检索结果-内蒙古大学图书馆

SSRN 2024年

作者： Peng, Sisi Qu, Dan Zhang, Wenlin Zhang, Hao Li, Shunhang Xu, Minchen Information Engineering University Zhengzhou China Laboratory for Advanced Computing and Intelligence Engineering Zhengzhou China

In recent years, knowledge-based dialogue generation has become a popular research topic due to its ability to generate informative and coherent responses by incorporating external knowledge into models. However, acquiring high-quality knowledge that corresponds to the dialogue content can be difficult and requires significant time and resources. To address the issue of limited dialogue data, most research efforts focus on data augmentation to obtain more training data. Unfortunately, these approaches do not consider knowledge augmentation, resulting in limited diversity of input data and only achieving improvements on a single metric. Real-life conversations are diverse, with repetitions, reversals, and interruptions, which present a higher demand for data diversity. In this paper, we propose a simple and effective data augmentation method called Multi-perspective Sentence Interaction to enhance the connections between sentences from different perspectives. By analyzing the target responses from multiple dialogue perspectives, we gain a better understanding of the relationships between dialogue sentences, thereby achieving an expanded knowledge-based dialogue data. We conduct experiments on multiple knowledge-based dialogue datasets and different models, and the results demonstrate that our method significantly improves the quality of model generation. Specifically, we achieved a 3.5% increase in reply accuracy and a 0.1506 improvement in diversity (DIST-2). Furthermore, the accuracy of knowledge selection improved by 19.04%, and the model perplexity decreased by 31.48%. © 2024, The Authors. All rights reserved.

关键词： Knowledge based systems

来源：评论

学校读者我要写书评

暂无评论

ProtoMix: Learnable Data Augmentation on Few-Shot Features with Vector Quantization in CTR Prediction 9th

ProtoMix: Learnable Data Augmentation on Few-Shot Features ...

引用

19th International Conference on advanced Data Mining and Applications, ADMA 2023

作者： Zhao, Haijun Xu, Ronghai Wang, Chang-Dong Jiang, Ying School of Computer Science and Engineering Sun Yat-sen University Guangdong Guangzhou China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education Guangzhou China

ISBN: (纸本)9783031466601

Click-Through Rate (CTR) prediction is a critical problem in recommendation systems since it involves enormous business interest. Most deep CTR model follows an Embedding & Feature Interaction paradigm. However, the feature interaction module cannot work well without a good embedding representation of features. Due to the long-tail phenomenon in real scenes, few samples are provided in the dataset for a large proportion of features. In this paper, we present ProtoMix, a model-agnostic framework for learnable data augmentation on few-shot features in CTR prediction. ProtoMix automatically extracts information from co-occurred features within the same instance to assign prototype embedding with vector quantization for few-shot features and further synthesize the embedding representation of the augmented virtual instance for training. Original embedding, feature interaction module, and the embedding generator are jointly trained on a well-designed objective in an end-to-end manner in ProtoMix. We experimentally validate the effectiveness and compatibility of ProtoMix by comparing it with baseline and other data augmentation methods on different deep CTR models and multiple real-world CTR benchmark datasets. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

ALGCN: Accelerated Light Graph Convolution Network for Recommendation 28th

ALGCN: Accelerated Light Graph Convolution Network for Reco...

引用

28th International Conference on Database Systems for advanced Applications, DASFAA 2023

作者： Xu, Ronghai Zhao, Haijun Li, Zhi-Yuan Wang, Chang-Dong School of Computer Science and Engineering Sun Yat-sen University Guangdong Guangzhou China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education Beijing China

ISBN: (纸本)9783031306716

Recently, Graph Convolutional Network (GCN) has been widely applied in the field of collaborative filtering (CF) with tremendous success, since its message-passing mechanism can efficiently aggregate neighborhood information between users and items. However, most of the existing GCN-based CF models suffer from low convergence rates during training, mainly because they follow the design of standard GCN using a simple uniform average to aggregate the neighborhood information. We also find that the scale of embedding across different layers oscillates. We argue that these issues can be alleviated by our proposed graph convolution framework, namely Accelerated Light Graph Convolutional Network (ALGCN). ALGCN mainly contains two components: influence-aware graph convolution operation and augmentation-free in-batch contrastive loss on the unit sphere. Empirical evaluations on three large and public datasets demonstrate that the proposed method achieves remarkable training speedups over LightGCN and substantially outperforms the state-of-the-art GCN-based CF models. Our method also shows a great improvement in long-tail recommendation. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Parameter-efficient fine-tuning of Whisper for low-resource speech recognition

Parameter-efficient fine-tuning of Whisper for low-resource ...

