检索结果-内蒙古大学图书馆

10th Conference on Sound and Music technology, CSMT 2023

作者： Gan, Xu Wu, Yifei Liu, Xinlu Duan, Ruilei Li, Wei School of Computer Science and Technology Fudan University Shanghai200438 China Department of Music Engineering ZheJiang Conservatory of Music Hangzhou310024 China Shanghai Key Laboratory of Intelligent Information Processing Fudan University Shanghai200433 China

ISBN: (纸本)9789819779611

The goal of the timbre transfer task is to apply the auditory characteristics of one instrument to another instrument. Past timbre transfer methods faced numerous challenges due to the limitations of the selected generation model. These challenges include difficulties in training, ambiguous results, poor diversity, and slow generation speed. To address these issues, we propose a diffusion model for timbre transfer, DiffTimb, which can be combined with vocoder to achieve single-note timbre transfer. Unlike traditional models that require training a specific model for each pair of timbre transfers, DiffTimb enables the conversion of various musical instrument timbres received in a single model, namely, many-to-many timbre transfer. We compare two Unet-based denoising frameworks and propose two feature fusion methods to fuse timbre and pitch features. We conduct unconditional and conditional generation pre-experiments to verify the generation ability of DiffTimb. To improve the generation effect, we clean the Nsynth dataset and get two subsets for our task. Experimental results show that DiffTimb has outperformed previous methods in both subjective and objective evaluations. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Audio acoustics

来源：评论

学校读者我要写书评

暂无评论

Topic-BiGRU-U-Net for Document-level Relation Extraction from Biomedical Literature

Topic-BiGRU-U-Net for Document-level Relation Extraction fro...

引用

2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023

作者： Zhao, Yali Yan, Rong Inner Mongolia University College of Computer Science Hohhot China Inner Mongolia Key Laboratory of Mongolian Information Processing Technology Hohhot China Natl. Loc. Jt. Eng. Research Center of Intelligent Information Processing Technology for Mongolian Hohhot China

ISBN: (纸本)9798350337488

Document-level biomedical relation extraction refers to extract relationship facts from unstructured biomedical literature. Due to the fact that many relationship facts span multiple sentences and involve complex interactions between entities, it requires models with strong logical reasoning capabilities. However, previous methods have had limitations in terms of logical reasoning, such as lack logical reasoning and the entities merely based on a single granularity. Although other efforts are based on multiple granularity, including mention pairs and entity pairs, the inference mechanism is weak. In this paper, we propose a module to enhance the model's logical reasoning capabilities from four kinds of granularity (mention, entity, mention pairs, and entity pairs). Specifically, we encode the entities and mentions separately using BiGRU to capture contextual features. Simultaneously, we employ a U-Net network to model entity pairs and mention pairs, and enhance inter-sentence reasoning abilities. Additionally, to support the model's reasoning capabilities, we introduce topic nodes into the traditional heterogeneous graph to fully extract document information. Experimental results on CDR and GDA datasets in the biomedical field demonstrate the outstanding performance of the proposed model. © 2023 IEEE.

关键词： BiGRU document-level relation extraction information extraction U-Net

来源：评论

学校读者我要写书评

暂无评论

Research on active defense decision-making method for cloud boundary networks based on reinforcement learning of intelligent agent

引用

High-Confidence Computing 2024年第2期4卷 50-61页

作者： Huan Wang Yunlong Tang Yan Wang Ning Wei Junyi Deng Zhiyan Bin Weilong Li School of Computer Science and Technology Guangxi University of Science and TechnologyLiuzhou 545000China Liuzhou Key Laboratory of Big Data Intelligent Processing and Security Liuzhou 545000China Guangxi Education System Network Security Monitoring Center Liuzhou 545000China Center of Graphics and Information Guangxi ECO-Engineering Vocational&Technical CollegeLiuzhou 545004China

The cloud boundary network environment is characterized by a passive defense strategy,discrete defense actions,and delayed defense feedback in the face of network attacks,ignoring the influence of the external environment on defense decisions,thus resulting in poor defense ***,this paper proposes a cloud boundary network active defense model and decision method based on the reinforcement learning of intelligent agent,designs the network structure of the intelligent agent attack and defense game,and depicts the attack and defense game process of cloud boundary network;constructs the observation space and action space of reinforcement learning of intelligent agent in the non-complete information environment,and portrays the interaction process between intelligent agent and environment;establishes the reward mechanism based on the attack and defense gain,and encourage intelligent agents to learn more effective defense *** designed active defense decision intelligent agent based on deep reinforcement learning can solve the problems of border dynamics,interaction lag,and control dispersion in the defense decision process of cloud boundary networks,and improve the autonomy and continuity of defense decisions.

