A5102812025- Quantum Chromodynamics and Particle Interactions
- Particle physics theoretical and experimental studies
- High-Energy Particle Collisions Research
- Black Holes and Theoretical Physics
- Dark Matter and Cosmic Phenomena
- Privacy-Preserving Technologies in Data
- Machine Learning and ELM
- Artificial Intelligence in Healthcare
- Chaos-based Image/Signal Encryption
- Digital Holography and Microscopy
- Chemical Synthesis and Analysis
- Digital Media Forensic Detection
- Mineralogy and Gemology Studies
- Complex Network Analysis Techniques
- Atomic and Subatomic Physics Research
- Enzyme Catalysis and Immobilization
- Machine Learning in Healthcare
- Domain Adaptation and Few-Shot Learning
- Clay minerals and soil interactions
- Image Processing Techniques and Applications
- Crystallography and Radiation Phenomena
- Neural Networks and Reservoir Computing
- Tactile and Sensory Interactions
- Crystal Structures and Properties
- Image and Signal Denoising Methods
Taiyuan University of Technology
2025
Institute of High Energy Physics
2004-2024
China University of Petroleum, East China
2024
Institute of Software
2024
Nanjing University of Science and Technology
2024
China Telecom
2023
China Telecom (China)
2023
Sports injuries can significantly impact athletes’ performance and career longevity, making their early prediction prevention a critical area of research. Traditional methods often fall short capturing the complex, nonlinear interactions between various risk factors that contribute to injuries. The sports is vital for well-being optimization athletes. This paper introduces Intrinsic Permutation Entropy Deep Learning (IPE-DL), novel framework synergizes permutation entropy with deep learning...
Session-based Recommendation (SR) aims to predict users' next click based on their behavior within a short period, which is crucial for online platforms. However, most existing SR methods somewhat ignore the fact that user preference not necessarily strongly related order of interactions. Moreover, they differences in importance between different samples, limits model-fitting performance. To tackle these issues, we put forward method, Mining Interest Trends and Adaptively Assigning Sample...
The decoder-only Transformer architecture with causal masking and relative position encoding (RPE) has become the de facto choice in language modeling. Despite its exceptional performance across various tasks, we have identified two limitations: First, it requires all attention scores to be non-zero sum up 1, even if current embedding sufficient self-contained information. This compels model assign disproportional excessive specific tokens. Second, RPE-based Transformers are not universal...
Community detection plays a pivotal role in uncovering closely connected subgraphs, aiding various real-world applications such as recommendation systems and anomaly detection. With the surge of rich information available for entities networks, community problem attributed networks has attracted widespread attention. While previous research effectively leveraged network topology attribute detection, these methods overlook two critical issues: (i) semantic similarity between node attributes...