检索结果-内蒙古大学图书馆

Deep Learning Operators Performance Tuning for Changeable Sized Input Data on Tensor Accelerate Hardware

IEEE Transactions on computers 2025年第6期74卷 2101-2113页

作者： Mu, Pengyu Liu, Yi Wang, Rui Liu, Guoxiang An, Hangcheng Zhao, Qianhe Yang, Hailong Xie, Chenhao Luan, Zhongzhi Gong, Chunye Qian, Depei Beihang University School of Computer Science and Engineering Beijing100191 China Beihang University Hangzhou Innovation Institute Hangzhou310051 China Shanghai Artificial Intelligence Laboratory Shanghai200003 China National Supercomputer Center Tianjin300457 China

The operator library is the fundamental infrastructure of deep learning acceleration hardware. Automatically generating the library and tuning its performance is promising because the manual development by well-trained and skillful programmers is costly in terms of both time and money. Tensor hardware has the best computing efficiency for deep learning applications, but the operator library programs are hard to tune because the tensor hardware primitives have many limitations. Otherwise, the performance is difficult to be fully explored. The recent advancement in LLM exacerbates this problem because the size of input data is not fixed. Therefore, mapping the computing tasks of operators to tensor hardware units is a significant challenge when the shape of the input tensor is unknown before the runtime. We propose DSAT, a deep learning operator performance autotuning technique for changeable-sized input data on tensor hardware. To match the input tensor's undetermined shape, we choose a group of abstract computing units as the basic building blocks of operators for changeable-sized input tensor shapes. We design a group of programming tuning rules to construct a large exploration space of the variant implementation of the operator programs. Based on these rules, we construct an intermediate representation of computing and memory access to describe the computing process and use it to map the abstract computing units to tensor primitives. To speed up the tuning process, we narrow down the optimization space by predicting the actual hardware resource requirement and providing an optimized cost model for performance prediction. DSAT achieves performance comparable to the vendor's manually tuned operator libraries. Compared to state-of-the-art deep learning compilers, it improves the performance of inference by 13% on average and decreases the tuning time by an order of magnitude. © 1968-2012 IEEE.

关键词： Tensors

来源：评论

学校读者我要写书评

暂无评论

Machine Learning for Brain Imaging Genomics Methods:A Review

引用

Machine intelligence Research 2023年第1期20卷 57-78页

作者： Mei-Ling Wang Wei Shao Xiao-Ke Hao Dao-Qiang Zhang College of Computer Science and Technology Nanjing University of Aeronautics and AstronauticsNanjing 211106China Key Laboratory of Pattern Analysis and Machine Intelligence Ministry of Industry and Information TechnologyNanjing 211106China School of Artificial Intelligence Hebei University of TechnologyTianjin 300401China

In the past decade,multimodal neuroimaging and genomic techniques have been increasingly *** an interdiscip-linary topic,brain imaging genomics is devoted to evaluating and characterizing genetic variants in individuals that influence phenotyp-ic measures derived from structural and functional brain *** technique is capable of revealing the complex mechanisms by macroscopic intermediates from the genetic level to cognition and psychiatric disorders in *** is well known that machine learn-ing is a powerful tool in the data-driven association studies,which can fully utilize priori knowledge(intercorrelated structure informa-tion among imaging and genetic data)for association *** addition,the association study is able to find the association between risk genes and brain structure or function so that a better mechanistic understanding of behaviors or disordered brain functions is *** this paper,the related background and fundamental work in imaging genomics are first ***,we show the univari-ate learning approaches for association analysis,summarize the main idea and modelling in genetic-imaging association studies based on multivariate machine learning,and present methods for joint association analysis and outcome ***,this paper discusses some prospects for future work.

关键词： Brain imaging genomics machine learning multivariate analysis association analysis outcome prediction

来源：评论

学校读者我要写书评

暂无评论

Fault Tolerance in The IoT: A Taxonomy Based on Techniques 3

Fault Tolerance in The IoT: A Taxonomy Based on Techniques

引用

3rd International Conference on Theoretical and Applicative Aspects of computer science, ICTAACS 2023

作者： Oussane, Sakina Benkaouha, Haroun Djouama, Amir Usthb University Lsi Laboratory Computer Science Faculty Algiers Algeria National School of Artificial Intelligence Lsi Laboratory Algiers Algeria

ISBN: (纸本)9798350385854

In order to ensure that IoT systems are reliable, efficient and robust, it is important for failures to be easily detectable and effectively managed to minimize their impact on users. This is why fault tolerance approaches are necessary in the Internet of Things (IoT). This article provides an overview of failures in IoT systems and proposes a classification of fault tolerance techniques. It also explores various existing fault tolerance approaches that can be used to develop effective solutions. © 2023 IEEE.

