检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Tang, Zhiwei Tang, Jiasheng Luo, Hao Wang, Fan Chang, Tsung-Hui School of Science and Engineering The Chinese University of Hong Kong Shenzhen China DAMO Academy Alibaba Group China Hupan Lab Zhejiang Province China Shenzhen Research Institute of Big Data Shenzhen China

Diffusion models have emerged as state-of-the-art generative models for image generation. However, sampling from diffusion models is usually time-consuming due to the inherent autoregressive nature of their sampling process. In this work, we propose a novel approach that accelerates the sampling of diffusion models by parallelizing the autoregressive process. Specifically, we reformulate the sampling process as solving a system of triangular nonlinear equations through fixed-point iteration. With this innovative formulation, we explore several systematic techniques to further reduce the iteration steps required by the solving process. Applying these techniques, we introduce ParaTAA, a universal and training-free parallel sampling algorithm that can leverage extra computational and memory resources to increase the sampling speed. Our experiments demonstrate that ParaTAA can decrease the inference steps required by common sequential sampling algorithms such as DDIM and DDPM by a factor of 4∼14 times. Notably, when applying ParaTAA with 100 steps DDIM for Stable Diffusion, a widely-used text-to-image diffusion model, it can produce the same images as the sequential sampling in only 7 inference steps. The code is available at https://***/TZW1998/ParaTAA-Diffusion. © 2024, CC BY.

关键词： Nonlinear equations

来源：评论

学校读者我要写书评

暂无评论

BADROBOT: JAILBREAKING EMBODIED LLMS IN THE PHYSICAL WORLD

arXiv

引用

arXiv 2024年

作者： Zhang, Hangtao Zhu, Chenyu Wang, Xianlong Zhou, Ziqi Yin, Changgan Li, Minghui Xue, Lulu Wang, Yichen Hu, Shengshan Liu, Aishan Guo, Peijin Zhang, Leo Yu National Engineering Research Center for Big Data Technology and System China Services Computing Technology and System Lab China Cluster and Grid Computing Lab China Hubei Engineering Research Center on Big Data Security China Hubei Key Laboratory of Distributed System Security China Beihang University China School of Cyber Science and Engineering Huazhong University of Science and Technology China School of Software Engineering Huazhong University of Science and Technology China School of Information and Communication Technology Griffith University Australia

Embodied AI represents systems where AI is integrated into physical entities. Large Language Model (LLM), which exhibits powerful language understanding abilities, has been extensively employed in embodied AI by facilitating sophisticated task planning. However, a critical safety issue remains overlooked: could these embodied LLMs perpetrate harmful behaviors? In response, we introduce BADROBOT, a novel attack paradigm aiming to make embodied LLMs violate safety and ethical constraints through typical voice-based user-system interactions. Specifically, three vulnerabilities are exploited to achieve this type of attack: (i) manipulation of LLMs within robotic systems, (ii) misalignment between linguistic outputs and physical actions, and (iii) unintentional hazardous behaviors caused by world knowledge’s flaws. Furthermore, we construct a benchmark of various malicious physical action queries to evaluate BADROBOT’s attack performance. Based on this benchmark, extensive experiments against existing prominent embodied LLM frameworks (e.g., Voxposer, Code as Policies, and ProgPrompt) demonstrate the effectiveness of our BADROBOT. Copyright © 2024, The Authors. All rights reserved.

关键词： Benchmarking

来源：评论

学校读者我要写书评

暂无评论

The periodic product recommendation system based on deep reinforcement learning and the multi-objective framework

The periodic product recommendation system based on deep rei...

引用

International Conference on Awareness science and Technology (iCAST)

作者： Dang Tien Dat Tran Ngoc Thang Nguyen Anh Minh Rung Ching Chen Nguyen Thi Ngoc Anh School of Applied Mathematics and Informatics Hanoi University of Science and Technology Vietnam Department of Information Management Chaoyang University of Technology Taiwan Big Data Lab CMC Institute of Science and Technology Vietnam

The primary objective of the recommendation system is to suggest suitable products to users. The need for a personalized recommendation system has become essential with the continuous growth in the number of users and the increasing diversity of products. In this paper, we proposed a novel approach that combines a reinforcement learning algorithm with user clustering to address the recommendation challenge for periodic products. Additionally, we extend our framework to accommodate multi-objective scenarios. Our contributions encompass the development of meaningful representation vectors for actions and states, along with the definition of reward vector aligning with users’ long-term preferences and optimizing profits. We apply our proposed method to the problem of recommending mobile packages in Vietnam. Notably, our approach yields favorable recommendation results and exhibits superior performance compared to alternative methods.

