检索结果-内蒙古大学图书馆

IEEE Transactions on Audio, Speech and Language Processing 2025年 33卷 797-811页

作者： Ruijie Tao Xinyuan Qian Yidi Jiang Junjie Li Jiadong Wang Haizhou Li Department of Electrical and Computer Engineering National University of Singapore Singapore Department of Computer Science and Technology University of Science and Technology Beijing Beijing China Chinese University of Hong Kong Shenzhen China Guangdong Provincial Key Laboratory of Big Data Computing Shenzhen Research Institute of Big Data The Chinese University of Hong Kong Shenzhen China University of Bremen Bremen Germany

Audio-visual target speaker extraction (AV-TSE) aims to extract the specific person's speech from the audio mixture given auxiliary visual cues. Previous methods usually search for the target voice through speech-lip synchronization. However, this strategy mainly focuseson the existence of target speech, while ignoring the variations of the noise characteristics, i.e., interference speaker and the background noise. That may result in extracting noisy signals from the incorrect sound source in challenging acoustic situations. To this end, we propose a novel selective auditory attention mechanism, which can suppress interference speakers and non-speech signals to avoid incorrect speaker extraction. By estimating and utilizing the undesired noisy signal through this mechanism, we design an AV-TSE framework named Subtraction-and-ExtrAction network (SEANet) to suppress the noisy signals. We conduct abundant experiments by re-implementing three popular AV-TSE methods as the baselines and involving nine metrics for evaluation. The experimental results show that our proposed SEANet achieves state-of-the-art results and performs well for all five datasets.

关键词： Noise measurement Lips Visualization Synchronization Noise Correlation data mining Speech processing Interference Speech recognition

来源：评论

学校读者我要写书评

暂无评论

High-Fidelity Editable Portrait Synthesis with 3D GAN Inversion

High-Fidelity Editable Portrait Synthesis with 3D GAN Invers...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Jindong Xie Jiachen Liu Yupei Lin Jinbao Wang Xianxu Hou Linlin Shen Computer Vision Institute School of Computer Science and Software Engineering Shenzhen University National Engineering Laboratory for Big Data System Computing Technology Shenzhen University School of Information Engineering Guangdong University of Technology Guangzhou China Guangdong Provincial Key Laboratory of Intelligent Information Processing School of AI and Advanced Computing Xi’an Jiaotong-Liverpool University

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

The 3D generative adversarial network (GAN) inversion converts an image into 3D representation to attain high-fidelity reconstruction and facilitate realistic image manipulation within the 3D latent space. However, previous approaches face challenges regarding the trade-off between the reconstruction ability and editability. That is, reversing a real-world image to a low-dimensional latent code would inevitably lead to information loss, and achieving a near-perfect reconstruction using high-rate triplane representation often limits the ability to manipulate the image freely in the latent space. To address these issues, we propose a novel latent conditioning encoder-based framework with the alignment between the low-dimensional latent and high-dimensional triplane. A non-semantic guided editing strategy bridges the intrinsic relation between the latent condition and triplane generation, making it possible to edit the high-dimensional representation by latent manipulation. As a result, our method can achieve high-fidelity reconstruction and editing simultaneously by directly controlling the latent code. Experimental results demonstrate that our approach excels in reconstruction and editing quality compared to previous 3D inversion methods. Furthermore, our method can also edit even real faces with large poses and out-of-domain cases.

关键词： Bridges Three-dimensional displays Codes Signal processing Generative adversarial networks Vectors Acoustics Speech processing Image reconstruction Faces

来源：评论

学校读者我要写书评

暂无评论

OUTLIER SYNTHESIS VIA HAMILTONIAN MONTE CARLO FOR OUT-OF-DISTRIBUTION DETECTION

arXiv

引用

arXiv 2025年

作者： Li, Hengzhuang Zhang, Teng National Engineering Research Center for Big Data Technology and System Service Computing Technology and Systems Laboratory Cluster and Grid Computing Lab School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China

Out-of-distribution (OOD) detection is crucial for developing trustworthy and reliable machine learning systems. Recent advances in training with auxiliary OOD data demonstrate efficacy in enhancing detection capabilities. Nonetheless, these methods heavily rely on acquiring a large pool of high-quality natural outliers. Some prior methods try to alleviate this problem by synthesizing virtual outliers but suffer from either poor quality or high cost due to the monotonous sampling strategy and the heavy-parameterized generative models. In this paper, we overcome all these problems by proposing the Hamiltonian Monte Carlo Outlier Synthesis (HamOS) framework, which views the synthesis process as sampling from Markov chains. Based solely on the in-distribution data, the Markov chains can extensively traverse the feature space and generate diverse and representative outliers, hence exposing the model to miscellaneous potential OOD scenarios. The Hamiltonian Monte Carlo with sampling acceptance rate almost close to 1 also makes our framework enjoy great efficiency. By empirically competing with SOTA baselines on both standard and large-scale benchmarks, we verify the efficacy and efficiency of our proposed HamOS. Our code is available at: https://***/Fir-lat/HamOS_OOD. © 2025, CC BY.

