检索结果-内蒙古大学图书馆

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Huang, Haocheng Chen, Jiaxin Guo, Jinyang Zhan, Ruiyi Wang, Yunhong State Key Laboratory of Virtual Reality Technology and Systems Beihang University Beijing China School of Computer Science and Engineering Beihang University Beijing China School of Artificial Intelligence Beihang University Beijing China

ISBN: (纸本)157735897X

Diffusion models have achieved remarkable success in the image and video generation tasks. Nevertheless, they often require a large amount of memory and time overhead during inference, due to the complex network architecture and considerable number of timesteps for iterative diffusion. Recently, the post-training quantization (PTQ) technique has proved a promising way to reduce the inference cost by quantizing the float-point operations to low-bit ones. However, most of them fail to tackle with the large variations in the distribution of activations across distinct channels and timesteps, as well as the inconsistent of input between quantization and inference on diffusion models, thus leaving much room for improvement. To address the above issues, we propose a novel method dubbed Timestep-Channel Adaptive Quantization for Diffusion Models (TCAQ-DM). Specifically, we develop a timestep-channel joint reparameterization (TCR) module to balance the activation range along both the timesteps and channels, facilitating the successive reconstruction procedure. Subsequently, we employ a dynamically adaptive quantization (DAQ) module that mitigate the quantization error by selecting an optimal quantizer for each post-Softmax layers according to their specific types of distributions. Moreover, we present a progressively aligned reconstruction (PAR) strategy to mitigate the bias caused by the input mismatch. Extensive experiments on various benchmarks and distinct diffusion models demonstrate that the proposed method substantially outperforms the state-of-the-art approaches in most cases, especially yielding comparable FID metrics to the full precision model on CIFAR-10 in the W6A6 setting, while enabling generating available images in the W4A4 settings. Copyright © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Dual Adaptive Representation Alignment for Cross-Domain Few-Shot Learning

引用

IEEE Transactions on Pattern Analysis and Machine Intelligence 2023年第10期45卷 11720-11732页

作者： Zhao, Yifan Zhang, Tong Li, Jia Tian, Yonghong Peking University The School of Computer Science Beijing100871 China Beihang University The State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and Engineering Beijing100191 China The Pengcheng Laboratory Guangdong Shenzhen518055 China Peking University Shenzhen Granduate School The School of Electronic and Computer Engineering Guangdong Shenzhen518055 China

Few-shot learning aims to recognize novel queries with limited support samples by learning from base knowledge. Recent progress in this setting assumes that the base knowledge and novel query samples are distributed in the same domains, which are usually infeasible for realistic applications. Toward this issue, we propose to address the cross-domain few-shot learning problem where only extremely few samples are available in target domains. Under this realistic setting, we focus on the fast adaptation capability of meta-learners by proposing an effective dual adaptive representation alignment approach. In our approach, a prototypical feature alignment is first proposed to recalibrate support instances as prototypes and reproject these prototypes with a differentiable closed-form solution. Therefore feature spaces of learned knowledge can be adaptively transformed to query spaces by the cross-instance and cross-prototype relations. Besides the feature alignment, we further present a normalized distribution alignment module, which exploits prior statistics of query samples for solving the covariant shifts among the support and query samples. With these two modules, a progressive meta-learning framework is constructed to perform the fast adaptation with extremely few-shot samples while maintaining its generalization capabilities. Experimental evidence demonstrates our approach achieves new state-of-the-art results on 4 CDFSL benchmarks and 4 fine-grained cross-domain benchmarks. © 1979-2012 IEEE.

关键词： Alignment

来源：评论

学校读者我要写书评

暂无评论

LogicST: A Logical Self-Training Framework for Document-Level Relation Extraction with Incomplete Annotations

LogicST: A Logical Self-Training Framework for Document-Leve...