引用

Artificial intelligence, Networking and Information Technology (AINIT), International Seminar on

作者： Yunpeng Liu Dan Qu Department of Artificial Intelligence Laboratory for Advanced Computing and Intelligence Engineering School of Information Systems Engineering University of Information Engineering Henan China

ISBN: (数字)9798350385557

ISBN: (纸本)9798350385564

Limited data availability remains a significant challenge for Whisper's low-resource speech recognition performance, falling short of practical application requirements. While previous studies have successfully reduced the recognition error rates of target language speech through fine-tuning, a comprehensive exploration and analysis of Whisper's parameter-efficient fine-tuning capabilities, the advantages and disadvantages are still lacking. This paper aims to fill this gap by conducting comprehensive experimental exploration for Whisper's low-resource speech recognition performance using two parameter-efficient fine-tuning strategies with limited supervised data from multiple low-resource languages. The results and analysis demonstrate that all fine-tuning strategies explored in this paper significantly enhance Whisper's performance. However, different strategies vary in their suitability and practical effectiveness, highlighting the need for careful selection based on specific use cases and resources available.

关键词： Seminars Costs Target recognition Error analysis Speech recognition Data models Task analysis

来源：评论

学校读者我要写书评

暂无评论

Mitigating Barren Plateaus of Variational Quantum Eigensolvers

IEEE Transactions on Quantum Engineering

引用

IEEE Transactions on Quantum engineering 2024年第1期5卷 1-19页

作者： Liu, Xia Liu, Geng Zhang, Hao-Kai Huang, Jiaxin Wang, Xin Thrust of Artificial Intelligence Information Hub Guangzhou510530 China Institute of Information Engineering Chinese Academy of Sciences Key Laboratory of Cyberspace Security Defense Beijing100085 China Baidu Research Institute for Quantum Computing Beijing100193 China Tsinghua University Institute for Advanced Study Beijing100084 China

Variational quantum algorithms (VQAs) are expected to establish valuable applications on near-term quantum computers. However, recent works have pointed out that the performance of VQAs greatly relies on the expressibility of the ansatzes and is seriously limited by optimization issues, such as barren plateaus (i.e., vanishing gradients). This article proposes the state-efficient ansatz (SEA) for accurate ground state preparation with improved trainability. We show that the SEA can generate an arbitrary pure state with much fewer parameters than a universal ansatz, making it efficient for tasks like ground state estimation. Then, we prove that barren plateaus can be efficiently mitigated by the SEA and the trainability can be further improved most quadratically by flexibly adjusting the entangling capability of the SEA. Finally, we investigate a plethora of examples in ground state estimation where we obtain significant improvements in the magnitude of the cost gradient and the convergence speed. © 2020 IEEE.

关键词： Timing circuits

来源：评论

学校读者我要写书评

暂无评论

Pushing AI to wireless network edge: an overview on integrated sensing, communication, and computation towards 6G

引用

Science China(Information Sciences) 2023年第3期66卷 7-25页

作者： Guangxu ZHU Zhonghao LYU Xiang JIAO Peixi LIU Mingzhe CHEN Jie XU Shuguang CUI Ping ZHANG Shenzhen Research Institute of Big Data Future Network of Intelligence Institute (FNii) The Chinese University of Hong Kong (Shenzhen) School of Science and Engineering (SSE) The Chinese University of Hong Kong (Shenzhen) State Key Laboratory of Advanced Optical Communication Systems and Networks School of Electronics Peking University Department of Electrical and Computer Engineering and Institute for Data Science and Computing University of Miami Peng Cheng Laboratory State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications

