检索结果-内蒙古大学图书馆

Large language models based vulnerability detection: How does it enhance performance?

INTERNATIONAL JOURNAL OF INFORMATION SECURITY 2025年第1期24卷 1-18页

作者： Xuan, Cho Do Quang, Dat Bui Quang, Vinh Dang Posts & Telecommun Inst Technol Dept Informat Secur Hanoi Vietnam Posts & Telecommun Inst Technol Dept Informat Technol Hanoi Vietnam

Detecting vulnerabilities in C/C + + source code has become a critical challenge in information security, especially as the growing number and severity of new vulnerabilities increasingly impact organizations. In this context, Large Language Models (LLMs) have emerged as a promising approach;however, building a model capable of effectively predicting and classifying various types of vulnerabilities from diverse datasets remains a complex problem, demanding innovative and comprehensive solutions. Our research proposes a breakthrough approach by developing the FG-CVD ensemble learning model, an advanced architecture that combines code embedding techniques and Knowledge Argument to enhance feature representation and the ability to learn complex relationships within source code. These improvements are specifically designed on the foundation of code embedding and the transformer architecture of LLMs to boost the detection and classification of sophisticated vulnerability patterns. To evaluate the model's effectiveness, we conducted extensive experiments on four representative datasets: Reveal, BigVul, RealVul, and FFMQ + QEmu. The experimental results demonstrated FG-CVD's superior performance with an average accuracy of 85%, a prediction precision of 43%, a recall of 65%, and an F1-score of 47%. Notably, the model exhibited flexible adaptability to datasets with different structures and efficiently addressed data imbalance between labels. Moreover, through rigorous cross-dataset testing, the model showcased strong generalization capabilities and high stability, underscoring not only the academic value of the approach but also its practical potential, outperforming traditional approaches across a range of metrics and experimental scenarios.

关键词： Software vulnerability detection Embedding code transformer encoder Large language models Ensemble learning

来源：评论

学校读者我要写书评

暂无评论

引用

NEUROCOMPUTING 2025年 616卷

作者： Wu, Lei Qin, Qibing Hou, Jinkui Dai, Jiangyan Huang, Lei Zhang, Wenfeng Qufu Normal Univ Sch Comp Sci Rizhao Peoples R China Weifang Univ Sch Comp Engn Weifang Peoples R China Ocean Univ China Coll Informat Sci & Engn Qingdao Peoples R China Chongqing Normal Univ Coll Comp & Informat Sci Chongqing Peoples R China

Due to low storage cost and efficient retrieval advantages, hashing technologies have gained broad attention in the field of cross-modal retrieval in recent years. However, most current cross-modal hashing usually employs random sampling or semi-hard negative mining to construct training batches for model optimization, which ignores the distribution relationships between raw samples, generating redundant and unbalanced pairs, and resulting in sub-optimal embedding spaces. In this work, we address this dilemma with a novel deep cross- modal hashing framework, called Deep Multi-similarity Hashing via Label-Guided Networks (DMsH-LN), to learn a high separability public embedding space and generate discriminative binary descriptors. Specifically, by utilizing pair mining and weighting to jointly calculate self-similarity and relative similarity between pairs, the multi-similarity loss is extended to cross-modal hashing to alleviate the negative impacts caused by redundant and imbalanced samples on hash learning, enhancing the distinguishing ability of the obtained discrete codes. Besides, to capture fine-grained semantic supervised signals, the Label-guided Network is proposed to learn class-specific semantic signals, which could effectively guide the parameter optimization of the Image Network and Text Network. Extensive experiments are conducted on four benchmark datasets, which demonstrate that the DMsH-LN framework achieves excellent retrieval performance. The source codes of DMsH-LN are downloaded from https://***/QinLab-WFU/DMsH-LN.

