A5110112480- Topic Modeling
- Speech Recognition and Synthesis
- Synthetic Aperture Radar (SAR) Applications and Techniques
- Music and Audio Processing
- Computational and Text Analysis Methods
- Video Analysis and Summarization
- Natural Language Processing Techniques
- Speech and Audio Processing
- Intelligent Tutoring Systems and Adaptive Learning
- Text and Document Classification Technologies
- Sentiment Analysis and Opinion Mining
- Advanced Text Analysis Techniques
- Meteorological Phenomena and Simulations
- Adversarial Robustness in Machine Learning
- Geophysical Methods and Applications
- Atmospheric aerosols and clouds
- Climate variability and models
- Soil Moisture and Remote Sensing
Minzu University of China
2025
Aalborg University
2014-2025
Wuhan University
2025
China Meteorological Administration
2022
Liaoning Meteorological Bureau
2022
Automatic speech recognition (ASR) systems are known to be vulnerable adversarial attacks. This paper addresses detection and defence against targeted white-box attacks on signals for ASR systems. While existing work has utilised diffusion models (DMs) purify examples, achieving state-of-the-art results in keyword spotting tasks, their effectiveness more complex tasks such as sentence-level remains unexplored. Additionally, the impact of number forward steps performance is not well...
With the advancement of pre-trained language models, prompt learning has emerged as a trend for text classification. It offers several advantages over traditional machine methods, particularly low-resource natural processing tasks. Prompt enables fine-tuning models on relatively small datasets, eliminating need large number expensive labeled samples. This paper proposes an effective approach that combines and ensemble learning, aiming to enhance performance individual in Tibetan...
This study presents a framework for automated evaluation of dynamically evolving topic taxonomies in scientific literature using Large Language Models (LLMs). In digital library systems, modeling plays crucial role efficiently organizing and retrieving scholarly content, guiding researchers through complex knowledge landscapes. As research domains proliferate shift, traditional human centric static methods struggle to maintain relevance. The proposed approach harnesses LLMs measure key...
<title>Abstract</title> As the probability and intensity of global drought events continue to increase, soil moisture is an important basis for monitoring assessment, need accurately obtain distribution information with high spatial-temporal resolution becoming extremely important. CYGNSS data based on spaceborne GNSS-R has advantage temporal resolution, while SAR can provide surface features spatial combination two provides favourable conditions obtaining resolution. This paper proposes a...
Accurate detection of non-linguistic vocal events in social signals can have a great impact on the applicability speech enabled interactive systems. In this paper, we investigate use random forest for event detection. Random technique has been successfully employed many areas such as object detection, face recognition, and audio This paper proposes to online detecting laughter filler analyzing importance various features classification through permutation. The results show that according...
Assessing students' answers is a labor-intensive task that could significantly burden educators. Technological advances have led to the development of automatic scoring systems. However, handwriting recognition errors impacts deployment in context paper-pencil assessment, which still most widely used format academic exams. Large language models (LLMs) proven be effective tasks and are promising for short answer grading. Nevertheless, LLMs' capability grade with underexplored. The current...
Blizzard is a severe weather-related disaster with significant environmental, economic, and social impacts. What worse, the blizzard increasingly frequent along climate change. To enhance resilience, it important to accurately estimate blizzards. While some studies have reported formation characteristics, limited not well presented backflow associated cold vortex in Northeast China. In order overcome this research gap, study aims analyze characteristics of Liaoning, China, reveal reasons...