检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Yuan, Xiaosong Shen, Chen Yan, Shaotian Zhang, Xiaofeng Xie, Liang Wang, Wenxiao Guan, Renchu Wang, Ying Ye, Jieping College of Computer Science and Technology Jilin University China Alibaba Cloud Computing China Shanghai Jiao Tong University China College of Software Zhejiang University China

Zero-shot Chain-of-Thought (CoT) prompting emerges as a simple and effective strategy for enhancing the performance of large language models (LLMs) in realworld reasoning tasks. Nonetheless, the efficacy of a singular, task-level prompt uniformly applied across the whole of instances is inherently limited since one prompt cannot be a good partner for all, a more appropriate approach should consider the interaction between the prompt and each instance meticulously. This work introduces an instance-adaptive prompting algorithm as an alternative zero-shot CoT reasoning scheme by adaptively differentiating good and bad prompts. Concretely, we first employ analysis on LLMs through the lens of information flow to detect the mechanism under zero-shot CoT reasoning, in which we discover that information flows from question to prompt and question to rationale jointly influence the reasoning results most. We notice that a better zero-shot CoT reasoning needs the prompt to obtain semantic information from the question, and then the rationale aggregates sufficient information from the question directly and via the prompt indirectly. On the contrary, lacking any of those would probably lead to a bad one. Stem from that, we further propose an instance-adaptive prompting strategy (IAP) for zero-shot CoT reasoning. Experiments conducted with LLaMA-2, LLaMA- 3, and Qwen on math, logic, and commonsense reasoning tasks (e.g., GSM8K, MMLU, Causal Judgement) obtain consistent improvement, demonstrating that the instance-adaptive zero-shot CoT prompting performs better than other task-level methods with some curated prompts or sophisticated procedures, showing the significance of our findings in the zero-shot CoT reasoning mechanism. Copyright © 2024, The Authors. All rights reserved.

关键词： Zero-shot learning

来源：评论

学校读者我要写书评

暂无评论

FEDGE: An Interference-Aware QoS Prediction Framework for Black-Box Scenario in IaaS clouds with Domain Generalization

FEDGE: An Interference-Aware QoS Prediction Framework for Bl...

引用

International Symposium on Parallel and Distributed Processing (IPDPS)

作者： Yunlong Cheng Xiuqi Huang Zifeng Liu Jiadong Chen Xiaofeng Gao Zhen Fang Yongqiang Yang Department of Computer Science and Engineering MoE Key Lab of Artificial Intelligence Shanghai Jiao Tong University Shanghai China Computing and Networking Innovation Lab Huawei Cloud Computing Technology Co. Ltd China

ISBN: (数字)9798350387117

ISBN: (纸本)9798350387124

Public cloud providers embrace multi-tenancy as a strategy to enhance the utilization and efficiency of resources. However, co-located virtual machines (VMs) suffer from qualityof-service (QoS) degradation caused by shared resource interference. Existing solutions for predicting QoS degradation often rely on the assumption of online access to application-level information. However, in a production environment, this assumption proves invalid as the VMs are black boxes to the providers. This intrinsic characteristic of the IaaS cloud necessitates the prediction model to generalize to unfamiliar applications and imposes specific criteria on the monitorable *** meet the black-box scenario under Infrastructure as a Service (IaaS) cloud computing, we present a novel framework, FEDGE, that can predict interference-aware QoS (IA-QoS) of co-located VMs using only low-level monitorable metrics before migration. Specifically, FEDGE utilizes a stochastic gates layer to select the most informative features from the high-dimensional resource and hardware metrics, which helps to reduce the monitoring overhead. Furthermore, we design a multi-domain MMD-based adversarial denoising autoencoder to regularize the learned hidden representations and prevent over-fitting on the source domains. Next, we employ a multi-layer perceptron (MLP) to accurately predict complex QoS degradation using the learned representations with domain generalization. Experimental results demonstrate that FEDGE outperforms other state-of-the-art methods in terms of both generalizability and effectiveness.