关键词： Cross-modal retrieval Deep hashing Random sampling Multi-similarity loss Label-guided network transformer encoder

来源：评论

学校读者我要写书评

暂无评论

Enhancing Drug Peptide Sequence Prediction Using Multi-view Feature Fusion Learning

引用

CURRENT BIOINFORMATICS 2025年第3期20卷 276-287页

作者： Zhang, Junyu Lu, Ronglin Zhou, Hongmei Jiang, Xinbo Shandong Univ Sch Software Jinan 250101 Peoples R China Shandong Univ Joint SDU NTU Ctr Artificial Intelligence Res FAIR Jinan 250101 Peoples R China Beidahuang Ind Grp Gen Hosp Harbin Peoples R China Shandong Univ Sch Qilu Transportat Jinan 250061 Peoples R China

Background Currently, various types of peptides have broad implications for human health and disease. Some drug peptides play significant roles in sensory science, drug research, and cancer biology. The prediction and classification of peptide sequences are of significant importance to various industries. However, predicting peptide sequences through biological experiments is a time-consuming and expensive process. Moreover, the task of protein sequence classification and prediction faces challenges due to the high dimensionality, nonlinearity, and irregularity of protein sequence data, along with the presence of numerous unknown or unlabeled protein sequences. Therefore, an accurate and efficient method for predicting peptide category is *** In our work, we used two pre-trained models to extract sequence features, TextCNN (Convolutional Neural Networks for Text Classification) and transformer. We extracted the overall semantic information of the sequences using transformer encoder and extracted the local semantic information between sequences using TextCNN and concatenated them into a new feature. Finally, we used the concatenated feature for classification prediction. To validate this approach, we conducted experiments on the BP dataset, THP dataset and DPP-IV dataset and compared them with some pre-trained *** Since TextCNN and transformer encoder extract features from different perspectives, the concatenated feature contains multi-view information, which improves the accuracy of the peptide *** Ultimately, our model demonstrated superior metrics, highlighting its efficacy in peptide sequence prediction and classification.

关键词： Drug peptide sequence multi-view feature fusion learning TextCNN transformer encoder

来源：评论

学校读者我要写书评

暂无评论

EpiBrCan-Lite: A lightweight deep learning model for breast cancer subtype classification using epigenomic data

引用

COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE 2025年 260卷 108553-108553页

作者： Bedi, Punam Rani, Surbhi Gupta, Bhavna Bhasin, Veenu Gole, Pushkar Univ Delhi Dept Comp Sci Delhi India Univ Delhi Keshav Mahavidyalaya New Delhi India Univ Delhi PGDAV Coll New Delhi India

Background and objectives: Early breast cancer subtypes classification improves the survival rate as it facilitates prognosis of the patient. In literature this problem was prominently solved by various Machine Learning and Deep Learning techniques. However, these studies have three major shortcomings: huge Trainable Weight Parameters (TWP), suffer from low performance and class imbalance problem. Methods: This paper proposes a lightweight model named EpiBrCan-Lite for classifying breast cancer subtypes using DNA methylation data. This model encompasses three blocks namely Data Encoding, TransGRU, and Classification blocks. In Data Encoding block, the input features are encoded into equal sized chunks and then passed down to TransGRU block which is a modified version of traditional transformer encoder (TE). In TransGRU block, MLP module of traditional TE is replaced by GRU module, consisting of two GRU layers to reduce TWP and capture the long-range dependencies of input feature data. Furthermore, output of TransGRU block is passed to Classification block for classifying breast cancer into their subtypes. Results: The proposed model is validated using Accuracy, Precision, Recall, F1-score, FPR, and FNR metrics on TCGA breast cancer dataset. This dataset suffers from the class imbalance problem which is mitigated using Synthetic Minority Oversampling Technique (SMOTE). Experimentation results demonstrate that EpiBrCan-Lite model attained 95.85 % accuracy, 95.96 % recall, 95.85 % precision, 95.90 % F1-score, 1.03 % FPR, and 4.12 % FNR despite of utilizing only 1/1500 of TWP than other state-of-the-art models. Conclusion: EpiBrCan-Lite model is efficiently classifying breast cancer subtypes, and being lightweight, it is suitable to be deployed on low computational powered devices.