关键词： Markov chains

来源：评论

学校读者我要写书评

暂无评论

It Takes Two to Tango: Serverless Workflow Serving via Bilaterally Engaged Resource Adaptation

arXiv

引用

arXiv 2025年

作者： Wu, Jing Wang, Lin Deng, Quanfeng Yu, Chen Zhang, Dong Yan, Bingheng Liu, Fangming National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab Huazhong University of Science and Technology China Paderborn University Germany Inspur Data Co. Ltd. China Peng Cheng Laboratory China

Serverless platforms typically adopt an early-binding approach for function sizing, requiring developers to specify an immutable size for each function within a workflow beforehand. Accounting for potential runtime variability, developers must size functions for worst-case scenarios to ensure service-level objectives (SLOs), resulting in significant resource inefficiency. To address this issue, we propose Janus, a novel resource adaptation framework for serverless platforms. Janus employs a late-binding approach, allowing function sizes to be dynamically adapted based on runtime conditions. The main challenge lies in the information barrier between the developer and the provider: developers lack access to runtime information, while providers lack domain knowledge about the workflow. To bridge this gap, Janus allows developers to provide hints containing rules and options for resource adaptation. Providers then follow these hints to dynamically adjust resource allocation at runtime based on real-time function execution information, ensuring compliance with SLOs. We implement Janus and conduct extensive experiments with real-world serverless workflows. Our results demonstrate that Janus enhances resource efficiency by up to 34.7% compared to the state-of-the-art. © 2025, CC BY-NC-ND.

关键词： Resource allocation

来源：评论

学校读者我要写书评

暂无评论

The role of testosterone in modulating positive and negative empathy in social interactions

引用

Neuropharmacology 2025年 274卷 110465页

作者： Zhuo, Shiwei Zhang, Yinhua Lin, Chennan Wu, Wen Peng, Weiwei School of Psychology Shenzhen University Guangdong Shenzhen China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Guangdong Shenzhen China Department of Rehabilitation Zhujiang Hospital Southern Medical University Guangdong Guangzhou China

Empathy encompasses both negative (e.g., distress) and positive (e.g., shared joy) dimensions, yet the effects of testosterone on positive empathy and its modulation of intrinsic neural dynamics remain underexplored. This double-blind, placebo-controlled study examined how testosterone influences neural sensitivity to empathy within social inclusion and exclusion contexts, as well as its impact on resting-state EEG microstates—millisecond-scale transient patterns of brain activity. Healthy male participants received either testosterone or placebo before completing resting-state EEG recordings and an empathy task featuring social scenarios. While self-reported empathy ratings remained unchanged, testosterone amplified neurophysiological responses: it enhanced anterior N2 amplitude (250–310 ms), associated with negative empathy toward social exclusion, and increased posterior α-event-related desynchronization (8.28–10 Hz;1226–1901 ms), linked to positive empathy during social inclusion. These findings suggest that testosterone enhances neural responsiveness to both threatening and affiliative social cues, reinforcing its role in adaptive social vigilance. Resting-state EEG microstate analysis further revealed that testosterone prolonged the temporal dominance of microstate E—a centro-parietal activity pattern associated with interoceptive awareness and emotional processing. Notably, these microstate E changes predicted increased emotional empathy across both positive and negative contexts. Together, our findings suggest that testosterone indirectly enhances empathy-related responsiveness by amplifying baseline interoceptive sensitivity to socially salient stimuli. These dual effects—enhanced intrinsic interoceptive processing and heightened neural reactivity to social cues—position testosterone as a key neuromodulator of context-adaptive social perception. © 2025 Elsevier Ltd

关键词： Empathy Event-related desynchronization Event-related potentials Microstate analysis Testosterone

来源：评论

学校读者我要写书评

暂无评论

EagerLog: Active Learning Enhanced Retrieval Augmented Generation for Log-based Anomaly Detection

EagerLog: Active Learning Enhanced Retrieval Augmented Gener...

