检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Wang, Wenxiao Gu, Lihui Zhang, Liye Luo, Yunxiang Dai, Yi Shen, Chen Xie, Liang Lin, Binbin He, Xiaofei Ye, Jieping Zhejiang University China Alibaba Cloud Computing China Zhejiang University of Technology China

The rapid advancement of large language models (LLMs) has opened new possibilities for automating the proposal of innovative scientific ideas. This process involves two key phases: literature retrieval and idea generation. However, existing approaches often fall short due to their reliance on keyword-based search tools during the retrieval phase, which neglects crucial semantic information and frequently results in incomplete retrieval outcomes. Similarly, in the idea generation phase, current methodologies tend to depend solely on the internal knowledge of LLMs or metadata from retrieved papers, thereby overlooking significant valuable insights contained within the full texts. To address these limitations, we introduce SciPIP, an innovative framework designed to enhance the LLM-based proposal of scientific ideas through improvements in both literature retrieval and idea generation. Our approach begins with the construction of a comprehensive literature database that supports advanced retrieval based not only on keywords but also on semantics and citation relationships. This is complemented by the introduction of a multi-granularity retrieval algorithm aimed at ensuring more thorough and exhaustive retrieval results. For the idea generation phase, we propose a dual-path framework that effectively integrates both the content of retrieved papers and the extensive internal knowledge of LLMs. This integration significantly boosts the novelty, feasibility, and practical value of proposed ideas. Our experiments, conducted across various domains such as natural language processing and computer vision, demonstrate SciPIP’s capability to generate a multitude of innovative and useful ideas. These findings underscore SciPIP’s potential as a valuable tool for researchers seeking to advance their fields with groundbreaking concepts. Copyright © 2024, The Authors. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Multi-Server Based Task Offloading Strategy in Mobile Edge computing

Multi-Server Based Task Offloading Strategy in Mobile Edge C...

引用

2023 International Conference on Algorithms, High Performance computing, and Artificial Intelligence, AHPCAI 2023

作者： Wei, Zeju Gao, Jingnan Ruan, Peiguang School of Computer Science and Technology Hangzhou Dianzi University Hangzhou China Alibaba Cloud Big Data Application College Zhuhai College of Science and Technology Zhuhai China

ISBN: (纸本)9781510671881

Edge computing, as a complement to cloud computing, has many unique advantages. In mobile edge computing (MEC) scenarios, tasks on terminal devices can be offloaded to edge servers for computing purposes, reducing the shortage of computing resources on terminal devices. Moreover, the resulting delay is much shorter than uploading the task to the cloud center directly. However, many scenarios in reality show that there are dependencies between tasks, and a task cannot be calculated until all its predecessor tasks have been calculated. In this paper, we summarize a variety of task dependencies. At the same time, we build an efficient task offloading model based on two levels of edge servers with different functions. In particular, the second-level edge server has the routing decision function, which provides us with diversified offloading methods. Finally, we used improved NSGA-Ⅱ for task offloading. Compared with other offloading methods, our algorithm performs better in latency and energy consumption. © 2023 SPIE. All rights reserved.

关键词： Energy utilization

来源：评论

学校读者我要写书评

暂无评论

Klotski v2: Improved DNN Model Orchestration Framework for Dataflow Architecture Accelerators

引用

IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems 2025年第3期44卷 1045-1058页

作者： Bai, Chen Wei, Xuechao Zhuo, Youwei Cai, Yi Zheng, Hongzhong Yu, Bei Xie, Yuan The Chinese University of Hong Kong Department of Computer Science and Engineering Hong Kong Hong Kong Alibaba DAMO Academy Computing Technology Laboratory Hangzhou311121 China Peking University School of Integrated Circuits Beijing100871 China The Hong Kong University of Science and Technology Department of Electronic and Computer Engineering Hong Kong Hong Kong

Dataflow architecture accelerators are a new kind of scalable DNN accelerators. For an instruction, the availability of input operands solely determines the beginning of executions. DNN model orchestration determines how to partition, schedule, and map the computation to the underlying hardware. In this article, we propose the Klotski v2 framework to solve DNN model orchestration for dataflow architecture accelerators. First, a Bayesian optimization-based entropy-directed partition algorithm is proposed to transform a DNN model into μ ops. Second, a unified formal formulation for μ ops scheduling and mapping is presented. Third, a two-stage methodology is proposed to decouple the scheduling and mapping. Fourth, a Hilbert curve-based mapping heuristic is proposed to enhance problem-solving efficiency, improving the tradeoff between solution quality and algorithm runtime. Extensive results show that Klotski v2 can achieve an average of 21.57% higher execution performance improvement than the previous methodologies. With the Hilbert curve-based mapping heuristic, we improve the algorithm efficiency by an average of 63.50% across different DNN workloads. © 1982-2012 IEEE.

