A5100570223- Advanced Graph Theory Research
- Advancements in Battery Materials
- Recommender Systems and Techniques
- Graph Labeling and Dimension Problems
- Advanced Graph Neural Networks
- Hepatocellular Carcinoma Treatment and Prognosis
- Computational Geometry and Mesh Generation
- Adversarial Robustness in Machine Learning
- Parallel Computing and Optimization Techniques
- Advanced Neural Network Applications
- Multimodal Machine Learning Applications
- IL-33, ST2, and ILC Pathways
- IoT and Edge/Fog Computing
- Macrophage Migration Inhibitory Factor
- Safety and Risk Management
- GDF15 and Related Biomarkers
- Domain Adaptation and Few-Shot Learning
- Sepsis Diagnosis and Treatment
- Complex Network Analysis Techniques
- Tropical and Extratropical Cyclones Research
- Liver Disease Diagnosis and Treatment
- Supply Chain and Inventory Management
- Nosocomial Infections in ICU
- Ship Hydrodynamics and Maneuverability
- Technology and Security Systems
Tongji University
2023-2025
Materials Science & Engineering
2025
Duke University
2022-2025
Jiangxi University of Traditional Chinese Medicine
2023-2024
East China Normal University
2022-2024
Qilu University of Technology
2024
Qilu Hospital of Shandong University
2024
University of Jinan
2024
Wuhan Institute of Technology
2023
Peking University
2023
Continual learning (CL) can help pre-trained vision-language models efficiently adapt to new or under-trained data distributions without re-training. Nevertheless, during the continual training of Contrastive Language-Image Pre-training (CLIP) model, we observe that model's zero-shot transfer ability significantly degrades due catastrophic forgetting. Existing CL methods mitigate forgetting by replaying previous data. However, since CLIP dataset is private, replay cannot access pre-training...
In the rapidly evolving domain of artificial intelligence, Large Language Models (LLMs) play a crucial role due to their advanced text processing and generation abilities. This study introduces new strategy aimed at harnessing on-device LLMs in invoking software APIs. We meticulously compile dataset derived from API documentation apply fine-tuning with capacities 2B, 3B 7B parameters, specifically enhance proficiency interactions. Our approach concentrates on refining models' grasp...
Introduction Macrophage-mediated inflammatory response may have crucial roles in the pathogenesis of a variety human diseases. Growth differentiation factor 15 (GDF15) is cytokine transforming growth factor-β superfamily, with potential anti-inflammatory activities. Previous studies observed lungs some macrophages which expressed high level GDF15. Methods In present study, we employed multiple techniques, including immunofluorescence, flow cytometry, and single-cell RNA sequencing, order to...
The exponential growth of “3C” (computer, communication, and consumer electronics) market generates an ever-increasing demand on recycling materials from the end-of-life LiCoO2 (LCO) batteries with ecological efficient methods. Herein we present a direct scalable approach to recycle degraded LCO by healing stabilizing their damaged structure via solid reactions. as-proposed constructs protective layer onto regenerated particles suppress O3 H1-3 phase transition at 4.55 V. Benefiting unique...
To investigate drug resistance of Acinetobacter baumannii and its related factors in intensive care unit (ICU), to provide clinical basis for prevention treatment.A retrospective analysis was conducted. Clinical data collected from 1 050 patients ICU Foshan Hospital Traditional Chinese Medicine January 2011 June 2013. The risk nosocomial infection were analyzed with univariate analysis. independent factor sieved the P<0.05 unconditional logistic regression analyze ICU.One hundred thirteen...
The rapid development of large language models (LLMs) has significantly transformed the field artificial intelligence, demonstrating remarkable capabilities in natural processing and moving towards multi-modal functionality. These are increasingly integrated into diverse applications, impacting both research industry. However, their deployment present substantial challenges, including need for extensive computational resources, high energy consumption, complex software optimizations. Unlike...
ABSTRACT Sepsis-induced myocardial dysfunction (SIMD) contributes significantly to cardiovascular during septic shock. We aimed evaluate the potential role of Xinmailong injection (XMLI), a polypeptide medicine extracted from Periplaneta americana , in reversing progression damage SIMD sepsis patients. This was multicenter, randomized, double-blind, parallel-group trial. recruited all patients consecutively admitted intensive care units (ICUs) who were aged 18 85 years old and met 3.0...
Accurate and efficient on-chip power modeling is crucial to runtime power, energy, voltage management. Such monitoring can be achieved by designing integrating meters (OPMs) into the target design. In this work, we propose a new method named DEEP automatically develop extremely OPM solutions for given selects inputs from all individual bits in RTL signals. bit-level selection provides an unprecedentedly large number of input candidates supports lower hardware cost, compared with signal-level...
Virtual Synchronous Generator (VSG) can effectively mitigate the inertia drop caused by grid connection of renewable energy, but VSG is prone to generate transient current at moment connection, resulting in switchover failure. This paper presents a pre-synchronization method for off-grid switching. First, control principle analyzed. Secondly, an improved PLL-free strategy proposed, which eliminates PLL and simplifies structure. Finally, correctness proposed verified simulation results.
Purpose Lymph node (LN) involvement is a key factor in ovarian clear cell carcinoma (OCCC) although, there several indicators can be used to define prognosis. This study examines the prognostic performances of each indicator for OCCC patients by comparing number lymph nodes examined (TNLE), positive (PLN), ratio (LNR), and log odds metastatic (LODDS). Methods 1,300 who underwent lymphadenectomy between 2004 2015 were extracted from Surveillance Epidemiology End Results (SEER) database....
Continual learning (CL) can help pre-trained vision-language models efficiently adapt to new or under-trained data distributions without re-training. Nevertheless, during the continual training of Contrastive Language-Image Pre-training (CLIP) model, we observe that model's zero-shot transfer ability significantly degrades due catastrophic forgetting. Existing CL methods mitigate forgetting by replaying previous data. However, since CLIP dataset is private, replay cannot access pre-training...
Abstract Macrophage-mediated inflammatory response may have crucial roles in the pathogenesis of a variety human diseases. Growth differentiation factor 15 (GDF15) is cytokine transforming growth factor-β superfamily, with potential anti-inflammatory activities. Previous studies observed lungs some macrophages which expressed high level GDF15. In present study, we employed multiple techniques, including immunofluorescence, flow cytometry, and single-cell RNA sequencing, order to further...
Non-alcoholic fatty liver disease (NAFLD) is increasingly recognized as a global public health concern characterized by excessive lipid accumulation in the liver. Ferulic acid (FA), common foodborne phenolic acid, has been shown to reduce activating adenosine monophosphate-activated protein kinase (AMPK) and suppressing expression of sterol regulatory element-binding protein-1 (SREBP-1). In this study, we synthesized new class (E)-3-(3-methoxy-4-substituted phenyl)-acrylic derivatives AMPK...
The demand for improving the efficiency and accuracy of handling classification logistics products has increased due to rapid development e-commerce express, new energy vehicles, intelligent factories, other fields. In this paper, in response above needs, we design produce a vehicle that can realise identification tracking. overall vehicle, mechanical structure design, control system algorithm are introduced. vehicle's main frame is constructed using aluminium profiles. motor's differential...