关键词： Active defense decision-making Cloud boundary network security intelligent agent reinforcement learning Offensive and defensive game

来源：评论

学校读者我要写书评

暂无评论

Polarization‑entangled photon‑pair source with van der Waals 3R‑WS_(2) crystal

引用

eLight 2024年第1期4卷 99-106页

作者： Jiangang Feng Yun‑Kun Wu Ruihuan Duan Jun Wang Weijin Chen Jiazhang Qin Zheng Liu Guang‑Can Guo Xi‑Feng Ren Cheng‑Wei Qiu Department of Electrical and Computer Engineering National University of SingaporeSingapore 117583Singapore CAS Key Laboratory of Quantum Information University of Science and Technology of ChinaHefei 230026China CAS Center for Excellence in Quantum Information and Quantum Physics University of Science and Technology of ChinaHefei 230026China Hefei National Laboratory University of Science and Technology of ChinaHefei 230088China School of Materials Science and Engineering Nanyang Technological UniversitySingapore 639798Singapore Department of Optical Science and Engineering and Shanghai Frontiers Science Research Base of Intelligent Optoelectronics and PerceptionFudan UniversityShanghai 200433China

Ultracompact entangled photon sources are pivotal to miniaturized quantum photonic *** der Waals(vdW)nonlinear crystals promise efficient photon-pair generation and on-chip monolithic integration with nanophotonic ***,it remains challenging to generate maximally entangled Bell states of photon pairs with high purity,generation rate,and fidelity required for practical ***,we realize a polarization-entangled photon-pair source based on spontaneous parametric down conversion in an ultrathin rhombohedral tungsten disulfide(3R-WS2)*** vdW entangled photonic source exhibits a high photon-pair purity with a coincidenceto-accidental ratio of above 800,a generation rate of 31 Hz,and two maximally polarization-entangled Bell states with fidelities exceeding 0.93 and entanglement degree over *** results stem from scalable optical nonlinearity,enhanced second-order susceptibility by electronic transitions,and a well-defined symmetry-enabled selection rule inherent in *** polarization entangled photon source can be integrated with photonic structures for generating more complex entangled states,thus paving an avenue for advanced quantum photonic systems toward computation and metrology.

关键词： purity maximal Polarization

来源：评论

学校读者我要写书评

暂无评论

Leveraging catastrophic forgetting to develop safe diffusion models against malicious finetuning 24

Leveraging catastrophic forgetting to develop safe diffusion...

引用

Proceedings of the 38th International Conference on Neural information processing Systems

作者： Jiadong Pan Hongcheng Gao Zongyu Wu Taihang Hu Li Su Qingming Huang Liang Li Key Laboratory of Intelligent Information Processing Institute of Computing Technology CAS and University of Chinese Academy of Sciences University of Chinese Academy of Sciences The Pennsylvania State University Nankai University Key Laboratory of Intelligent Information Processing Institute of Computing Technology CAS

ISBN: (纸本)9798331314385

Diffusion models (DMs) have demonstrated remarkable proficiency in producing images based on textual prompts. Numerous methods have been proposed to ensure these models generate safe images. Early methods attempt to incorporate safety filters into models to mitigate the risk of generating harmful images but such external filters do not inherently detoxify the model and can be easily bypassed. Hence, model unlearning and data cleaning are the most essential methods for maintaining the safety of models, given their impact on model parameters. However, malicious fine-tuning can still make models prone to generating harmful or undesirable images even with these methods. Inspired by the phenomenon of catastrophic forgetting, we propose a training policy using contrastive learning to increase the latent space distance between clean and harmful data distribution, thereby protecting models from being fine-tuned to generate harmful images due to forgetting. The experimental results demonstrate that our methods not only maintain clean image generation capabilities before malicious fine-tuning but also effectively prevent DMs from producing harmful images after malicious fine-tuning. Our method can also be combined with other safety methods to maintain their safety against malicious fine-tuning further.