关键词： Breast cancer disease transformer encoder Gated recurrent unit Epigenomic DNA methylation data SMOTE

来源：评论

学校读者我要写书评

暂无评论

Lower Limb Motion Recognition Based on sEMG and CNN-TL Fusion Model

引用

SENSORS 2024年第21期24卷 7087页

作者： Zhou, Zhiwei Tao, Qing Su, Na Liu, Jingxuan Chen, Qingzheng Li, Bowen Xinjiang Univ Coll Intelligent Mfg Modern Ind Urumqi 830017 Peoples R China Xinjiang Med Univ Affiliated Hosp 1 Urumqi 830017 Peoples R China

To enhance the classification accuracy of lower limb movements, a fusion recognition model integrating a surface electromyography (sEMG)-based convolutional neural network, transformer encoder, and long short-term memory network (CNN-transformer-LSTM, CNN-TL) was proposed in this study. By combining these advanced techniques, significant improvements in movement classification were achieved. Firstly, sEMG data were collected from 20 subjects as they performed four distinct gait movements: walking upstairs, walking downstairs, walking on a level surface, and squatting. Subsequently, the gathered sEMG data underwent preprocessing, with features extracted from both the time domain and frequency domain. These features were then used as inputs for the machine learning recognition model. Finally, based on the preprocessed sEMG data, the CNN-TL lower limb action recognition model was constructed. The performance of CNN-TL was then compared with that of the CNN, LSTM, and SVM models. The results demonstrated that the accuracy of the CNN-TL model in lower limb action recognition was 3.76%, 5.92%, and 14.92% higher than that of the CNN-LSTM, CNN, and SVM models, respectively, thereby proving its superior classification performance. An effective scheme for improving lower limb motor function in rehabilitation and assistance devices was thus provided.

关键词： surface electromyography signals lower limb action recognition convolutional neural network transformer encoder long short-term memory

来源：评论

学校读者我要写书评

暂无评论

ClKI: closed-loop and knowledge iterative via self-distillation for image sentiment analysis

引用

INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS 2024年第7期15卷 2843-2862页

作者： Zhang, Hongbin Yuan, Meng Hu, Lang Wang, Wengang Li, Zhijie Ye, Yiyuan Ren, Yafeng Ji, Donghong East China Jiaotong Univ Sch Software Nanchang Peoples R China East China Jiaotong Univ Sch Informat Engn Nanchang Peoples R China Guangdong Univ Foreign Studies Sch Interpreting & Translat Studies Guangzhou Peoples R China Wuhan Univ Sch Cyber Sci & Engn Wuhan Peoples R China

Image sentiment analysis has received a lot of attention due to the fact that users prefer to express their private emotions using images on social platforms. Researchers strive to obtain more discriminative features to characterize these images by designing diverse networks, such as convolutional neural networks (CNNs) and transformer. This method requires prior knowledge in advance and its feature fusion ignores the correlation between heterogeneous networks. An efficient but effective integration of heterogeneous transformer and CNN deserves further study. To this end, we propose a novel model called Closed-loop and Knowledge Iterative (ClKI) via self-distillation for image sentiment analysis. First, CNN features at different scales are extracted to capture the complementary multi-scale information, and then input into a transformer stacked by multiple encoders with the ProbSparse self-attention to acquire more discriminative representations. Second, we adopt the true labels and deepest classification results jointly guide the training procedure of shallow modules via self-distillation. Thus, the ClKI model achieves an efficient interaction between the deep and shallow modules, providing sufficient knowledge for image sentiment analysis. Our model achieves 83.35%, 65.64%, and 91.01% accuracy on the FI, Emotion 6, and Twitter I datasets respectively, which outperforms state-of-the-art methods. Additional experiments further validate that ClKI can resolve the data imbalance problem to some degree and it has good generalization ability. Notably, ClKI is plug-and-play, promoting its practicality. Briefly, this study provides an innovative approach in image sentiment analysis, offering valuable insights for the integration of heterogeneous networks as well as knowledge iterative method.