关键词： Measurement Degradation cloud computing Closed box Stochastic processes Quality of service Production

来源：评论

学校读者我要写书评

暂无评论

Approximating to the Real Translation Quality for Neural Machine Translation via Causal Motivated Methods

引用

ACM Transactions on Asian and Low-Resource Language Information Processing 2023年第5期22卷 1-26页

作者： Xuewen Shi Heyan Huang Ping Jian Yi-Kun Tang School of Computer Science and Technology Beijing Institute of Technology Beijing Engineering Research Center of High Volume Language Information Processing and Cloud Computing Applications Haidian District Beijing China

It is hard to evaluate translations objectively and accurately, which limits the applications of machine translation. In this article, we assume that the above phenomenon is caused by noise interference during translation evaluation, and we handle the problem through a perspective of causal inference. We assume that the observable translation score is affected by the unobservable true translation quality and some noise simultaneously. If there is a variable that is related to the noise and independent to the true translation quality, the related noise can be eliminated by removing the effect of that variable from the observed score. Based on the above causality hypothesis, this article studies the length bias problem of beam search for neural machine translation (NMT) and the input related noise problem of translation quality estimation (QE). For the NMT length bias problem, we conduct the experiments on four typical NMT tasks (Uyghur–Chinese, Chinese–English, English–German, and English–French) with different scales of datasets. Comparing with previous approaches, the proposed causal motivated method is model-agnostic and does not require supervised training. For QE tasks, we conduct the experiments on the WMT’20 submissions. Experimental results show that the denoised QE results gain better Pearson’s correlation scores with human assessed scores compared to the original submissions. Further analyses on the NMT and QE tasks also demonstrate the rationality of the empirical assumptions made on our methods.

关键词： Neural machine translation causal inference half-sibling regression quality estimation

来源：评论

学校读者我要写书评

暂无评论

Road Extraction with Satellite Images and Partial Road Maps

arXiv

引用

arXiv 2023年

作者： Xu, Qianxiong Long, Cheng Yu, Liang Zhang, Chen The School of Computer Science and Engineering Nanyang Technological University Singapore Alibaba Cloud Computing Alibaba Hangzhou China The Department of Computing Hong Kong Polytechnic University Hong Kong

Road extraction is a process of automatically generating road maps mainly from satellite images. Existing models all target to generate roads from the scratch despite that a large quantity of road maps, though incomplete, are publicly available (e.g. those from OpenStreetMap) and can help with road extraction. In this paper, we propose to conduct road extraction based on satellite images and partial road maps, which is new. We then propose a two-branch Partial to Complete Network (P2CNet) for the task, which has two prominent components: Gated Self-Attention Module (GSAM) and Missing Part (MP) loss. GSAM leverages a channel-wise self-attention module and a gate module to capture long-range semantics, filter out useless information, and better fuse the features from two branches. MP loss is derived from the partial road maps, trying to give more attention to the road pixels that do not exist in partial road maps. Extensive experiments are conducted to demonstrate the effectiveness of our model, e.g. P2CNet achieves state-of-the-art performance with the IoU scores of 70.71% and 75.52%, respectively, on the SpaceNet and OSM datasets. Copyright © 2023, The Authors. All rights reserved.

关键词： Remote sensing

来源：评论

学校读者我要写书评

暂无评论

Pronounced advance on typhoon track forecast with global convection-permitting model

引用

science Bulletin 2025年第9期70卷 1424-1428页

作者： Gu, Jun Zhao, Chun Li, Gudongze Feng, Jiawang Xu, Mingyue Du, Qiuyan Xia, Zihan Li, Yubin Chen, Guanghua Hao, Xiaoyu Chen, Junshi An, Hong Deep Space Exploration Laboratory/School of Earth and Space Sciences/CMA-USTC Laboratory of Fengyun Remote Sensing/State Key Laboratory of Fire Science/Institute of Advanced Interdisciplinary Research on High-Performance Computing Systems and Software University of Science and Technology of China Hefei230026 China Laoshan Laboratory Qingdao266237 China CAS Center for Excellence in Comparative Planetology University of Science and Technology of China Hefei230026 China School of Atmospheric Physics Nanjing University of Information Science and Technology Nanjing210044 China Key Laboratory of Cloud-Precipitation Physics and Severe Storms Institute of Atmospheric Physics Chinese Academy of Sciences Beijing100029 China School of Computer Science and Technology University of Science and Technology of China Hefei230026 China

来源：评论

学校读者我要写书评

暂无评论

A secured GA-WPC framework for scheduling the independent tasks in cloud environment

