect concentration is rationally modulated to explore the impact on electrochemical behavior in NaNi1/3Fe1/3Mn1/3O2 layered oxides. Bulk vacancies are increased through annealing an oxygen-rich atmosphere, demonstrated by electron paramagnetic resonance measurement. It found that cathode with enriched oxygen exhibits...
This work introduces Weaver, our first family of large language models (LLMs) dedicated to content creation. Weaver is pre-trained on a carefully selected corpus that focuses improving the writing capabilities models. We then fine-tune for creative and professional purposes align it preference writers using suit novel methods instruction data synthesis LLM alignment, making able produce more human-like texts follow diverse instructions The consists Mini (1.8B), Base (6B), Pro (14B), Ultra...
INTRODUCTION: Most airline pilots reported having suffered from sleep disorders and fatigue due to circadian disruption, a potential risk flight safety. This study attempted uncover the actual scenario of disruption working load status among pilots. METHODS: In 1, 21 were invited participate in 14-d monitoring dual 2-back test monitor their patterns cognitive function level. To provide an in-depth view, data scheduled flights, including 567 pilots, was analyzed Study 2. The present used...
In recent years, deep learning has been widely applied to the detection of malicious URLs and made a great success. However, latent patterns change constantly, which requires frequent model update for system. Existing URL system basically relies on retaining, is time-consuming resource-consuming. this paper, we propose an incremental based closed-loop system, where continually updates fuse knowledge samples with different from previously known in training set. We develop convolutional neural...