关键词： Image sentiment analysis Closed-loop and knowledge iterative Self-distillation Multi-scale features Convolutional neural network transformer encoder

来源：评论

学校读者我要写书评

暂无评论

TEC-miTarget: enhancing microRNA target prediction based on deep learning of ribonucleic acid sequences

引用

BMC BIOINFORMATICS 2024年第1期25卷 1页

作者： Yang, Tingpeng Wang, Yu He, Yonghong Peng Cheng Lab Shenzhen 518055 Peoples R China Tsinghua Shenzhen Int Grad Sch Shenzhen 518055 Peoples R China

Background MicroRNAs play a critical role in regulating gene expression by binding to specific target sites within gene transcripts, making the identification of microRNA targets a prominent focus of research. Conventional experimental methods for identifying microRNA targets are both time-consuming and expensive, prompting the development of computational tools for target prediction. However, the existing computational tools exhibit limited performance in meeting the demands of practical applications, highlighting the need to improve the performance of microRNA target prediction *** In this paper, we utilize the most popular natural language processing and computer vision technologies to propose a novel approach, called TEC-miTarget, for microRNA target prediction based on transformer encoder and convolutional neural networks. TEC-miTarget treats RNA sequences as a natural language and encodes them using a transformer encoder, a widely used encoder in natural language processing. It then combines the representations of a pair of microRNA and its candidate target site sequences into a contact map, which is a three-dimensional array similar to a multi-channel image. Therefore, the contact map's features are extracted using a four-layer convolutional neural network, enabling the prediction of interactions between microRNA and its candidate target sites. We applied a series of comparative experiments to demonstrate that TEC-miTarget significantly improves microRNA target prediction, compared with existing state-of-the-art models. Our approach is the first approach to perform comparisons with other approaches at both sequence and transcript levels. Furthermore, it is the first approach compared with both deep learning-based and seed-match-based methods. We first compared TEC-miTarget's performance with approaches at the sequence level, and our approach delivers substantial improvements in performance using the same datasets and evaluation metrics. Moreover, w

关键词： MicroRNAs miRNA targets Target prediction Deep learning transformer encoder Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Fast-Focused-Net: Enhancing Small Object Encoding With VDP Layer in Deep Neural Networks

引用

IEEE ACCESS 2024年 12卷 130603-130616页

作者： Ali, Tofik Pratim Roy, Partha Saini, Rajkumar Indian Inst Technol Roorkee Dept Comp Sci & Engn Roorkee 247667 India Lulea Univ Technol Dept Comp Sci Elect & Space Engn S-97187 Lulea Sweden

In this paper, we introduce Fast&Focused-Net (FFN), a novel deep neural network architecture tailored for efficiently encoding small objects into fixed-length feature vectors. Contrary to conventional Convolutional Neural Networks, FFN employs a series of our newly proposed layer, the Volume-wise Dot Product (VDP) layer, designed to address several inherent limitations of CNNs. Specifically, CNNs exhibit a smaller effective receptive field (ERF) than their theoretical counterparts, limiting their vision span. Additionally, the initial layers in CNNs produce low-dimensional feature vectors, presenting a bottleneck for subsequent learning. Lastly, the computational overhead of CNNs, particularly in capturing diverse image regions by parameter sharing, is significantly high. The VDP layer, at the heart of FFN, aims to remedy these issues by efficiently covering the entire image patch information with reduced computational. Experimental results demonstrate the prowess of FFN in a variety of applications. Our network outperformed state-of-the-art methods for small object classification tasks on datasets such as CIFAR-10, CIFAR-100, STL-10, SVHN-Cropped, and Fashion-MNIST. In the context of larger image classification, when combined with a transformer encoder (ViT), FFN produced competitive results for OpenImages V6, ImageNet-1K, and Places365 datasets. Moreover, the same combination showcased unparalleled performance in text recognition tasks across SVT, IC15, SVTP, and HOST datasets.