关键词：

来源：评论

学校读者我要写书评

暂无评论

keypoints and Action Units Jointly Drive Talking Head Generation for Video Conferencing

引用

IEEE Transactions on Circuits and Systems for Video technology 2025年

作者： Shi, Wuzhen Xue, Zibang Wen, Yang Shenzhen University Shenzhen Key Laboratory of Digital Creative Technology Guangdong Province Engineering Laboratory for Digital Creative Technology Guangdong Provincial Key Laboratory of Intelligent Information Processing College of Electronics and Information Engineering Shenzhen518060 China

This paper introduces a high-quality talking head generation method that is jointly driven by keypoints and action units, aiming to strike a balance between low-bandwidth transmission and high-quality generation in video conference scenarios. Existing methods for talking head generation often face limitations: they either require an excessive amount of driving information or struggle with accuracy and quality when adapted to low-bandwidth conditions. To address this, we decompose the talking head generation task into two components: a driving task, focused on information-limited control, and an enhancement task, aimed at achieving high-quality, high-definition output. Our proposed method innovatively incorporates the joint driving of keypoints and action units, improving the accuracy of pose and expression generation while remaining suitable for low-bandwidth environments. Furthermore, we implement a multistep video quality enhancement process, targeting both the entire frame and key regions, while incorporating temporal consistency constraints. By leveraging attention mechanisms, we enhance the realism of the challenging-to-generate mouth regions and mitigate background jitter through background fusion. Finally, a prior-driven super-resolution network is employed to achieve high-quality display. Extensive experiments demonstrate that our method effectively supports low-resolution recording, low-bandwidth transmission, and high-definition display. © 1991-2012 IEEE.

关键词： Bandwidth

来源：评论

学校读者我要写书评

暂无评论

Adaptive Neural Consensus Tracking Control for Nonlinear Multiagent Systems with Dead Zone and Sensor Faults 42

Adaptive Neural Consensus Tracking Control for Nonlinear Mul...

引用

42nd Chinese Control Conference, CCC 2023

作者： Huang, Chengjie Chen, Ci Liu, Zhi Li, Zhenni Guangdong University of Technology Guangzhou510006 China Guangdong Key Laboratory of IoT Information Technology Guangzhou510006 China 111 Center for Intelligent Batch Manufacturing Based on IoT Technology Guangzhou510006 China Guangdong-HongKong-Macao Joint Laboratory for Smart Discrete Manufacturing Guangzhou510006 China Key Laboratory of Intelligent Information Processing and System Integration of IoT Ministry of Education Guangzhou510006 China Key Laboratory of Intelligent Detection and the Internet of Things in Manufacturing Ministry of Education Guangzhou510006 China

ISBN: (纸本)9789887581543

In this paper, an adaptive neural tracking control design is investigated for uncertain multiagent systems (MASs) with actuator dead zone and sensor faults. The adaptive control technique based on neural networks is employed to handle the uncertainty in the systems, which, however, results in a bounded tracking error from most existing works. To achieve the asymptotic tracking performance, we proposed a direct adaptive strategy via the tuning function method. Technically, a Lyapunov function with a series of smooth functions is designed, which benefits proposing of an asymptotic controller. Then, under the framework of direct adaptive control, an inverse compensation scheme, as well as a fault-tolerant mechanism, are considered for disposing of the actuator dead zone and sensor failures. As a result, user-preset tracking performance is achieved and the system error in the MASs converges to a predefined range. An example in the simulation section illustrates the effectiveness of the proposed method. © 2023 Technical Committee on Control Theory, Chinese Association of Automation.

关键词： Multi agent systems

来源：评论

学校读者我要写书评

暂无评论

Region of interest image encryption method based on panoramic segmentation and a novel coupled chaotic map

Region of interest image encryption method based on panorami...