引用

Journal of Ambient Intelligence and Humanized computing 2023年第9期14卷 13003-13015页

作者： Dubey, Kalka Sharma, S.C. Kumar, Mohit Kumar, Prashant Nasr, Aida A. Cloud Computing and Wireless Sensor Lab Indian Institute of Technology Roorkee Uttarakhand Roorkee247667 India Department of Computer Science and Engineering Rajiv Gandhi Institute of Petroleum Technology Jais Uttar Pradesh Amethi229304 India Department of Information Technology Dr BR Ambedkar National Institute of Technology Punjab Jalandhar144011 India Information Technology Department Faculty of Computers and Informatics Tanta University Tanta Egypt

cloud computing proved itself a powerful and popular platform for providing computing services. It offers on-demand services to users with highly efficient resources over the internet. With the exponential increase in the number of internet users and due to the high acceptance rate to use the cloud services in everything, the demand for cloud platforms has also increased significantly. This leads to heavy workloads on the servers, which is a difficult task to manage due to the heterogeneity and uncertainty of the resources. However, several traditional, heuristic, and evolutionary-based approaches have been proposed to solve the problem of scheduling. Existing state-of-art techniques and frameworks have solved the problem up to a limited level. Still, these solutions have not given any effective and concrete solution for the overall resource scheduling problem. Therefore, this paper tries to overcome the bottleneck of existing solutions by presenting a new framework named Genetic Algorithm Water Pressure Change (GAWPC) to solve task scheduling problems efficiently. The GAWPC framework exploited the feature of the Genetic Algorithm (GA) and Water Pressure Change optimization (WPCO) techniques to form a new hybrid mechanism. In the proposed framework, GA has been deployed to find the optimal solution for a particular population. WPCO has been used to optimize fitness function for a fast convergence rate. The computer-based simulation has demonstrated promising benefits like improved computation cost, make-span time, and load balancing when compared with the most recent existing baseline techniques using cloudSim. © 2022, The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Historical Information-Based Intent Detection for Multiturn Dialogue 22

Historical Information-Based Intent Detection for Multiturn ...

引用

8th International Conference on computing and Artificial Intelligence, ICCAI 2022

作者： Sun, Xin Zheng, Hongchao Tang, Zheng Beijing Eng. Res. Ctr. of High Vol. Language Information Processing and Cloud Computing Applications School of Computer Science and Technology Beijing Institute of Technology Beijing100081 China School of Computer Science and Technology Beijing Institute of Technology Beijing100081 China

ISBN: (纸本)9781450396110

Intent detection aims to determine the intent of users, an important task in natural language processing and dialogue systems. As one of the key modules of task-based dialogue systems, intent detection directly influences the meaning analysis of spoken language and the performance of dialogue systems. However, existing intent detection algorithms mainly focus on single-turn conversation, ignoring the valuable historical conversation message. To alleviate the problem, we propose a multiturn dialogue intent detection model based on historical information. Specifically, the semantic information of user statements and system statements is extracted by a role-distinguished self-attention network. Then, semantic relevance based on context information adjustment is used to combine the system statement and user statement. Finally, long short-term memory (LSTM) is used to capture the historical message. Experimental results prove that the proposed method improves the effect of intent detection and performs better than seven mainstream baselines. © 2022 ACM.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

Exploring and Evaluating Hallucinations in LLM-Powered Code Generation

arXiv

引用

arXiv 2024年

作者： Liu, Fang Liu, Yang Shi, Lin Huang, Houkun Wang, Ruifeng Yang, Zhen Zhang, Li Li, Zhongqi Ma, Yuchi School of Computer Science and Engineering Beihang University Beijing China School of Software Beihang University Beijing China School of Computer Science and Technology Shandong University Qingdao China Huawei Cloud Computing Technologies Co. Ltd China

The rise of Large Language Models (LLMs) has significantly advanced many applications on software engineering tasks, particularly in code generation. Despite the promising performance, LLMs are prone to generate hallucinations, which means LLMs might produce outputs that deviate from users’ intent, exhibit internal inconsistencies, or misalign with the factual knowledge, making the deployment of LLMs potentially risky in a wide range of applications. Existing work mainly focuses on investing the hallucination in the domain of natural language generation (NLG), leaving a gap in understanding the types and extent of hallucinations in the context of code generation. To bridge the gap, we conducted a thematic analysis of the LLM-generated code to summarize and categorize the hallucinations present in it. Our study established a comprehensive taxonomy of hallucinations in LLM-generated code, encompassing 5 primary categories of hallucinations depending on the conflicting objectives and varying degrees of deviation observed in code generation. Furthermore, we systematically analyzed the distribution of hallucinations, exploring variations among different LLMs and their correlation with code correctness. Based on the results, we proposed HALLUCODE, a benchmark for evaluating the performance of code LLMs in recognizing hallucinations. Hallucination recognition and mitigation experiments with HALLUCODE and HumanEval show existing LLMs face great challenges in recognizing hallucinations, particularly in identifying their types, and are hardly able to mitigate hallucinations. We believe our findings will shed light on future research about hallucination evaluation, detection, and mitigation, ultimately paving the way for building more effective and reliable code LLMs in the future. 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