引用

2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024

作者： Fan, Shengda Wang, Yanting Mo, Shasha Niu, Jianwei School of Cyber Science and Technology Beihang University Beijing100191 China Zhongguancun Laboratory Beijing100191 China State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and Engineering Beihang University Beijing100191 China

ISBN: (纸本)9798891761643

Document-level relation extraction (DocRE) aims to identify relationships between entities within a document. Due to the vast number of entity pairs, fully annotating all fact triplets is challenging, resulting in datasets with numerous false negative samples. Recently, self-training-based methods have been introduced to address this issue. However, these methods are purely black-box and sub-symbolic, making them difficult to interpret and prone to overlooking symbolic interdependencies between relations. To remedy this deficiency, our insight is that symbolic knowledge, such as logical rules, can be used as diagnostic tools to identify conflicts between pseudo-labels. By resolving these conflicts through logical diagnoses, we can correct erroneous pseudo-labels, thus enhancing the training of neural models. To achieve this, we propose LogicST, a neural-logic self-training framework that iteratively resolves conflicts and constructs the minimal diagnostic set for updating models. Extensive experiments demonstrate that LogicST significantly improves performance and outperforms previous state-of-the-art methods. For instance, LogicST achieves an increase of 7.94% in F1 score compared to CAST (Tan et al., 2023a) on the DocRED benchmark (Yao et al., 2019). Additionally, LogicST is more time-efficient than its self-training counterparts, requiring only 10% of the training time of CAST. Code is available at https://***/XingYing-stack/LogicST. © 2024 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

DualFed: Enjoying both Generalization and Personalization in Federated Learning via Hierachical Representations 24

DualFed: Enjoying both Generalization and Personalization in...

引用

32nd ACM International Conference on Multimedia, MM 2024

作者： Zhu, Guogang Liu, Xuefeng Niu, Jianwei Tang, Shaojie Wu, Xinghao Zhang, Jiayuan State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and Engineering Beihang University Beijing China Zhongguancun Laboratory Beijing China Department of Management Science and Systems University at Buffalo BuffaloNY United States

ISBN: (纸本)9798400706868

In personalized federated learning (PFL), it is widely recognized that achieving both high model generalization and effective personalization poses a significant challenge due to their conflicting nature. As a result, existing PFL methods can only manage a trade-off between these two objectives. This raises an interesting question: Is it feasible to develop a model capable of achieving both objectives simultaneously? Our paper presents an affirmative answer, and the key lies in the observation that deep models inherently exhibit hierarchical architectures, which produce representations with various levels of generalization and personalization at different stages. A straightforward approach stemming from this observation is to select multiple representations from these layers and combine them to concurrently achieve generalization and personalization. However, the number of candidate representations is commonly huge, which makes this method infeasible due to high computational costs. To address this problem, we propose DualFed, a new method that can directly yield dual representations correspond to generalization and personalization respectively, thereby simplifying the optimization task. Specifically, DualFed inserts a personalized projection network between the encoder and classifier. The pre-projection representations are able to capture generalized information shareable across clients, and the post-projection representations are effective to capture task-specific information on local clients. This design minimizes the mutual interference between generalization and personalization, thereby achieving a win-win situation. Extensive experiments show that DualFed can outperform other FL methods. Code is available at https://***/GuogangZhu/DualFed. © 2024 ACM.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Relationship between mechanical load and surface erosion degradation of a shape memory elastomer poly(glycerol-dodecanoate)for soft tissue implant

引用

Regenerative Biomaterials 2023年第1期10卷 1085-1095页

作者： Kaixiang Jin Hanqin Li Mingkai Liang Yuqi Li Lizhen Wang Yubo Fan Key Laboratory of Biomechanics and Mechanobiology(Beihang University) Ministry of EducationBeijing Advanced Innovation Center for Biomedical EngineeringSchool of Biological Science and Medical EngineeringSchool of Engineering MedicineBeihang UniversityBeijing 100083China State Key Laboratory of Virtual Reality Technology and Systems Beihang UniversityBeijing 100083China