引用

2022 International Conference on Artificial Intelligence, Virtual Reality, and Visualization, AIVRV 2022

作者： Zhang, Hangming Hu, Hanping School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan430074 China Key Laboratory of Image Information Processing and Intelligent Control Ministry of Education Wuhan430074 China

ISBN: (纸本)9781510662988

Social media is a platform for people to share their lives and interact with others. Image sharing is an integral component. Privacy information will inevitably be compromised during the process of sharing whole photos. Sometimes, it is not even caused by the person who shares the image but the third party who forwards the image. In the majority of cases, we do not attempt to protect the entirety of the media;rather, we seek to protect the most crucial portion, whether it is the background or specific objects, which decreasing superfluous cryptography procedures. Unfortunately, current research on such field is extremely rare. In this research, we established a novel pixel-level image encryption technique relying on Panoptic FCN panoramic segmentation and chaos for client-intended images on social media sites. Our suggested technology is capable of automated picture encryption on either the whole images or user-selected areas, whether they are rectangular or irregular, which is suited for all region of interest (ROI) encryption. Relying on a novel coupled chaotic map, this universal new encryption method flattens the array of the image ROI into a series of pixels. The module of Panoptic FCN can be replaced by any other panoptic segmentation models which are stronger in performance. Statistical and cryptographic evaluations demonstrate that our technique preserves the high efficiency for practical applications. © 2023 SPIE. All rights reserved.

关键词： Cryptography

来源：评论

学校读者我要写书评

暂无评论

A Policy-Controlled Signature Scheme for end-to-end Secure Communication 12

A Policy-Controlled Signature Scheme for end-to-end Secure C...

引用

12th International Conference on Advanced Cloud and Big Data, CBD 2024

作者： Yu, Feng Meng, Linghui Xi, Jing Li, Xianxian Lu, Qianhui Zhu, Weidong Guangxi Normal University Key Lab of Education Blockchain and Intelligent Technology GuiLin China University of SouthEast Jiangsu Provincial Key Laboratory of Netwrok and Information Security Nanjing China Guangxi Normal University Guangxi Key Lab of Multi-Source Information Mining and Security GuiLin China Guangxi Collaborative Innovation Center of Multi-source Information Integration and Intelligent Processing Guangxi Normal University GuiLin China

ISBN: (纸本)9798331511074

In the digitalized society, end-to-end security communication is receiving increasing attention. Issues such as ma-licious sender impersonation and over-privilege communication continue to prevail in end-to-end communication. This paper pro-poses a policy-controlled signature scheme aimed at addressing these concerns through identity verification and permission man-agement, thus establishing a secure mode of communication. The scheme designs policy-controlled signatures with partial policy hiding functionality, employing expressive access structures and garbled Bloom Filter (GBF) to protect the sender's attributes. In comparison to related approaches, this scheme provides benefits in terms of signature computation cost, thereby enhancing the resilience and efficiency of the communication system against attacks. © 2024 IEEE.

关键词： Authentication Protocol

来源：评论

学校读者我要写书评

暂无评论

DD-Net: Dynamic Network Architecture for Optimized Curve Segmentation and Reduce Computational Redundancy 27th

DD-Net: Dynamic Network Architecture for Optimized Curve Seg...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Cao, Yunxiang Chen, Li Wang, Yubo Feng, Zhida Tian, Jing School of Computer Science and Technology Wuhan University of Science and Technology Wuhan China Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System Wuhan University of Science and Technology Wuhan China Institute of Systems Science National University of Singapore Singapore Singapore

ISBN: (纸本)9783031783821

Curvilinear Structure segmentation has numerous applications in various fields, including providing a better understanding of defects such as cracks on roads or walls, thereby assessing the structural safety of buildings. Numerous curve segmentation methods based on Convolutional Neural Networks(CNN) have been developed in recent years. However, preserving feature integrity remains a challenge. To address this issue, this paper introduces a double dynamic network, called DD-Net, which consists of a dynamic structure for training and dynamic masking for inference. Firstly, to enhance the nonlinearity of the model, a paradigm to auto-adjust the CNN structure via a trainable module is formed for better fine-grained feature extraction. This dynamic structure (DS) paradigm enables a trade-off to extract or discard the feature data in the model training. Due to the data heterogeneity between the training and inference data, models may contain some useless nodes that are less effective during inference. The dynamic masking (DM) is proposed to omit the useless nodes based on the score difference between the train and inference feature statistics, thereby reducing redundant computations. To further improve the model’s performance on thin-curve segmentation and to preserve feature integrity, we introduce a non-curve suppression (NCS) module. This module focuses on background information while considering foreground prediction to address noisy conditions. The experimental results show that our DD-Net achieves promising results on three benchmark datasets and outperforms state-of-the-art curve segmentation models. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Convolutional neural networks

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：