Pushing artificial intelligence(AI) from central cloud to network edge has reached board consensus in both industry and academia for materializing the vision of artificial intelligence of things(AIoT) in the sixth-generation(6G) era. This gives rise to an emerging research area known as edge intelligence, which concerns the distillation of human-like intelligence from the vast amount of data scattered at the wireless network edge. Typically, realizing edge intelligence corresponds to the processes of sensing, communication,and computation, which are coupled ingredients for data generation, exchanging, and processing, ***, conventional wireless networks design the three mentioned ingredients separately in a task-agnostic manner, which leads to difficulties in accommodating the stringent demands of ultra-low latency, ultra-high reliability, and high capacity in emerging AI applications like auto-driving and metaverse. This thus prompts a new design paradigm of seamlessly integrated sensing, communication, and computation(ISCC) in a taskoriented manner, which comprehensively accounts for the use of the data in downstream AI tasks. In view of its growing interest, this study provides a timely overview of ISCC for edge intelligence by introducing its basic concept, design challenges, and enabling techniques, surveying the state-of-the-art advancements, and shedding light on the road ahead.

关键词： sixth-generation (6G) edge intelligence artificial intelligence of things (AIoT) integrated sensing communication and computation (ISCC)

来源：评论

学校读者我要写书评

暂无评论

iGrasp: An Interactive 2D-3D Framework for 6-DoF Grasp Detection 27th

iGrasp: An Interactive 2D-3D Framework for 6-DoF Grasp Detec...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Jiang, Jian-Jian Wu, Xiao-Ming Chen, Zibo Wei, Yi-Lin Zheng, Wei-Shi School of Computer Science and Engineering Sun Yat-sen University Guangzhou China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education Guangzhou China Guangdong Province Key Laboratory of Information Security Technology Sun Yat-sen University Guangzhou China

ISBN: (纸本)9783031781124

For 6-DoF grasp detection, we aim at introducing a new interactive 2D-3D framework which filters out irrelevant information and makes both modalities collaborate effectively to generate robust grasps and accelerate inference speed greatly. This cannot be accomplished by existing works on learning to grasp that merely utilize 3D point clouds or leverage both 2D textures and 3D point clouds. Our framework is called iGrasp, a novel three-step design between 2D textures and 3D point clouds, where the interaction modelling enhances both modalities. Concretely, we propose the 2D-to-3D interaction to leverage objectness masks generated from 2D textures to filter out target-irrelevant information in 3D point clouds. Then, we introduce the 3D-to-2D interaction to leverage structural priors from 3D point cloud features with cross-attention and cylinder grouping to refine 2D texture features. Finally, we combine the refined 2D texture features and 3D point cloud features for generating high-quality 6-DoF grasp poses. Our experiments on the large-scale real-world dataset, namely GraspNet-1Billion, demonstrate that iGrasp surpasses state-of-the-art methods by 4.66/3.53 mAP on RealSense/Kinect and reduces the inference time by 28%. Real-world experiments further verify the effectiveness of iGrasp. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Depth-Enhanced Alignment for Label-Free 3D Semantic Segmentation 27th

Depth-Enhanced Alignment for Label-Free 3D Semantic Segment...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Xie, Shangjin Feng, Jiawei Chen, Zibo Liu, Zhixuan Zheng, Wei-Shi School of Computer Science and Engineering Sun Yat-sen University Guangzhou China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education Guangzhou China Guangdong Province Key Laboratory of Information Security Technology Sun Yat-sen University Guangzhou China

