A5100396049- Topic Modeling
- Natural Language Processing Techniques
- Structural Response to Dynamic Loads
- Data Quality and Management
- Advanced Text Analysis Techniques
- Transportation Safety and Impact Analysis
- Structural Behavior of Reinforced Concrete
- Smart Grid Energy Management
- Microgrid Control and Optimization
- Advanced Graph Neural Networks
- Multimodal Machine Learning Applications
- Optimal Power Flow Distribution
- Engineering and Test Systems
- Web Data Mining and Analysis
- Simulation and Modeling Applications
- Geotechnical Engineering and Underground Structures
- Material Properties and Processing
- Corrosion Behavior and Inhibition
- Text Readability and Simplification
- Semantic Web and Ontologies
- Cellular and Composite Structures
- Geophysical Methods and Applications
- High-Velocity Impact and Material Behavior
- Concrete Corrosion and Durability
- Energy Load and Power Forecasting
Institute of Computing Technology
2018-2024
Chinese Academy of Sciences
2015-2024
Shanghai Maritime University
2024
State Grid Corporation of China (China)
2023
Tsinghua University
2023
Southeast University
2014-2023
China Southern Power Grid (China)
2023
Southeast University
2023
PLA Army Engineering University
2013-2022
Zhongyuan University of Technology
2022
Large language models (LLMs) have revolutionized the field of artificial intelligence, enabling natural processing tasks that were previously thought to be exclusive humans. In this work, we introduce Qwen, first installment our large model series. Qwen is a comprehensive series encompasses distinct with varying parameter counts. It includes base pretrained models, and Qwen-Chat, chat finetuned human alignment techniques. The consistently demonstrate superior performance across multitude...
This report introduces the Qwen2 series, latest addition to our large language models and multimodal models. We release a comprehensive suite of foundational instruction-tuned models, encompassing parameter range from 0.5 72 billion, featuring dense Mixture-of-Experts model. surpasses most prior open-weight including its predecessor Qwen1.5, exhibits competitive performance relative proprietary across diverse benchmarks on understanding, generation, multilingual proficiency, coding,...
Entity resolution (ER) is a core problem of data integration. The state-of-the-art (SOTA) results on ER are achieved by deep learning (DL) based methods, trained with lot labeled matching/non-matching entity pairs. This may not be when using well-prepared benchmark datasets. Nevertheless, for many real-world applications, the situation changes dramatically, painful issue to collect large-scale In this paper, we seek answer: If have well-labeled source dataset, can train DL-based model target...
Abstract Fissured coal mass under triaxial unloading condition exhibits higher burst potential than the loading condition, which poses challenge to safety and productivity of resources extraction underground space utilization. To comprehensively understand mechanism unloading-induced during excavation process, this study investigated fracture energy evolution samples with different fissure types such as single, two parallel, coplanar-parallel using PFC 2D modelling. Triaxial tests were...
Relation extraction (RE) consistently involves a certain degree of labeled or unlabeled data even if under zero-shot setting. Recent studies have shown that large language models (LLMs) transfer well to new tasks out-of-the-box simply given natural prompt, which provides the possibility extracting relations from text without any and parameter tuning. This work focuses on study exploring LLMs, such as ChatGPT, relation extractors. On one hand, we analyze drawbacks existing RE prompts attempt...
In maritime logistics optimization, considerable research efforts are focused on the extraction of deep behavioral characteristics from comprehensive shipping data to discern patterns in vessel behavior. The effective linkage these with infrastructure, such as berths, is critical for enhancement ship navigation systems. This endeavor paramount not only a focus within information science but also progression intelligent Traditional methodologies have primarily emphasized analysis navigational...
Polyurea coating and carbon fibre reinforced polymer reinforcing techniques were applied to retrofit a severely damaged urban utility tunnel (UUT). The blast responses of the retrofitted UUT investigated through in-filed explosion experimental tests, displacements, strains, accelerations, damage compared. exhibited comparable or even better resistance than intact UUT. Although polyurea was flexible, it excellent performance in improving As with basalt reinforce (BFRP) bars has smaller...
With the continuous integration of renewable energy and complexity power system, dynamic voltage regulation low-voltage distribution network has become an urgent problem to be solved. Through in-depth analysis theoretical discussion, mathematical model is established, control strategy based on designed for regulation. simulation experiment field verification, superiority proposed method compared with existing demonstrated. In addition, sensitivity parameters stability are also...
In the field of military research, manufacturing and management weapons equipment are very important. Due to continuous advancement science technology, many databases have a loose structure, which makes them difficult be utilized efficiently, resulting in low efficiency, chaotic management, other issues. order solve these problems, an entity-relation extraction method based on CRF syntactic analysis tree is proposed according latest text algorithm. Finally, knowledge graph construction...
NER (Named Entity Recognition) is of great significance for the construction a knowledge map. The purpose to guarantee recognition effect named entity method in application scenario vertical field, proposed based on BI-LSTM-CRF [BI(Bidirectional) LSTM (Long-Short Term Memory) CRF (Conditional Random Field)] equipment support which improves domain and provides technical subsequent First, Chinese characters are represented by word embedding input into model. Then, feature vector sequence...