关键词：

来源：评论

学校读者我要写书评

暂无评论

PipePrune: Pipeline Parallel Based on Convolutional Layer Pruning for Distributed Deep Learning 23

PipePrune: Pipeline Parallel Based on Convolutional Layer Pr...

引用

23rd IEEE International Conference on High Performance Computing and Communications, 7th IEEE International Conference on data science and Systems, 19th IEEE International Conference on Smart City and 7th IEEE International Conference on Dependability in Sensor, Cloud and big data Systems and Applications, HPCC-DSS-SmartCity-DependSys 2021

作者： Tan, Daojun Jiang, Wenbin Qin, Shang Jin, Hai School of Computer Science and Technology Huazhong University of Science and Technology Natl. Eng. Res. Ctr. for Big Data Technol. and Syst. Serv. Comp. Technol. and Syst. Lab Cluster and Grid Comp. Lab Wuhan430074 China

ISBN: (纸本)9781665494571

Benefitting from the combination of the idea of pipeline with model parallelism and data parallelism, pipeline parallelism improves the efficiency of distributed deep learning systems significantly. However, suffering from the bubbles and gaps caused by the imbalance of different stages in pipeline, it can not output ideal performance yet. To explore the potential of pipeline parallelism further, we propose a novel approach called PipePrune, which adds a convolutional layer pruning strategy to pipeline creatively to reduce the bubbles and gaps. For the convolutional layers with heavy overheads, some unimportant kernels are pruned by the measurement of the L1-norm. This approach makes the processing overheads of different pipeline stages more balanced. The experimental results show that, compared with state-of-the-art pipeline methods, PipePrune can improve the training speeds obviously (e.g. for ResNet50 on ImageNet, about 30% speed improvement is realized with only 1.1% loss of training accuracies). © 2021 IEEE.

关键词： Training Deep learning Smart cities Pipelines Neural networks Distributed databases Parallel processing

来源：评论

学校读者我要写书评

暂无评论

On Pipelined GCN with Communication-Efficient Sampling and Inclusion-Aware Caching

On Pipelined GCN with Communication-Efficient Sampling and I...

引用

IEEE Annual Joint Conference: INFOCOM, IEEE Computer and Communications Societies

作者： Shulin Wang Qiang Yu Xiong Wang Yuqing Li Hai Jin 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 Huazhong University of Science and Technology Wuhan China School of Cyber Science and Engineering Wuhan University Wuhan China

ISBN: (数字)9798350383508

ISBN: (纸本)9798350383515

Graph convolutional network (GCN) has achieved enormous success in learning structural information from unstructured data. As graphs become increasingly large, distributed training for GCNs is severely prolonged by frequent cross-worker communications. Existing efforts to improve the training efficiency often come at the expense of GCN performance, while the communication overhead persists. In this paper, we propose PSC-GCN, a holistic pipelined framework for distributed GCN training with communication-efficient sampling and inclusion-aware caching, to address the communication bottleneck while ensuring satisfactory model performance. Specifically, we devise an asynchronous pre-fetching scheme to retrieve stale statistics (features, embedding, gradient) of boundary nodes in advance, such that the embedding aggregation and model update are pipelined with statistics transmission. To alleviate communication volume and staleness effect, we introduce a variance-reduction based sampling policy, which prioritizes inner nodes over boundary ones for reducing the access frequency to remote neighbors, thus mitigating cross-worker statistics exchange. Complementing graph sampling, a feature caching module is co-designed to buffer hot nodes with high inclusion probability, ensuring that frequently sampled nodes will be available in local memory. Extensive evaluations on real-world datasets show the superiority of PSC-GCN over state-of-the-art methods, where we can reduce training time by 72%-80% without sacrificing model accuracy.

关键词： Training Estimation error Accuracy Graph convolutional networks Computational modeling Probability Benchmark testing

来源：评论

学校读者我要写书评

暂无评论

SPA: a graph spectral alignment perspective for domain adaptation 23

SPA: a graph spectral alignment perspective for domain adapt...