Plan-on-Graph: Self-Correcting Adaptive Planning of Large Language Model on Knowledge Graphs

arXiv

引用

arXiv 2024年

作者： Chen, Liyi Tong, Panrong Jin, Zhongming Sun, Ying Ye, Jieping Xiong, Hui University of Science and Technology China Alibaba Cloud Computing China Thrust of Artificial Intelligence The Hong Kong University of Science and Technology Guangzhou China Department of Computer Science and Engineering The Hong Kong University of Science and Technology Hong Kong

Large Language Models (LLMs) have shown remarkable reasoning capabilities on complex tasks, but they still suffer from out-of-date knowledge, hallucinations, and opaque decision-making. In contrast, Knowledge Graphs (KGs) can provide explicit and editable knowledge for LLMs to alleviate these issues. Existing paradigm of KG-augmented LLM manually predefines the breadth of exploration space and requires flawless navigation in KGs. However, this paradigm cannot adaptively explore reasoning paths in KGs based on the question semantics and self-correct erroneous reasoning paths, resulting in a bottleneck in efficiency and effect. To address these limitations, we propose a novel self-correcting adaptive planning paradigm for KG-augmented LLM named Plan-on-Graph (PoG), which first decomposes the question into several sub-objectives and then repeats the process of adaptively exploring reasoning paths, updating memory, and reflecting on the need to self-correct erroneous reasoning paths until arriving at the answer. Specifically, three important mechanisms of Guidance, Memory, and Reflection are designed to work together, to guarantee the adaptive breadth of self-correcting planning for graph reasoning. Finally, extensive experiments on three real-world datasets demonstrate the effectiveness and efficiency of PoG. © 2024, CC BY-NC-ND.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

EdgeCross: cloud Scale Traffic Management at Peering Edges

引用

Proceedings of the ACM on Networking 2024年第CONEXT4期2卷 1-23页

作者： Xiaoliang Wang Penghui Mi Yong Zhu Baoyi An Yinhua Wang Lixiang Wang Xuezhi Yu Qiong Xie Xiang Huang Mingliang Yin Chaoyang Ji Wei Sun Yihang Lv Yuhang Chen Cam-Tu Nguyen Chen Tian Xiaoming Fu State Key Laboratory for Novel Software Technology Nanjing University Nanjing China Huawei Cloud Computing Technologies Co. Ltd. Shenzhen China Institute of Computer Science University of Göttingen Göttingen Germany

cloud providers deployed dozens of PoPs and data centers globally to serve billions of geo-distributed users. The traffic management at peering edges has become a key capability of cloud network operators to meet the diverse demands of users. With the rapid growth of cloud applications, users have recently announced new performance requirements, e.g., achieving latency as low as possible instead of maintaining a specified delay. The conventional inter-domain bandwidth allocation approach, which aims to reduce the high operating expenditures of bandwidth usage, fails to meet these new requirements. We further reveal that the flow scheduling among PoPs may fail due to the limited link capacity hidden by the cloud private backbone network controller. Therefore we seek a new traffic management at peering *** propose a new controller framework, EdgeCross, that satisfies not only users' emerging demands but maintains low operating costs. The large number of fine-grain application-aware flows and the consideration of backbone links' capacity lead to very high complexity of routing computation and verification for the controller. EdgeCross introduces a two-phase operation that first achieves the low-expense bandwidth allocation according to the standard 95th percentile billing model and then allocates specified flows to peering edges based on users' requirements. EdgeCross further reduces large memory consumption by proposing an effective routing table compression approach. The evaluation based on a production network with 16 PoPs has shown that EdgeCross can successfully process the routes of 1 billion flows in 10 seconds, reduce the average delay for performance-sensitive flows by 2 milliseconds compared to traditional BGP, and is able to save the bandwidth cost by 10-26% compared to the state-of-the-art Cascara.

关键词： cloud network flow scheduling peering edges sdn controller

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：