引用

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

作者： Duan, Chiming Jia, Tong Yang, Yong Liu, Guiyang Liu, Jinbu Zhang, Huxing Zhou, Qi Li, Ying Huang, Gang School of Software and Microelectronics Peking University Beijing China Institute for Artificial Intelligence Peking University Beijing China National Engineering Research Center for Software Engineering Peking University Beijing China National Key Laboratory of Data Space Technology and System Beijing China Alibaba Cloud Computing Company Hangzhou China

ISBN: (纸本)9798350368741

Logs record essential information about system operations and serve as a critical source for anomaly detection, which has generated growing research interest. Utilizing large language models (LLMs) within a retrieval-augmented generation (RAG) framework for log-based anomaly detection is an effective approach due to its strong generalization capabilities and efficient few-shot performance. However, the effectiveness of this method hinges on the quality of the knowledge source, which can be impacted by noise and changes within the software systems. Facing these problems, in this paper, we propose a novel log-based anomaly detection method named EagerLog, employing active learning to choose the logs for humans to label, thereby adding them to the knowledge source, thus enhancing the knowledge source and maintaining its quality. Our experiments on three open datasets (BGL, Thunderbird, Zookeeper) and one industrial dataset demonstrate that EagerLog can achieve 93.65% F1 score with approximately 10 labeled log sequences, surpassing existing methods by 15.32%. © 2025 IEEE.

关键词： Active Learning Anomaly Detection Log Analysis Retrieval Augmented Generation

来源：评论

学校读者我要写书评

暂无评论

Working Smarter Not Harder: Hybrid Cooling for Deep Learning in Edge datacenters

IEEE Transactions on Sustainable Computing

引用

IEEE Transactions on Sustainable computing 2025年

作者： Pei, Qiangyu Yuan, Yongjie Hu, Haichuan Wang, Lin Zhang, Dong Yan, Bingheng Yu, Chen Liu, Fangming Huazhong University of Science and Technology National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab School of Computer Science and Technology 1037 Luoyu Road Wuhan430074 China Paderborn University TU Darmstadt Germany Jinan Inspur Data Co. Ltd. China Huazhong University of Science and Technology Peng Cheng Laboratory China

The proliferation of deep-learning-based mobile and IoT applications has driven the increasing deployment of edge datacenters equipped with domain-specific accelerators. The unprecedented computing power offered by these accelerators puts a heavy burden on the cooling system, motivating more potent cooling techniques like cold water cooling. However, we observe that cold water cooling results in significant energy waste in edge datacenters due to the fluctuating resource utilization both spatially and temporally. To tackle this issue, we propose the concept of "working smarter"by slowing down accelerators deliberately whenever possible and enabling warm water cooling during these times to achieve cooling efficiency. Based on this concept, we develop Hyco - a hybrid water cooling system tailored for edge datacenters running deep learning workloads. First, Hyco features a zone-based cooling architecture enabling dynamic switching between cold water and warm water cooling. Then, based on a lightweight latency estimation method, Hyco incorporates a learning-based scheduling scheme to determine "which"accelerator workers and "when"to slow down through an adaptive and intelligent power-latency trade-off for deep learning models. The simulation with real-world traces shows that Hyco reduces the cooling energy consumption by up to 34.74× while satisfying latency constraints more than 99% of the time for deep-learning-based applications. © 2016 IEEE.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

SAV-SE: Scene-aware Audio-Visual Speech Enhancement with Selective State Space Model

引用

IEEE Journal on Selected Topics in Signal Processing 2025年

作者： Qian, Xinyuan Gao, Jiaran Zhang, Yaodan Zhang, Qiquan Liu, Hexin Garcia, Leibny Paola Li, Haizhou University of Science and Technology Beijing School of Computer and Communication Engineering Beijing100083 China University of New South Wales School of Electrical Engineering and Telecommunications Sydney2052 Australia Nanyang Technological University College of Computing and Data Science Singapore Singapore Johns Hopkins University CLSP and HLT-COE United States Chinese University of HongKong Guangdong Provincial Key Laboratory of Big Data Computing Shenzhen518172 China Shenzhen Research Institute of Big data Shenzhen51872 China

Speech enhancement plays an essential role in various applications, and the integration of visual information has been demonstrated to bring substantial advantages. However, the majority of current research concentrates on the examination of facial and lip movements, which can be compromised or entirely inaccessible in scenarios where occlusions occur or when the camera view is distant. Whereas contextual visual cues from the surrounding environment have been overlooked: for example, when we see a dog bark, our brain has the innate ability to discern and filter out the barking noise. To this end, in this paper, we introduce a novel task, i.e. Scene-aware Audio-Visual Speech Enhancement SAV-SE. To our best knowledge, this is the first proposal to use rich contextual information from synchronized video as auxiliary cues to indicate the type of noise, which eventually improves the speech enhancement performance. Specifically, we propose the VC-S $^{2}$ E method, which incorporates the Conformer and Mamba modules for their complementary strengths. Extensive experiments are conducted on public MUSIC, AVSpeech and AudioSet datasets, where the results demonstrate the superiority of VC-S $^{2}$ E over other competitive methods. We will make the source code publicly available. Project demo page: https://***/ © 2007-2012 IEEE.