引用

Proceedings of the 37th International Conference on Neural Information Processing Systems

作者： Zhiqing Xiao Haobo Wang Ying Jin Lei Feng Gang Chen Fei Huang Junbo Zhao College of Computer Science and Technology Zhejiang University and Key Lab of Intelligent Computing based Big Data of Zhejiang Province Zhejiang University School of Software Technology Zhejiang University and Key Lab of Intelligent Computing based Big Data of Zhejiang Province Zhejiang University CUHK-SenseTime Joint Lab The Chinese University of Hong Kong School of Computer Science and Engineering Nanyang Technological University Alibaba Group

Unsupervised domain adaptation (UDA) is a pivotal form in machine learning to extend the in-domain model to the distinctive target domains where the data distributions differ. Most prior works focus on capturing the inter-domain transferability but largely overlook rich intra-domain structures, which empirically results in even worse discriminability. In this work, we introduce a novel graph SPectral Alignment (SPA) framework to tackle the tradeoff. The core of our method is briefly condensed as follows: (i)-by casting the DA problem to graph primitives, SPA composes a coarse graph alignment mechanism with a novel spectral regularizer towards aligning the domain graphs in eigenspaces; (ii)-we further develop a finegrained message propagation module — upon a novel neighbor-aware self-training mechanism — in order for enhanced discriminability in the target domain. On standardized benchmarks, the extensive experiments of SPA demonstrate that its performance has surpassed the existing cutting-edge DA methods. Coupled with dense model analysis, we conclude that our approach indeed possesses superior efficacy, robustness, discriminability, and transferability. Code and data are available at: https://***/CrownX/SPA.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Towards Memory- and Time-Efficient Backpropagation for Training Spiking Neural Networks

Towards Memory- and Time-Efficient Backpropagation for Train...

引用

International Conference on Computer Vision (ICCV)

作者： Qingyan Meng Mingqing Xiao Shen Yan Yisen Wang Zhouchen Lin Zhi-Quan Luo The Chinese University of Hong Kong Shenzhen Shenzhen Research Institute of Big Data National Key Lab. of General AI School of Intelligence Science and Technology Peking University Center for Data Science Peking University Institute for Artificial Intelligence Peking University Peng Cheng Laboratory

Spiking Neural Networks (SNNs) are promising energy-efficient models for neuromorphic computing. For training the non-differentiable SNN methods, the backpropagation through time (BPTT) with surrogate gradients (SG) method has achieved high performance. However, this method suffers from considerable memory cost and training time during training. In this paper, we propose the Spatial Learning Through Time (SLTT) method that can achieve high performance while greatly improving training efficiency compared with BPTT. First, we show that the backpropagation of SNNs through the temporal domain contributes just a little to the final calculated gradients. Thus, we propose to ignore the unimportant routes in the computational graph during backpropagation. The proposed method reduces the number of scalar multiplications and achieves a small memory occupation that is independent of the total time steps. Furthermore, we propose a variant of SLTT, called SLTT-K, that allows backpropagation only at K time steps, then the required number of scalar multiplications is further reduced and is independent of the total time steps. Experiments on both static and neuromorphic datasets demonstrate superior training efficiency and performance of our SLTT. In particular, our method achieves state-of-the-art accuracy on ImageNet, while the memory cost and training time are reduced by more than 70% and 50%, respectively, compared with BPTT. Our code is available at https://***/qymeng94/SLTT.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Enabling Efficient Large Recommendation Model Training with Near CXL Memory Processing

Enabling Efficient Large Recommendation Model Training with ...

引用

Annual International Symposium on Computer Architecture, ISCA

作者： Haifeng Liu Long Zheng Yu Huang Jingyi Zhou Chaoqiang Liu Runze Wang Xiaofei Liaot Hai Jinf Jingling Xue 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 School of Computer Science and Engineering University of New South Wales Australia Zhejiang Lab Hangzhou China

ISBN: (数字)9798350326581

ISBN: (纸本)9798350326598

Personalized recommendation systems have become one of the most important Internet services nowadays. A critical challenge of training and deploying the recommendation models is their high memory capacity and bandwidth demands, with the embedding layers occupying hundreds of GBs to TBs of storage. The advent of memory disaggregation technology and Compute Express Link (CXL) provides a promising solution for memory capacity scaling. However, relocating memory-intensive embedding layers to CXL memory incurs noticeable performance degradation due to its limited transmission bandwidth, which is significantly lower than the host memory bandwidth. To address this, we introduce ReCXL, a CXL memory disaggregation system that utilizes near-memory processing for scalable, efficient recommendation model training. ReCXL features a unified, hardwareefficient NMP architecture that processes the entire embedding training within CXL memory, minimizing data transfers over the bandwidth-limited CXL and enhancing internal bandwidth. To further improve the performance, ReCXL incorporates softwarehardware co-optimizations, including sophisticated dependencyfree prefetching and fine-grained update scheduling, to maximize hardware utilization. Evaluation results show that ReCXL outperforms the CPU-GPU baseline and the naïve CXL memory by $7.1 \times \sim 10.6 \times(9.4 \times$ on average) and $12.7 \times \sim 31.3 \times(22.6 \times$ on average), respectively.

