检索结果-内蒙古大学图书馆

31st International Conference on Computational Linguistics, COLING 2025

作者： Peng, Xingyu Wu, Junran Liu, Ruomei Xu, Ke State Key Laboratory of Complex & Critical Software Environment Beihang University Beijing China

ISBN: (纸本)9798891761964

Traditional methods for detecting rumors on social media primarily focus on analyzing textual content, often struggling to capture the complexity of online interactions. Recent research has shifted towards leveraging graph neural networks to model the hierarchical conversation structure that emerges during rumor propagation. However, these methods tend to overlook the temporal aspect of rumor propagation and may disregard potential noise within the propagation structure. In this paper, we propose a novel approach that incorporates temporal information by constructing a weighted propagation tree, where the weight of each edge represents the time interval between connected posts. Drawing upon the theory of structural entropy, we transform this tree into a coding tree. This transformation aims to preserve the essential structure of rumor propagation while reducing noise. Finally, we introduce a recursive neural network to learn from the coding tree for rumor veracity prediction. Experimental results on two common datasets demonstrate the superiority of our approach. © 2025 Association for Computational Linguistics.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Complementary Learning System Theory-based Active Learning for Audio Classification

Complementary Learning System Theory-based Active Learning f...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Geng, Hui Gao, Zijian Wan, Tianjiao Feng, Dawei Wang, Changjian Xu, Kele College of Computer Science and Technology National University of Defense Technology Changsha China State Key Laboratory of Complex & Critical Software Environment Changsha China

ISBN: (纸本)9798350368741

Deep learning has significantly advanced the audio classification, achieving remarkable results. However, these successes often rely on extensive manual annotation of audio, a labor-intensive and costly process. Active Learning (AL) presents a promising solution by minimizing the required amount of annotation through the iterative selection of the most informative audio samples. Current AL methods for audio classification typically depend solely on the latest model checkpoint, overlooking the dynamics of the entire training process. The Complementary Learning Systems (CLS) theory posits that the interplay between short-term and long-term memory systems can effectively measure sample uncertainty, offering a means to capture training dynamics. In this work, we introduce a novel AL framework for audio classification, termed CLS-AL, which addresses the limitations of existing methods by simultaneously maintaining both short-term and long-term memory models. This dual-memory approach allows for a more comprehensive consideration of training dynamics. The divergence in predictions between these memory models provides a new metric for evaluating the uncertainty of unlabeled samples, enhancing the effectiveness of the AL sample selection process. We demonstrate the effectiveness and generalizability of CLS-AL through extensive experiments on a diverse set of audio datasets, showing that CLS-AL obviously outperforms existing state-of-the-art methods. © 2025 IEEE.

关键词： Active Learning Audio Classification Complementary Learning System Uncertainty Estimation

来源：评论

学校读者我要写书评

暂无评论

On the Applicability of Code Language Models to Scientific Computing Programs

引用

IEEE Transactions on software Engineering 2025年

作者： Zhao, Qianhui Liu, Fang Long, Xiao Wu, Chengru Zhang, Li State Key Laboratory of Complex & Critical Software Environment Beihang University Beijing China

Scientific Computing Programming Languages (SCPLs), like MATLAB and R, are popular and widely used for computational mathematics. In recent years, pre-trained code language models (CLMs) have automated many code-related tasks, covering various general programming languages. SCPLs share many similarities with general programming languages, including similar syntactic structures and the semantics of identifiers. Despite the similarities, there exist many differences between them. For example, lots of numerical operations and dedicated libraries exist in SCPLs. However, there has been little comprehensive work analyzing CLMs’ capabilities in the understanding and generation of pragmatic scientific computing programs. To this end, we investigate the applicability of code language models for the SCPL analysis, especially focus on real-world code in open-source repositories. We first create a benchmark that contains programs and documentation from three widely used scientific computing programming languages, then perform an adequate evaluation of existing advanced code language models on both code understanding and generation tasks using the new benchmark, and study the relations of different training strategies, model types, and model sizes to the performance of different tasks and languages. Evaluation results confirm that, compared to general programming languages, SCPLs are more challenging to understand, and especially to generate, but the use of code language models is nevertheless feasible, and the knowledge obtained from the general languages can be transferred to SCPL analysis. A deeper analysis reveals additional challenges in generating code that incorporates API calls relevant to computational mathematics. We believe that our findings can provide guidance on improving tooling and analyses for the scientific programming language, and also inspire and motivate researchers to improve the robustness of existing code language models. © 1976-2012 IEEE.

