A5086038509- Topic Modeling
- Natural Language Processing Techniques
- Multimodal Machine Learning Applications
- Advanced Graph Neural Networks
- Explainable Artificial Intelligence (XAI)
- Surface Modification and Superhydrophobicity
- Text Readability and Simplification
- Domain Adaptation and Few-Shot Learning
- Supercapacitor Materials and Fabrication
- Graphene and Nanomaterials Applications
- Bone Tissue Engineering Materials
- Advanced battery technologies research
- Carbon and Quantum Dots Applications
- Speech and dialogue systems
- Advancements in Battery Materials
- Advanced Image and Video Retrieval Techniques
- Selenium in Biological Systems
- Material Properties and Applications
- Advanced Photocatalysis Techniques
- Spine and Intervertebral Disc Pathology
- Text and Document Classification Technologies
- Information Retrieval and Search Behavior
- Music and Audio Processing
- Nanoplatforms for cancer theranostics
- Education and Learning Interventions
Soochow University
2023-2025
Shaanxi Normal University
2025
Zhejiang University
2020-2024
Northwest Normal University
2022-2024
Harbin Institute of Technology
2024
Zhejiang University of Science and Technology
2020-2023
Alibaba Group (China)
2021-2023
Beijing University of Technology
2019
The emergence of large language models (LLMs) has marked a significant breakthrough in natural processing (NLP), fueling paradigm shift information acquisition. Nevertheless, LLMs are prone to hallucination, generating plausible yet nonfactual content. This phenomenon raises concerns over the reliability real-world retrieval (IR) systems and attracted intensive research detect mitigate such hallucinations. Given open-ended general-purpose attributes inherent LLMs, LLM hallucinations present...
The emergence of large language models (LLMs) has marked a significant breakthrough in natural processing (NLP), leading to remarkable advancements text understanding and generation. Nevertheless, alongside these strides, LLMs exhibit critical tendency produce hallucinations, resulting content that is inconsistent with real-world facts or user inputs. This phenomenon poses substantial challenges their practical deployment raises concerns over the reliability scenarios, which attracts...
Abstract Repair of infectious bone defects remains a serious problem in clinical practice owing to the high risk infection and excessive reactive oxygen species (ROS) during early stage, residual bacteria delayed Osseo integrated interface later which jointly creates complex dynamic microenvironment leads non‐union. The melatonin carbon dots (MCDs) possess antibacterial osteogenesis abilities, greatly simplifying composition multifunctional material. Therefore, hydrogel containing MCDs...
Intervertebral disc degeneration (IVDD) is a global public health issue. The injury of annulus fibrosus (AF) caused by acupuncture or discectomy can trigger IVDD again. However, there currently no suitable method for treating AF injury. In this study, the high-strength smart microneedles (MNs) which penetrate tissue through local and minimally invasive method, achieve remote control speeded-up release drug hyperthermia Near Infrared developed. PDA/GelMA composite MNs loaded with diclofenac...
Chain-of-thought reasoning, a cognitive process fundamental to human intelligence, has garnered significant attention in the realm of artificial intelligence and natural language processing. However, there still remains lack comprehensive survey for this arena. To end, we take first step present thorough research field carefully widely. We use X-of-Thought refer Chain-of-Thought broad sense. In detail, systematically organize current according taxonomies methods, including XoT construction,...
Abstract Stem cell implantation holds promise for enhancing bone repair, but risks of pathogen transmission and malignant transformation should not be ignored. Compared to stem implantation, recruitment endogenous cells injured sites is more critical in situ regeneration. In this study, based on the acidic microenvironment injury, an HG‐AA 1:1 ‐SDF‐1α composite hydrogel with a dual‐control intelligent switch function developed by incorporating stromal cell‐derived factor (SDF‐1α), arginine...
Morphology optimization of organic electrodes with more exposed redox-active sites and large contact area between active materials conductive carbon is crucial for higher actual capacity, improved rate performance, however, also challenging. In this work, we report a self-assembled perylene diimide disodium salts (PDI-ONa)-based cathode sodium batteries (SOBs) unveil an in situ electro-induced reorganization process that optimizes the nanostructured morphology significantly improves ionic...
The emerging pollutant selenium has an immeasurable impact on production and life, selective adsorption effective utilization of excessive in water bodies is very challenging. In this paper, nanoscale Al/Fe-MOF was obtained by situ modification Fe-MOF Al3+, Al/Fe-MOF-X/PAN nanofibrous membranes were prepared combining with electrostatic spinning technology for the enrichment Se(IV) bodies. experiments showed that Al/Fe-MOF-2/PAN had a high capacity good selectivity Se(IV), maximum up to...
In order to facilitate natural language understanding, the key is engage commonsense or background knowledge. However, how effectively in question answering systems still under exploration both research academia and industry. this paper, we propose a novel question-answering method by integrating multiple knowledge sources, i.e. ConceptNet, Wikipedia, Cambridge Dictionary, boost performance. More concretely, first introduce graph-based iterative retrieval module, which iteratively retrieves...
Path planning is a key part of ensuring that smart vehicles are safe and fast to reach their destination. Currently, global path mainly uses Astar algorithm for search planning. heuristic algorithm, which has been widely used in planning, but it many disadvantages, such as: low efficiency, poor real-time performance, there no constraints search. In view the shortcomings this paper presents Bidirectional search-Binary tree algorithm(BB-Astar), BB-Astar following improvements: First, binary...
Recent efforts of multimodal Transformers have improved Visually Rich Document Understanding (VrDU) tasks via incorporating visual and textual information. However, existing approaches mainly focus on fine-grained elements such as words document image patches, making it hard for them to learn from coarse-grained elements, including natural lexical units like phrases salient regions prominent regions. In this paper, we attach more importance containing high-density information consistent...