关键词： Fault tolerance

来源：评论

学校读者我要写书评

暂无评论

Joint Ability Assessment for Talent Recruitment: A Neural Cognitive Diagnosis Approach

引用

ACM Transactions on Management Information Systems 2025年第2期16卷 1-25页

作者： Ma, Haiping Li, Manwei Qin, Chuan Shen, Dazhong Zhu, Hengshu Zhang, Xingyi Xiong, Hui Institutes of Physical Science and Information Technology Anhui University Hefei China Computer Network Information Center Chinese Academy of Sciences Beijing China University of Science and Technology of China Hefei China Shanghai Artificial Intelligence Laboratory Shanghai China School of Computer Science and Technology Anhui University Hefei China The Hong Kong University of Science and Technology Guangzhou Campus Guangzhou China

Ability assessment is a critical task in talent recruitment that aims at identifying the most suitable job candidates by evaluating the alignment of their skills with job requirements. Indeed, traditional ability assessment involves multiple staffing processes with various forms of evaluation methods, such as written tests and face-to-face interviews, which usually result in fragmented, noisy, and inconsistent conclusions. Therefore, a long-standing challenge in talent recruitment is how to comprehensively evaluate candidates by integrating multi-source heterogeneous assessment results. To this end, in this article, we propose a holistic framework, JCD-TR (Joint Cognitive Diagnosis for Talent Recruitment), for enhancing the performance of ability assessment in talent recruitment by jointly modeling the multi-source heterogeneous assessment results. Specifically, we first construct a skill graph based on the co-occurrence relations of skills in multi-source recruitment data. Along this line, we can learn the skill representations that maintain both the semantic and structural information with graph embedding. Then, we design a multi-source candidate ability profiling module with the guidance of item response theory in psychometrics and the neural topic model. As a result, the candidates' ability profiles can be explored from their resumes, written tests, and interview assessment data, respectively. Furthermore, we propose a joint cognitive diagnosis module by integrating those multi-view ability profiles and skill representations to assess the candidates' skill proficiency state. Extensive experiments on a real-world dataset demonstrate the effectiveness of our JCD-TR. © 2025 Copyright held by the owner/author(s). Publication rights licensed to ACM.

关键词： Ability testing

来源：评论

学校读者我要写书评

暂无评论

On Unsupervised Uncertainty-Driven Speech Pseudo-Label Filtering and Model Calibration 48

On Unsupervised Uncertainty-Driven Speech Pseudo-Label Filte...

引用

48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023

作者： Dawalatabad, Nauman Khurana, Sameer Laurent, Antoine Glass, James MIT Computer Science and Artificial Intelligence Laboratory CambridgeMA United States LIUM - Le Mans University France

ISBN: (纸本)9781728163277

Pseudo-label (PL) filtering forms a crucial part of Self-Training (ST) methods for unsupervised domain adaptation. Dropout-based Uncertainty-driven Self-Training (DUST) proceeds by first training a teacher model on source domain labeled data. Then, the teacher model is used to provide PLs for the unlabeled target domain data. Finally, we train a student on augmented labeled and pseudo-labeled data. The process is iterative, where the student becomes the teacher for the next DUST iteration. A crucial step that precedes the student model training in each DUST iteration is filtering out noisy PLs that could lead the student model astray. In DUST, we proposed a simple, effective, and theoretically sound PL filtering strategy based on the teacher model's uncertainty about its predictions on unlabeled speech utterances. We estimate the model's uncertainty by computing disagreement amongst multiple samples drawn from the teacher model during inference by injecting noise via dropout. In this work, we show that DUST's PL filtering, as initially used, fail under severe source and target domain mismatch. We suggest several approaches to eliminate or alleviate this issue. Further, we bring insights from the research in neural network model calibration to DUST and show that a well-calibrated model correlates strongly with a positive outcome of the DUST PL filtering step. We demonstrate effectiveness of our methods on target domain GigaSpeech YouTube dataset. © 2023 IEEE.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Stress-Testing of Multimodal Models in Medical Image-Based Report Generation 39

Stress-Testing of Multimodal Models in Medical Image-Based R...