关键词： System program documentation

来源：评论

学校读者我要写书评

暂无评论

Deep learning-based software engineering: progress,challenges, and opportunities

引用

Science China(Information Sciences) 2025年第1期68卷 57-144页

作者： Xiangping CHEN Xing HU Yuan HUANG He JIANG Weixing JI Yanjie JIANG Yanyan JIANG Bo LIU Hui LIU Xiaochen LI Xiaoli LIAN Guozhu MENG Xin PENG Hailong SUN Lin SHI Bo WANG Chong WANG Jiayi WANG Tiantian WANG Jifeng XUAN Xin XIA Yibiao YANG Yixin YANG Li ZHANG Yuming ZHOU Lu ZHANG School of Journalism and Communication Sun Yat-sen University School of Software Technology Zhejiang University School of Software Engineering Sun Yat-sen University School of Software Dalian University of Technology School of Computer Science and Technology Beijing Institute of Technology Key Laboratory of High Confidence Software Technologies (Peking University) Ministry of EducationSchool of Computer Science Peking University State Key Laboratory for Novel Software Technology Nanjing University School of Computer Science and Engineering Beihang University Institute of Information Engineering Chinese Academy of Sciences School of Computer Science Fudan University State Key Laboratory of Complex & Critical Software Environment (CCSE) School of Software Beihang University School of Computer and Information Technology Beijing Jiaotong University School of Computer Science and Technology Harbin Institute of Technology School of Computer Science Wuhan University Huawei Technologies

Researchers have recently achieved significant advances in deep learning techniques, which in turn has substantially advanced other research disciplines, such as natural language processing, image processing, speech recognition, and software engineering. Various deep learning techniques have been successfully employed to facilitate software engineering tasks, including code generation, software refactoring, and fault localization. Many studies have also been presented in top conferences and journals, demonstrating the applications of deep learning techniques in resolving various software engineering tasks. However,although several surveys have provided overall pictures of the application of deep learning techniques in software engineering,they focus more on learning techniques, that is, what kind of deep learning techniques are employed and how deep models are trained or fine-tuned for software engineering tasks. We still lack surveys explaining the advances of subareas in software engineering driven by deep learning techniques, as well as challenges and opportunities in each subarea. To this end, in this study, we present the first task-oriented survey on deep learning-based software engineering. It covers twelve major software engineering subareas significantly impacted by deep learning techniques. Such subareas spread out through the whole lifecycle of software development and maintenance, including requirements engineering, software development, testing, maintenance, and developer collaboration. As we believe that deep learning may provide an opportunity to revolutionize the whole discipline of software engineering, providing one survey covering as many subareas as possible in software engineering can help future research push forward the frontier of deep learning-based software engineering more systematically. For each of the selected subareas,we highlight the major advances achieved by applying deep learning techniques with pointers to the available datasets i

关键词： deep learning software engineering software benchmark software artifact representation survey

来源：评论

学校读者我要写书评

暂无评论

SSAST-Adapter: A Parameter-efficient Incremental Learning Algorithm for Underwater Acoustic Target Recognition

SSAST-Adapter: A Parameter-efficient Incremental Learning Al...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Zhu, Qian Xu, Qisheng Zhu, Boqing Gao, Zijian Zeng, Lingbin Xu, Kele College of Computer Science and Technology National University of Defense Technology Changsha China State Key Laboratory of Complex & Critical Software Environment Changsha China College of Meteorology and Oceanography National University of Defense Technology Changsha China

ISBN: (纸本)9798350368741

Underwater acoustic target recognition involves identifying and classifying targets in underwater environments using acoustic signals. In recent years, deep learning has made significant progress in this field. However, the models require the entire dataset to be available upfront, and classification categories must be predefined. In practical scenarios, new objects or species may appear in underwater environments over time, making it impractical to retrain a model from scratch each time new data is introduced. At the same time, training on new data inevitably leads to catastrophic forgetting of past data. To address these challenges, we propose a large-scale pre-training strategy combined with adapters to enable incremental learning without the need for complete retraining. To minimize the number of parameters during model fine-tuning, we employ a adapter structure at various network layers, reducing the number of trainable parameters to less than 2%. Experimental results demonstrate that our proposed method effectively reduces the need for full retraining by allowing the model to update in a resource-efficient manner using only new data, and the model has achieved high recognition accuracy in underwater acoustic target recognition tasks. © 2025 IEEE.