关键词： Computer architecture Radio frequency Computational efficiency Convolutional neural networks Encoding Text recognition Image analysis Small object classification efficient image encoding convolutional neural network transformer encoder text recognition

来源：评论

学校读者我要写书评

暂无评论

UTran-DSR: a novel transformer-based model using feature enhancement for dysarthric speech recognition

引用

EURASIP JOURNAL ON AUDIO SPEECH AND MUSIC PROCESSING 2024年第1期2024卷 54页

作者： Irshad, Usama Mahum, Rabbia Ganiyu, Ismaila Butt, Faisal Shafique Hidri, Lotfi Ali, Tamer G. El-Sherbeeny, Ahmed M. Univ Engn & Technol Taxila Dept Comp Sci Taxila 47050 Pakistan COMSATS Univ Islamabad Dept Comp Sci Wah Campus Wah Cantt Pakistan King Saud Univ Coll Engn Ind Engn Dept POB 800 Riyadh 11421 Saudi Arabia King Salman Ctr Disabil Res POB 94682 Riyadh 11614 Saudi Arabia

Over the past decade, the prevalence of neurological diseases has significantly risen due to population growth and aging. Individuals suffering from spastic paralysis, brain attack, and idiopathic Parkinson's disease (PD), among other neurological illnesses, commonly suffer from dysarthria. Early detection and treatment of dysarthria in these patients are essential for effectively managing the progression of their disease. This paper provides UTrans-DSR, a novel encoder-decoder architecture for analyzing Mel-spectrograms (generated from audios) and classifying speech as healthy or dysarthric. Our model employs transformer encoder features based on a hybrid design, which includes the feature enhancement block (FEB) and the vision transformer (ViT) encoders. This combination effectively extracts global and local pixel information regarding localization while optimizing the mel-spectrograms feature extraction process. We keep up with the original class-token grouping sequence in the vision transformer while generating a new equivalent expanding route. More specifically, two unique growing pathways use a deep-supervision approach to increase spatial data recovery and expedite model convergence. We add consecutive residual connections to the system to reduce feature loss while increasing spatial data retrieval. Our technique is based on identifying gaps in mel-spectrograms distinguishing between normal and dysarthric speech. We conducted several experiments on UTrans-DSR using the UA speech and TORGO datasets, and it outperformed the existing top models. The model performed significantly in pixel's localized and spatial feature extraction, effectively detecting and classifying spectral gaps. The Tran-DSR model outperforms previous research models, achieving an accuracy of 97.75%.

关键词： Dysarthria speech recognition transformer encoder Deep learning Vision transformer Feature enhancement block

来源：评论

学校读者我要写书评

暂无评论

Graph neural collaborative filtering with medical content-aware pre-training for treatment pattern recommendation

引用

PATTERN RECOGNITION LETTERS 2024年 185卷 210-217页

作者： Min, Xin Li, Wei Han, Ruiqi Ji, Tianlong Xie, Weidong Northeastern Univ Sch Comp Sci & Engn Shenyang 11000 Peoples R China Northeastern Univ Key Lab Intelligent Comp Med Image MIIC Shenyang 11000 Peoples R China China Med Univ Hosp 1 Dept Radiat Oncol Shenyang 11000 Peoples R China

Recently, considering the advancement of information technology in healthcare, electronic medical records (EMRs) have become the repository of patients' treatment processes in hospitals, including the patient's treatment pattern (standard treatment process), the patient's medical history, the patient's admission diagnosis, etc. In particular, EMRs-based treatment recommendation systems have become critical for optimizing clinical decision-making. EMRs contain complex relationships between patients and treatment patterns. Recent studies have shown that graph neural collaborative filtering can effectively capture the complex relationships in EMRs. However, none of the existing methods take into account the impact of medical content such as the patient's admission diagnosis, and medical history on treatment recommendations. In this work, we propose a graph neural collaborative filtering model with medical content-aware pre-training (CAPRec) for learning initial embeddings with medical content to improve recommendation performance. First the model constructs a patient-treatment pattern interaction graph from EMRs data. Then we attempt to use the medical content for pre-training learning and transfer the learned embeddings to a graph neural collaborative filtering model. Finally, the learned initial embedding can support the downstream task of graph collaborative filtering. Extensive experiments on real world datasets have consistently demonstrated the effectiveness of the medical content-aware training framework in improving treatment recommendations.

关键词： Graph neural collaborative filtering Medical content-aware Pre-training generative transformer encoder

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：