关键词： Scheduling algorithms

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 24

Plan-on-graph: self-correcting adaptive planning of large la...

引用

Proceedings of the 38th International Conference on Neural Information Processing Systems

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

ISBN: (纸本)9798331314385

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.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A reposition algorithm for e-hailing based on quantum annealing and intuitive reasoning

引用

Intelligent and Converged Networks 2024年第4期5卷 317-335页

作者： Chao Wang Yiyun Shi Sumin Wang Key Laboratory of Specialty Fiber Optics and Optical Access Networks Shanghai UniversityShanghai 200444China School of Information Engineering Ganzhou Key Laboratory of Cloud Computing and Big Data ResearchGannan University of Science and TechnologyGanzhou 341000China

Currently,the challenge lies in the traditional intelligent algorithm’s ability to effectively address the e-hailing repositioning *** identifying the underlying characteristics in extensive traffic data within a limited timeframe is difficult,ultimately preventing the achievement of the most optimal *** paper suggests a hybrid computing architecture involving reinforcement learning and quantum annealing based on intuitive *** reasoning aims to enhance performance in scenarios with poor system robustness,complex tasks,and diverse goals.A deep learning model is constructed,trained to extract scene features,and combined with expert knowledge,then transformed into a quantum annealable *** final strategy is obtained using a D-wave quantum computer with quantum tunneling effect,which helps in finding optimal solutions by jumping out of local suboptimal *** on 400000 real data,four algorithms are compared:minimum-cost flow,sequential markov decision process,hot-dot strategy,and driver-prefer *** average total revenue increases by about 10%and vehicle utilization by about 15%in various *** summary,the proposed architecture effectively solves the e-hailing reposition problem,offering new directions for robust artificial intelligence in big data decision problems.

关键词： intuitive reasoning reinforcement learning quantum annealing

来源：评论

学校读者我要写书评

暂无评论

Delving into the Reversal Curse: How Far Can Large Language Models Generalize? 38

Delving into the Reversal Curse: How Far Can Large Language ...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Lin, Zhengkai Fu, Zhihang Liu, Kai Xie, Liang Lin, Binbin Wang, Wenxiao Cai, Deng Wu, Yue Ye, Jieping State Key Lab of CAD&CG Zhejiang University China Alibaba Cloud China College of Biomedical Engineering & Instrument Science Zhejiang University China College of Computer Science and Technology Zhejiang University of Technology China School of Software Technology Zhejiang University China Fullong Inc. China

While large language models (LLMs) showcase unprecedented capabilities, they also exhibit certain inherent limitations when facing seemingly trivial tasks. A prime example is the recently debated "reversal curse", which surfaces when models, having been trained on the fact "A is B", struggle to generalize this knowledge to infer that "B is A". In this paper, we examine the manifestation of the reversal curse across various tasks and delve into both the generalization abilities and the problem-solving mechanisms of LLMs. This investigation leads to a series of significant insights: (1) LLMs are able to generalize to "B is A" when both A and B are presented in the context as in the case of a multiple-choice question. (2) This generalization ability is highly correlated to the structure of the fact "A is B" in the training documents. For example, this generalization only applies to biographies structured in "[Name] is [Description]" but not to "[Description] is [Name]". (3) We propose and verify the hypothesis that LLMs possess an inherent bias in fact recalling during knowledge application, which explains and underscores the importance of the document structure to successful learning. (4) The negative impact of this bias on the downstream performance of LLMs can hardly be mitigated through training alone. These findings offer a novel perspective on interpreting LLMs' generalization through their intrinsic mechanisms and provide insights for developing more effective learning methods. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Plausible Heterogeneous Graph k-Anonymization for Social Networks

引用

Tsinghua science and technology 2022年第6期27卷 912-924页

作者： Kaiyang Li Ling Tian Xu Zheng Bei Hui School of Computer Science and Engineering University of Electronic Science and Technology of ChinaChengdu 611731China School of Information and Software Engineering University of Electronic Science and Technology of ChinaChengdu 610054China Trusted Cloud Computing and Big Data Key Laboratory of Sichuan Province Chengdu 610000China

The inefficient utilization of ubiquitous graph data with combinatorial structures necessitates graph embedding methods,aiming at learning a continuous vector space for the graph which is amenable to be adopted in traditional machine learning algorithms in favor of vector *** embedding methods build an important bridge between social network analysis and data analytics as social networks naturally generate an unprecedented volume of graph data *** social network data not only bring benefit for public health,disaster response,commercial promotion,and many other applications,but also give birth to threats that jeopardize each individual’s privacy and ***,most existing works in publishing social graph embedding data only focus on preserving social graph structure with less attention paid to the privacy issues inherited from social *** be specific,attackers can infer the presence of a sensitive relationship between two individuals by training a predictive model with the exposed social network *** this paper,we propose a novel link-privacy preserved graph embedding framework using adversarial learning,which can reduce adversary’s prediction accuracy on sensitive links while persevering sufficient non-sensitive information such as graph topology and node attributes in graph *** experiments are conducted to evaluate the proposed framework using ground truth social network datasets.