关键词： Adapter Audio Spectrum Transformer Incremental Learning Underwater Acoustics Target Recognition

来源：评论

学校读者我要写书评

暂无评论

Pushing the Limit of Post-Training Quantization

引用

IEEE Transactions on Pattern Analysis and Machine Intelligence 2025年 PP卷 PP页

作者： Gong, Ruihao Liu, Xianglong Li, Yuhang Fan, Yunqiang Wei, Xiuying Guo, Jinyang Beihang University State Key Laboratory of Complex & Critical Software Environment Beijing100191 China Yale University New HavenCT06511 United States ShanghaiTech University Shanghai201210 China Beihang University State Key Laboratory of Complex & Critical Software Environment Institute of Artificial Intelligence Beijing100191 China

Recently, post-training quantization (PTQ) has become the de facto way to produce efficient low-precision neural networks without long-time retraining. Despite its low cost, current PTQ works fail to succeed under the extremely low-bit setting. In this work, we delve into extremely low-bit quantization and construct a unified theoretical analysis, which provides an in-depth understanding of the reason for the failure of low-bit quantization. According to the theoretical study, we argue that the existing methods fail in low-bit schemes due to significant perturbation on weights and lack of consideration of activation quantization. To this end, we propose Brecq and QDrop to respectively solve these two challenges, based on which a Q-Limit framework is constructed. Then the Q-Limit framework is further extended to support a mixed precision quantization scheme. To the best of our knowledge, this is the first work that can push the limit of PTQ down to INT2. Extensive experiments on various handcrafted and searched neural architectures are conducted for both visual recognition/detection tasks and language processing tasks. Without bells and whistles, our PTQ framework can attain low-bit ResNet and MobileNetV2 comparable with quantization-aware training (QAT), establishing a new state-of-the-art for PTQ. Our code has been open-sourced at https://***/ModelTC/MQBench/. © 1979-2012 IEEE.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

A Feature-Aware Transformer Approach for Enhanced Maritime Trajectory Prediction

A Feature-Aware Transformer Approach for Enhanced Maritime T...

引用

2024 International Conference Optoelectronic Information and Optical Engineering, OIOE 2024

作者： Wang, Wuyong Ma, Xingkong Liu, Bo College of Computer Science and Technology National University of Defense Technology Changsha410073 China State Key Laboratory of Complex & Critical Software Environment Changsha410073 China Strategic Evaluation and Consultation Center Academy of Military Sciences Beijing100000 China

ISBN: (数字)9781510688209

ISBN: (纸本)9781510688193

Predicting ship trajectories is vital for enhancing maritime safety, optimizing route planning, and boosting the efficiency of maritime traffic management. Current research often neglects the complex marine environment and the effect of ship performance on prediction accuracy. To address these deficiencies, we present a feature-aware transformer-based methodology that integrates meteorological and ship performance data. Our method begins with a time interpolation technique to synchronize the inconsistent time resolutions between meteorological and trajectory data, thereby establishing a comprehensive multi-source dataset. We develop a feature-aware module using transformer architecture to consolidate ship and meteorological features into a unified low-dimensional space, thus enabling the modeling of feature interactions. Experiments on Automatic Identification System (AIS) datasets indicate that our approach reduces the average prediction error to within 4 kilometers over a prediction period of up to 6 hours, achieving a 13% improvement over the state-of-the-art transformer-based methods. © 2025 SPIE.