Poly(glycerol-dodecanoate)(PGD)has aroused increasing attention in biomedical engineering for its degradability,shape memory and rubber-like mechanical properties,giving it potential to fabricate intelligent implants for soft *** degradation is important for biodegradable implants and is affected by various *** mechanical load has been shown to play an important role in regulating polymer degradation in *** in-depth investigation of PGD degradation under mechanical load is essential for adjusting its degradation behavior after implantation,further guiding to regulate degradation behavior of soft tissue implants made by *** vitro degradation of PGD under different compressive and tensile load has proceeded in this study and describes the relationships by empirical *** on the equations,a continuum damage model is designed to simulate surface erosion degradation of PGD under stress through finite element analysis,which provides a protocol for PGD implants with different geometric structures at varied mechanical conditions and provides solutions for predicting in vivo degradation processes,stress distribution during degradation and optimization of the loaded drug release.

关键词： poly(glycerol-dodecanoate) in vitro degradation mechanical load surface erosion numerical simulation

来源：评论

学校读者我要写书评

暂无评论

Adaptive information fusion network for multi-modal personality recognition

Adaptive information fusion network for multi-modal personal...

引用

作者： Bao, Yongtang Liu, Xiang Qi, Yue Liu, Ruijun Li, Haojie School of Computer Science and Engineering Shandong University of Science and Technology Qingdao China State Key Laboratory of Virtual Reality Technology and Systems Beihang University Beijing China Virtual Reality Research Institute Beihang University Qingdao Research Institute Qingdao China School of Software Beihang University Beijing China

Personality recognition is of great significance in deepening the understanding of social relations. While personality recognition methods have made significant strides in recent years, the challenge of heterogeneity between modalities during feature fusion still needs to be solved. This paper introduces an adaptive multi-modal information fusion network (AMIF-Net) capable of concurrently processing video, audio, and text data. First, utilizing the AMIF-Net encoder, we process the extracted audio and video features separately, effectively capturing long-term data relationships. Then, adding adaptive elements in the fusion network can alleviate the problem of heterogeneity between modes. Lastly, we concatenate audio-video and text features into a regression network to obtain Big Five personality trait scores. Furthermore, we introduce a novel loss function to address the problem of training inaccuracies, taking advantage of its unique property of exhibiting a peak at the critical mean. Our tests on the ChaLearn First Impressions V2 multi-modal dataset show partial performance surpassing state-of-the-art networks. © 2024 John Wiley & Sons Ltd.

关键词： Information fusion

来源：评论

学校读者我要写书评

暂无评论

Adaptive Boundary and Semantic Composite Segmentation Method for Individual Objects in Aerial Images

引用

computer Modeling in Engineering & sciences 2023年第9期136卷 2237-2265页

作者： Ying Li Guanghong Gong Dan Wang Ni Li School of Automation Science and Electrical Engineering Beihang UniversityBeijing100191China Shenyuan Honors College Beihang UniversityBeijing100191China State Key Laboratory of Virtual Reality Technology and Systems Beijing100191China

There are two types of methods for image *** is traditional image processing methods,which are sensitive to details and boundaries,yet fail to recognize semantic *** other is deep learning methods,which can locate and identify different objects,but boundary identifications are not accurate *** of them cannot generate entire segmentation *** order to obtain accurate edge detection and semantic information,an Adaptive Boundary and Semantic Composite Segmentation method(ABSCS)is *** method can precisely semantic segment individual objects in large-size aerial images with limited GPU *** includes adaptively dividing and modifying the aerial images with the proposed principles and methods,using the deep learning method to semantic segment and preprocess the small divided pieces,using three traditional methods to segment and preprocess original-size aerial images,adaptively selecting traditional results tomodify the boundaries of individual objects in deep learning results,and combining the results of different *** object semantic segmentation experiments are conducted by using the AeroScapes dataset,and their results are analyzed qualitatively and *** experimental results demonstrate that the proposed method can achieve more promising object boundaries than the original deep learning *** work also demonstrates the advantages of the proposed method in applications of point cloud semantic segmentation and image inpainting.