引用

39th Annual AAAI Conference on artificial intelligence, AAAI 2025

作者： Carvalhido, Flávia Cardoso, Henrique Lopes Cerqueira, Vítor LIACC - Artificial Intelligence and Computer Science Laboratory Faculty of Engineering University of Porto Rua Dr. Roberto Frias Porto 4200-465 Portugal

ISBN: (纸本)157735897X

Multimodal models, namely vision-language models, present unique possibilities through the seamless integration of different information mediums for data generation. These models mostly act as a black-box, making them lack transparency and explicability. Reliable results require accountable and trustworthy artificial intelligence (AI), namely when in use for critical tasks, such as the automatic generation of medical imaging reports for healthcare diagnosis. By exploring stress-testing techniques, multimodal generative models can become more transparent by disclosing their shortcomings, further supporting their responsible usage in the medical field. Copyright © 2025, Association for the Advancement of artificial intelligence (***). All rights reserved.

关键词： Report generators

来源：评论

学校读者我要写书评

暂无评论

LQTTrack:Multi-Object Tracking by Focusing on Low-Quality Targets Association

引用

computers, Materials & Continua 2024年第10期81卷 1449-1470页

作者： Suya Li Ying Cao Hengyi Ren Dongsheng Zhu Xin Xie Henan Key Laboratory of Big Data Analysis and Processing School of Computer and Information EngineeringHenan UniversityKaifeng475001China College of Information Science and Technology&Artificial Intelligence Nanjing Forestry UniversityNanjing210037China School of Computer Nanjing University of Posts and TelecommunicationsNanjing210023China

Multi-object tracking(MOT)has seen rapid improvements in recent ***,frequent occlusion remains a significant challenge in MOT,as it can cause targets to become smaller or disappear entirely,resulting in lowquality targets,leading to trajectory interruptions and reduced tracking *** from some existing methods,which discarded the low-quality targets or ignored low-quality target ***,with a lowquality association strategy(LQA),is proposed to pay more attention to low-quality *** the association scheme of LQTTrack,firstly,multi-scale feature fusion of FPN(MSFF-FPN)is utilized to enrich the feature information and assist in subsequent data ***,the normalized Wasserstein distance(NWD)is integrated to replace the original Inter over Union(IoU),thus overcoming the limitations of the traditional IoUbased methods that are sensitive to low-quality targets with small sizes and enhancing the robustness of low-quality target ***,the third association stage is proposed to improve the matching between the current frame’s low-quality targets and previously interrupted trajectories from earlier frames to reduce the problem of track fragmentation or error tracking,thereby increasing the association success rate and improving overall multi-object tracking *** experimental results demonstrate the competitive performance of LQTTrack on benchmark datasets(MOT17,MOT20,and DanceTrack).

关键词： Low-quality targets association strategy feature fusion multi-object tracking tracking-by-detection

来源：评论

学校读者我要写书评

暂无评论

StoryLLaVA: Enhancing Visual Storytelling with Multi-Modal Large Language Models 31

StoryLLaVA: Enhancing Visual Storytelling with Multi-Modal L...

引用

31st International Conference on Computational Linguistics, COLING 2025

作者： Yang, Li Xiao, Zhiding Huang, Wenxin Zhong, Xian Hubei Key Laboratory of Big Data Intelligent Analysis and Application School of Computer Science and Information Engineering Hubei University China Hubei Key Laboratory of Transportation Internet of Things School of Computer Science and Artificial Intelligence Wuhan University of Technology China