关键词： Vehicle performance

来源：评论

学校读者我要写书评

暂无评论

Towards practical data alignment in production federated learning

引用

Frontiers of Computer Science 2025年第1期19卷 167-169页

作者： Yexuan SHI Wei YU Yuanyuan ZHANG Chunbo XUE Yuxiang ZENG Zimu ZHOU Manxue GUO Lun XIN Wenjing NIE State Key Laboratory of Complex&Critical Software Environment and Advanced Innovation Center for Future Blockchain and Privacy Computing Beihang UniversityBeijing 100191China Zhongguancun Pan Connected Mobile Communication Technology Innovation and Application Research Institute Beijing 100088China China Mobile Research Institute Beijing 100053China School of Data Science City University of Hong KongHong Kong 999077China

1 Introduction Federated learning has emerged as a promising par-adigm for collaborative model training that facilitates cooperation among multiple parties while ensuring data privacy[1].Successful alignment of data across parties is crucial for effective federated learning[2].This alignment involves harmonizing heterogeneous data from different parties to identify shared data for joint model *** set intersection(PSI)is a technique that allows the alignment of common entities between parties without revealing additional ***,efficiently performing data alignment with PSI in federated learning[3],especially when dealing with highly unbalanced data,remains challenging due to the low efficiency.

关键词： alignment dealing parties

来源：评论

学校读者我要写书评

暂无评论

Structural Entropy Guided Unsupervised Graph Out-Of-Distribution Detection

arXiv

引用

arXiv 2025年

作者： Hou, Yue Zhu, He Liu, Ruomei Su, Yingke Xia, Jinxiang Wu, Junran Xu, Ke State Key Laboratory of Complex & Critical Software Environment Beihang University China Shen Yuan Honors College Beihang University China

With the emerging of huge amount of unlabeled data, unsupervised out-of-distribution (OOD) detection is vital for ensuring the reliability of graph neural networks (GNNs) by identifying OOD samples from in-distribution (ID) ones during testing, where encountering novel or unknown data is inevitable. Existing methods often suffer from compromised performance due to redundant information in graph structures, which impairs their ability to effectively differentiate between ID and OOD data. To address this challenge, we propose SEGO, an unsupervised framework that integrates structural entropy into OOD detection regarding graph classification. Specifically, within the architecture of contrastive learning, SEGO introduces an anchor view in the form of coding tree by minimizing structural entropy. The obtained coding tree effectively removes redundant information from graphs while preserving essential structural information, enabling the capture of distinct graph patterns between ID and OOD samples. Furthermore, we present a multi-grained contrastive learning scheme at local, global, and tree levels using triplet views, where coding trees with essential information serve as the anchor view. Extensive experiments on real-world datasets validate the effectiveness of SEGO, demonstrating superior performance over state-of-the-art baselines in OOD detection. Specifically, our method achieves the best performance on 9 out of 10 dataset pairs, with an average improvement of 3.7% on OOD detection datasets, significantly surpassing the best competitor by 10.8% on the FreeSolv/ToxCast dataset pair. © 2025, CC BY.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Enhancing Cervical Segmentation via Alternate Diverse Teaching in Semi-Supervised Learning

Enhancing Cervical Segmentation via Alternate Diverse Teachi...

引用

IEEE International Symposium on Biomedical Imaging

作者： Jiawei Yu College of Computer Science National University of Defense Technology Changsha China State Key Laboratory of Complex & Critical Software Environment Changsha China

ISBN: (数字)9798331520526

ISBN: (纸本)9798331520533

In clinical practice, the cervical structure plays a critical role in predicting preterm birth, thus making accurate segmentation of this structure in ultrasound images of paramount importance. However, the performance of existing deep learning methods is limited by the high cost of labeling ultrasound images. In this context, semi-supervised medical image segmentation techniques have emerged as promising solutions, particularly recent teacher-student frameworks that have demonstrated notable efficacy. Despite their success, these methods are often subject to confirmation bias, which undermines their reliability. To address this issue, we introduce a novel approach, Alternate Diverse Teaching (AD-MT), for the semi-supervised segmentation of cervical ultrasound images. AD-MT enhances model performance by alternately updating two teacher models, offering diverse perspectives during learning. Furthermore, a conftict-combating module is incorporated to ensure that the model learns from the discrepancies between teacher predictions. This methodology effectively harnesses both labeled and unlabeled data for segmentation tasks. Experimental results demonstrate that AD-MT achieves competitive segmentation performance, as evidenced by its strong showing in the ISBI 2025 Fetal Ultrasound Grand Challenge.

关键词： Training Deep learning Image segmentation Ultrasonic imaging Biological system modeling Semisupervised learning Predictive models Reliability Labeling Biomedical imaging

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：