关键词： Training Degradation Scalability Prefetching Memory management Web and internet services Bandwidth

来源：评论

学校读者我要写书评

暂无评论

Rethinking CLIP-based Video Learners in Cross-Domain Open-Vocabulary Action Recognition

arXiv

引用

arXiv 2024年

作者： Lin, Kun-Yu Ding, Henghui Zhou, Jiaming Tang, Yu-Ming Peng, Yi-Xing Zhao, Zhilin Loy, Chen Change Zheng, Wei-Shi School of Computer Science and Engineering Sun Yat-sen University China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education China S-Lab Nanyang Technological University Singapore Institute of Big Data Fudan University China AI Thrust Hong Kong University of Science and Technology Guangzhou China Data Science Lab School of Computing & DataX Research Centre Macquarie University Australia

Building upon the impressive success of CLIP (Contrastive Language-Image Pretraining), recent pioneer works have proposed to adapt the powerful CLIP to video data, leading to efficient and effective video learners for open-vocabulary action recognition. Inspired by that humans perform actions in diverse environments, our work delves into an intriguing question: Can CLIP-based video learners effectively generalize to video domains they have not encountered during training? To answer this, we establish a CROSS-domain Open-Vocabulary Action recognition benchmark named XOV-Action, and conduct a comprehensive evaluation of five state-of-the-art CLIP-based video learners under various types of domain gaps. The evaluation demonstrates that previous methods exhibit limited action recognition performance in unseen video domains, revealing potential challenges of the cross-domain open-vocabulary action recognition task. In this paper, we focus on one critical challenge of the task, namely scene bias, and accordingly contribute a novel scene-aware video-text alignment method. Our key idea is to distinguish video representations apart from scene-encoded text representations, aiming to learn scene-agnostic video representations for recognizing actions across domains. Extensive experiments demonstrate the effectiveness of our method. The benchmark and code will be available at https://***/KunyuLin/XOV-Action/. Copyright © 2024, The Authors. All rights reserved.

关键词： Video recording

来源：评论

学校读者我要写书评

暂无评论

Aligning Language Models Using Follow-up Likelihood as Reward Signal 39

Aligning Language Models Using Follow-up Likelihood as Rewar...

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Zhang, Chen Chong, Dading Jiang, Feng Tang, Chengguang Gao, Anningzhe Tang, Guohua Li, Haizhou National University of Singapore Singapore Peking University China The Chinese University of Hong Kong Shenzhen China Shenzhen Research Institute of Big Data China University of Science and Technology of China China Tencent AI Lab China

ISBN: (纸本)157735897X

In natural human-to-human conversations, participants often receive feedback signals from one another based on their follow-up reactions. These reactions can include verbal responses, facial expressions, changes in emotional state, and other non-verbal cues. Similarly, in human-machine interactions, the machine can leverage the user’s follow-up utterances as feedback signals to assess whether it has appropriately addressed the user’s request. Therefore, we propose using the likelihood of follow-up utterances as rewards to differentiate preferred responses from less favored ones, without relying on human or commercial LLM-based preference annotations. Our proposed reward mechanism, “Follow-up Likelihood as Reward” (FLR), matches the performance of strong reward models trained on large-scale human or GPT-4 annotated data on 8 pairwise-preference and 4 rating-based benchmarks. Building upon the FLR mechanism, we propose to automatically mine preference data from the online generations of a base policy model. The preference data are subsequently used to boost the helpfulness of the base model through direct alignment from preference (DAP) methods, such as direct preference optimization (DPO). Lastly, we demonstrate that fine-tuning the language model that provides follow-up likelihood with natural language feedback significantly enhances FLR’s performance on reward modeling benchmarks and effectiveness in aligning the base policy model’s helpfulness. © 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

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：