ISBN: (纸本)9783031784552

Labeling every point in a scene is a laborious journey for 3D understanding. To achieve annotation-free training, existing works introduce Contrastive Language-Image Pre-training (CLIP) to transfer the pre-trained capability of visual-linguistic correspondence to 3D-linguistic matching. However, directly adopting this CLIP-driven strategy can inevitably introduce bias: The overrated roles of the color and texture from an RGB image could overshadow the geometric nature of the corresponding 3D scene, resulting in a sub-optimal alignment. We note that different from RGB images, a depth map contains rich geometric information. Inspired by this, we propose Depth-Enhanced Alignment (D-EA) for label-free 3D semantic segmentation. D-EA aims to explore the rich geometric cues in depth maps and mitigate the color and texture biases rooted in the original CLIP-driven strategy. Specifically, we first tune a geometry-enhanced CLIP by aligning its depth prediction to the paired RGB prediction given by the original CLIP. Next, the point cloud feature space is matched with the RGB-Depth aggregated CLIP space by aligning point prediction to RGB and depth predictions. Moreover, to mitigate the semantic ambiguity caused by view-specific noise, we propose a View-Integrated Pseudo Label Generation paradigm. Experiments demonstrate the effectiveness of the proposed D-EA on the ScanNet (indoor) and GraspNet-1Billion (desktop) datasets in the label-free setting. Our method is also competitive in limited annotation semantic segmentation. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Privacy-Preserving Face Recognition with Adaptive Generative Perturbations 27th

Privacy-Preserving Face Recognition with Adaptive Generative...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Zhang, Delong Peng, Yixing Wu, Ancong Zheng, Wei-Shi School of Computer Science and Engineering Sun Yat-sen University Guangzhou China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education Guangzhou China Guangdong Province Key Laboratory of Information Security Technology Sun Yat-sen University Guangzhou China

ISBN: (纸本)9783031783401

For online face recognition services, the potential leakage of facial features and reconstruction techniques gives malicious attackers the opportunity to reconstruct face images, raising public concern about privacy. Previous privacy-preserving face recognition methods either require retraining the face recognition model or iterative perturbation that increases inference time. To overcome these limitations, we propose an efficient plug-and-play method that crafts Adaptive Generative perturbations for frozen Face recognition model (AGFace) to defend the reconstruction attacks. To generate perturbation with a single forward for shallow features extracted by frozen recognition model, we learn a Perturbation Generator to efficiently mine the adversarial perturbations by simultaneously minimizing the perturbation and maximizing the reconstruction error of a proxy reconstructor. To achieve privacy-utility trade-off, we propose the Adaptive Channel Selector to identify top-k reconstruction-sensitive channels for the features. By selecting these channels for perturbation, the dominant visual privacy information is protected with reconstruction-insensitive discriminative information preserved. Extensive experiments demonstrate that AGFace achieves state-of-the-art performance in terms of both privacy and utility among retraining-free methods and is comparable to retraining-dependent methods. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Differential privacy

来源：评论

学校读者我要写书评

暂无评论

Towards Completeness: A Generalizable Action Proposal Generator for Zero-Shot Temporal Action Localization 27th

Towards Completeness: A Generalizable Action Proposal Genera...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Du, Jia-Run Lin, Kun-Yu Meng, Jingke Zheng, Wei-Shi School of Computer Science and Engineering Sun Yat-sen University Guangzhou China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education Guangzhou China Guangdong Province Key Laboratory of Information Security Technology Sun Yat-sen University Guangzhou China

ISBN: (纸本)9783031784439

To address the zero-shot temporal action localization (ZSTAL) task, existing works develop models that are generalizable to detect and classify actions from unseen categories. They typically develop a category-agnostic action detector and combine it with the Contrastive Language-Image Pre-training (CLIP) model to solve ZSTAL. However, these methods suffer from incomplete action proposals generated for unseen categories, since they follow a frame-level prediction paradigm and require hand-crafted post-processing to generate action proposals. To address this problem, in this work, we propose a novel model named Generalizable Action Proposal generator (GAP), which can interface seamlessly with CLIP and generate action proposals in a holistic way. Our GAP is built in a query-based architecture and trained with a proposal-level objective, enabling it to estimate proposal completeness and eliminate the hand-crafted post-processing. Based on this architecture, we propose an Action-aware Discrimination loss to enhance the category-agnostic dynamic information of actions. Besides, we introduce a Static-Dynamic Rectifying module that incorporates the generalizable static information from CLIP to refine the predicted proposals, which improves proposal completeness in a generalizable manner. Our experiments show that our GAP achieves state-of-the-art performance on two challenging ZSTAL benchmarks, i.e., Thumos14 and ActivityNet1.3. Specifically, our model obtains significant performance improvement over previous works on the two benchmarks, i.e., +3.2% and +3.4% average mAP, respectively. The code is available at https://***/Run542968/GAP. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Zero-shot learning

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：