关键词： Speech enhancement

来源：评论

学校读者我要写书评

暂无评论

Efficient Hybrid Multi-Population Genetic Algorithm for Multi-UAV Task Assignment in Consumer Electronics Applications

引用

IEEE Transactions on Consumer Electronics 2025年

作者： Bai, Xiaoshan Jiang, Haoyu Li, Chao Ullah, Inam Dabel, Maryam M. Al Bashir, Ali Kashif Wu, Zongze Sam, Shuzhi Shenzhen University College of Mechatronics and Control Engineering National Engineering Laboratory for Big Data System Computing Technology Shenzhen518060 China Shenzhen University College of Mechatronics and Control Engineering Shenzhen518060 China Shenzhen University Shenzhen City Joint Laboratory of Autonomous Unmanned Systems and Intelligent Manipulation Shenzhen518060 China Tashkent State University of Economics Department of Artificial Intelligence Tashkent100066 Uzbekistan University of Hafr Al Batin Department of Computer Science and Engineering College of Computer Science and Engineering Saudi Arabia Manchester Metropolitan University Manchester United Kingdom Shenzhen518060 China National University of Singapore Department of Electrical and Computer Engineering Singapore117576 Singapore

In recent years, as people’s living standards have improved and consumption concepts have been transformed, the demand for purchasing consumer electronics online has continued to grow, further stimulating the development of the logistics industry. Consequently, how to deliver consumer electronics to households at minimal cost has become a crucial factor that limits the development of the consumer technology industry. To tackle this problem, this paper studies the task assignment problem for multiple initially dispersed UAVs to deliver products to target locations while minimizing their total operation time. Each UAV can continuously provide delivery services to multiple target locations within its limited loading capacity and operation time. To solve this problem, we propose several hybrid multipopulation genetic algorithms. First, a novel crossover operator for the genetic algorithms is designed, through which a single parent chromosome can generate offspring individually. Second, two mutation mechanisms are performed to increase gene diversity. Third, multiple local search strategies are employed to enhance the populations’ fitness during each iteration of evolution. An improved 2-opt local search strategy is applied to optimize individual chromosomes when their similarity with the current best chromosome falls below a prescribed threshold. Alternatively, local search strategies are utilized for 1-opt, 2h-opt and interchange processes. Combining local search strategies, genetic operators, and the multi-population mechanism leads to several hybrid multi-population genetic algorithms. Numerical simulations and experimental tests demonstrate that the hybrid multi-population genetic algorithm, integrated with the improved 2-opt and 1-opt local search strategies, exhibits superior performance among the designed hybrid genetic algorithms, the minimum marginal cost algorithm (MMA), and the existing popular Co-evolutionary Multi-population Genetic Algorithm (CMGA). In exp

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Comprehensive Architecture Search for Deep Graph Neural Networks

引用

IEEE Transactions on big data 2025年

作者： Dong, Yukang Pan, Fanxing Gui, Yi Jiang, Wenbin Wan, Yao Zheng, Ran Jin, Hai National Engineering Research Center for Big Data Technology Huazhong University of Science and Technology Wuhan430074 China Huazhong University of Science and Technology Service Computing Technology and System Laboratory Wuhan430074 China Huazhong University of Science and Technology Cluster and Grid Computing Laboratory Wuhan430074 China Huazhong University of Science and Technology School of Computer Science and Technology Wuhan430074 China Zhejiang Lab Hangzhou311121 China

In recent years, Neural Architecture Search (NAS) has emerged as a promising approach for automatically discovering superior model architectures for deep Graph Neural Networks (GNNs). Different methods have paid attention to different types of search spaces. However, due to the time-consuming nature of training deep GNNs, existing NAS methods often fail to explore diverse search spaces sufficiently, which constrains their effectiveness. To crack this hard nut, we propose CAS-DGNN, a novel comprehensive architecture search method for deep GNNs. It encompasses four kinds of search spaces that are the composition of aggregate and update operators, different types of aggregate operators, residual connections, and hyper-parameters. To meet the needs of such a complex situation, a phased and hybrid search strategy is proposed to accommodate the diverse characteristics of different search spaces. Specifically, we divide the search process into four phases, utilizing evolutionary algorithms and Bayesian optimization. Meanwhile, we design two distinct search methods for residual connections (All-connected search and Initial Residual search) to streamline the search space, which enhances the scalability of CAS-DGNN. The experimental results show that CAS-DGNN achieves higher accuracy with competitive search costs across ten public datasets compared to existing methods. © 2015 IEEE.

关键词： Neural network models

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：