ISBN: (纸本)9798891761964

The rapid development of multi-modal large language models (MLLMs) has positioned visual storytelling as a crucial area in content creation. However, existing models often struggle to maintain temporal, spatial, and narrative coherence across image sequences and frequently lack the depth and engagement of human-authored stories. To address these challenges, we propose Story with Large Language and Vision Assistant (StoryLLaVA), a novel framework for enhancing visual storytelling. Our approach introduces a Topic-Driven Narrative Optimizer (TDNO) that improves both the training data and MLLM models by integrating image descriptions, topic generation, and GPT-4-based refinements. Furthermore, we employ a preference-based ranked story sampling method that aligns model outputs with human storytelling preferences through positive-negative pairing. These two phases of the framework differ in their training methods: the former uses supervised fine-tuning, while the latter incorporates reinforcement learning with positive and negative sample pairs. Experimental results demonstrate that StoryLLaVA outperforms current models in visual relevance, coherence, and fluency, with LLM-based evaluations confirming the generation of richer and more engaging narratives. The enhanced dataset and model are available at https://***/XxxZzD/StoryLLaVA. © 2025 Association for Computational Linguistics.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Dynamic community detection algorithm based on hyperbolic graph convolution

引用

International Journal of Intelligent Computing and Cybernetics 2024年第3期17卷 632-653页

作者： Weijiang Wu Heping Tan Yifeng Zheng College of Information Science and Engineering/College of Artificial Intelligence China University of Petroleum BeijingBeijingChina Beijing Key Laboratory of Petroleum Data Mining China University of Petroleum BeijingBeijingChina School of Computer Science Minnan Normal UniversityZhangzhouChina

Purpose:Community detection is a key factor in analyzing the structural features of complex ***,traditional dynamic community detection methods often fail to effectively solve the problems of deep network information loss and computational complexity in hyperbolic *** address this challenge,a hyperbolic space-based dynamic graph neural network community detection model(HSDCDM)is ***/methodology/approach:HSDCDM first projects the node features into the hyperbolic space and then utilizes the hyperbolic graph convolution module on the Poincare and Lorentz models to realize feature fusion and information *** addition,the parallel optimized temporal memory module ensures fast and accurate capture of time domain information over extended ***,the community clustering module divides the community structure by combining the node characteristics of the space domain and the time *** evaluate the performance of HSDCDM,experiments are conducted on both artificial and real ***:Experimental results on complex networks demonstrate that HSDCDM significantly enhances the quality of community detection in hierarchical *** shows an average improvement of 7.29%in NMI and a 9.07%increase in ARI across datasets compared to traditional *** complex networks with nonEuclidean geometric structures,the HSDCDM model incorporating hyperbolic geometry can better handle the discontinuity of the metric space,provides a more compact embedding that preserves the data structure,and offers advantages over methods based on Euclidean geometry ***/value:This model aggregates the potential information of nodes in space through manifoldpreserving distribution mapping and hyperbolic graph topology ***,it optimizes the Simple Recurrent Unit(SRU)on the hyperbolic space Lorentz model to effectively extract time series data in hyperbolic space,thereby enhancing computing efficiency by eliminating t

关键词： Complex networks Dynamic community detection Hyperbolic graph neural network

来源：评论

学校读者我要写书评

暂无评论

Causal Representation Enhances Cross-Domain Named Entity Recognition in Large Language Models

引用

computers, Materials & Continua 2025年第5期83卷 2809-2828页

作者： Jiahao Wu Jinzhong Xu Xiaoming Liu Guan Yang Jie Liu School of Artificial Intelligence Zhongyuan University of TechnologyZhengzhou450007China School of Computer Science Zhongyuan University of TechnologyZhengzhou450007China Zhengzhou Key Laboratory of Text Processing and Image Understanding Zhengzhou450007China School of Information Science and Technology North China University of TechnologyBeijing100144China

Large language models cross-domain named entity recognition task in the face of the scarcity of large language labeled data in a specific domain,due to the entity bias arising from the variation of entity information between different domains,which makes large language models prone to spurious correlations problems when dealing with specific domains and *** order to solve this problem,this paper proposes a cross-domain named entity recognition method based on causal graph structure enhancement,which captures the cross-domain invariant causal structural representations between feature representations of text sequences and annotation sequences by establishing a causal learning and intervention module,so as to improve the utilization of causal structural features by the large languagemodels in the target domains,and thus effectively alleviate the false entity bias triggered by the false relevance problem;meanwhile,through the semantic feature fusion module,the semantic information of the source and target domains is effectively *** results show an improvement of 2.47%and 4.12%in the political and medical domains,respectively,compared with the benchmark model,and an excellent performance in small-sample scenarios,which proves the effectiveness of causal graph structural enhancement in improving the accuracy of cross-domain entity recognition and reducing false correlations.

关键词： Large language model entity bias causal graph structure

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：