关键词： Semantic segmentation aerial images composite method traditional image processing deep learning

来源：评论

学校读者我要写书评

暂无评论

Research on Aviation Combat Simulation Based on STK and FMI 7th

Research on Aviation Combat Simulation Based on STK and FMI

引用

7th Chinese Conference on Swarm Intelligence and Cooperative Control, CCSICC 2023

作者： Li, Weitong Xiao, Jin State Key Laboratory of Virtual Reality Technology and Systems Beihang University Beijing China School of Automation Science and Electrical Engineering Beihang University Beijing China

ISBN: (纸本)9789819733316

With the integration of avionics system from single platform to multi-platform into a larger and highly complex system, the technical requirements of avionics system are more complex, more extensive, and higher precision requirements. In the field of model-based systems engineering, the rationality verification of requirements by analysts and the performance verification of algorithms and functions by professional designers have always been the key considerations for quality assurance of design process. This paper develops a set of general aviation combat simulation platform. Through Socket interface in STK, STK/Connect module is used to obtain real-time information of combat objects in STK scene, and this information is sent to FMU file generated by Simulink or Rhapsody for solving. The calculated data is sent back to the combat scene through the interface again, and the state of the combat object is updated to realize the real-time digital battlefield calculation and simulation. The aviation combat simulation platform can receive the relevant data of the system equipment in the given combat scenario. By designing verification models and connecting them with combat scenarios, professionals can verify the rationality of combat plans and equipment performance requirements, as well as verify the performance of simulation algorithms and functions that meet the actual requirements. This method also supports the adjustment of the combat strategy through the combat results, to optimize the combat performance. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Simulation platform

来源：评论

学校读者我要写书评

暂无评论

Application Research of Edge Computing in Airborne Networks Algorithm 7th

Application Research of Edge Computing in Airborne Networks ...

引用

7th Chinese Conference on Swarm Intelligence and Cooperative Control, CCSICC 2023

作者： Li, Chuxin Xiao, Jin State Key Laboratory of Virtual Reality Technology and Systems Beihang University Beijing China School of Automation Science and Electrical Engineering Beihang University Beijing China

ISBN: (纸本)9789819733316

Due to the requirement of low cost and multi-interaction for UAV formation in future combat mode, in order to solve the problem of time and energy cost optimization through task allocation under the organization mode of UAV formation, this paper proposes a computational offloading model based on BAS algorithm under edge computing platform. With the aim of multi-objective optimization, the UAV formation is abstracted into a constrained communication cost model, and the system benefit function of time and energy cost is constructed. Finally, the multi-objective optimization problem is solved by using the improved BAS algorithm. The dynamic choice of time cost and energy cost in different scenarios can be realized by changing the weight of system income function. This scheme can effectively meet the requirements of low cost and multi-interaction of unmanned formation in the future combat style. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Multiobjective optimization

来源：评论

学校读者我要写书评

暂无评论

LipText: Lip Tracking Based Text Entry in VR 1st

LipText: Lip Tracking Based Text Entry in VR

引用

1st International Conference on Extended reality, ICXR 2024

作者： Leng, Jiaye Wang, Zijun Wu, Jian Wang, Lili School of Creative Media City University of Hong Kong Kowloon Tong Hong Kong State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and Engineering Beihang University Beijing100191 China

ISBN: (纸本)9789819636785

Text entry is an important task in virtual reality (VR), and most existing methods require hand involvement, while hands-free typing has great potential for applications in mobile scenarios. Existing hands-free text entry methods are usually implemented by combining the head and eyes with techniques such as Dwell, Blink and Gesture, which can easily fatigue the user. In this paper, we propose LipText, a lip-tracking-based text entry method in VR. We use a neural network to perform letter-level prediction on the lip data captured by the facial tracker and use head-based selection as an auxiliary to improve the accuracy. We conduct a user study to evaluate our method, the results show a typing speed of 8.63 WPM for the novice group, 9.81 WPM for the potential expert group, and the highest recorded typing speed is 11.13 WPM achieved by a potential expert. Our method is also novice-friendly, and their typing speed increased by 64.38% over a six-day practice. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： virtual environments

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：