关键词： social network graph embedding privacy preservation adversarial learning

来源：评论

学校读者我要写书评

暂无评论

VideoCLIP-XL: Advancing Long Description Understanding for Video CLIP Models

arXiv

引用

arXiv 2024年

作者： Wang, Jiapeng Wang, Chengyu Huang, Kunzhe Huang, Jun Jin, Lianwen South China University of Technology Guangzhou China Alibaba Cloud Computing Hangzhou China

Contrastive Language-Image Pre-training (CLIP) has been widely studied and applied in numerous applications. However, the emphasis on brief summary texts during pre-training prevents CLIP from understanding long descriptions. This issue is particularly acute regarding videos given that videos often contain abundant detailed contents. In this paper, we propose the VideoCLIP-XL (eXtra Length) model, which aims to unleash the long-description understanding capability of video CLIP models. Firstly, we establish an automatic data collection system and gather a large-scale VILD pre-training dataset1 with VIdeo and Long-Description pairs. Then, we propose Text-similarity-guided Primary Component Matching (TPCM) to better learn the distribution of feature space while expanding the long description capability. We also introduce two new tasks namely Detail-aware Description Ranking (DDR) and Hallucination-aware Description Ranking (HDR) for further understanding improvement. Finally, we construct a Long Video Description Ranking (LVDR) benchmark2 for evaluating the long-description capability more comprehensively. Extensive experimental results on widely-used text-video retrieval benchmarks with both short and long descriptions and our LVDR benchmark can fully demonstrate the effectiveness of our method.3 Copyright © 2024, The Authors. All rights reserved.

关键词： Footage counters

来源：评论

学校读者我要写书评

暂无评论

Wordle Data Exploration and Prediction Based on BP-GA Model 4

Wordle Data Exploration and Prediction Based on BP-GA Model

引用

4th International Conference on Computers, Information Processing and Advanced Education, CIPAE 2023

作者： Zhou, Luyao Ruan, Peiguang Wenzhou University of Technology School of Data Science and Artificial Intelligence Wenzhou China Zhuhai College of Science and Technology Alibaba Cloud Big Data Application College Zhuhai China

ISBN: (纸本)9798350342710

Wordle, a fun puzzle game, the acquisition of Wordle by the New York Times led to a surge in the game's popularity. In this article, we aim to explore various data related to Wordle, including the number of daily reported results, words of varying difficulty, and the distribution of user attempts. We use the ARIMA model to predict the number of reported results in the world in the short term, and verify it with the LSTM model, and the results show that our interval prediction is correct. We forecast to reach 14,737 within the forecast range (13,667, 15,807) in the coming months. In addition, we built a BP-GA model to predict the distribution of future reported outcomes and found a number of word attributes, including seven numerical factors and two potential uncertainties that could affect reported outcomes. Based on the random forest model and grid search, the difficulty classification model of words is constructed, and it is found that word frequency and ASCII mean have great influence on the difficulty classification of words. The classification accuracy is 88.89%, which verifies the superiority of the model. © 2023 IEEE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Three-Stage Progressive Pre-Analysis Framework for VMAF Controllable Image Coding

Three-Stage Progressive Pre-Analysis Framework for VMAF Cont...

引用

Data Compression Conference (DCC)

作者： Shaokang Wang Guoqing Xiang Wenzhao Li Mingyuan Yang Fan Yang Shanghang Zhang Xiaodong Xie Huizhu Jia School of Computer Science Peking University Beijing China Alibaba Cloud Computing Beijing China

ISBN: (数字)9798331534714

ISBN: (纸本)9798331534721

To achieve controllable subjective quality in image coding, this paper proposes a Video Multi-method Assessment Fusion (VMAF)-oriented image coding pre-analysis algorithm, enabling the adaptive derivation of quantization parameters corresponding to a specified quality target. First, a $Q-\mathcal{V}$ model is constructed to describe the relationship between encoding quantization and VMAF distortion. Then, a Three-stage Progressive Control (TPC) algorithm, shown in Fig. 1(a), is designed to adapt quantization parameters using the discovered $Q-\mathcal{V}$ model. The first two stages, based on lightweight feature extraction, iteratively fit the distortion metrics intrinsically calculated by VMAF to predict the VMAF value for a given sample under specified distortion conditions. The final stage fits the $Q-\mathcal{V}$ model parameters using multi-point VMAF distortion data and outputs the corresponding quantization step for encoder control. A two-pass refinement algorithm, depicted in Fig. 1(b), further adjusts the quantization parameters based on the first encoding pass, improving quality control accuracy and framework robustness. Experiments on four datasets show that the quality control error remains below 1.293% for various VMAF targets, and the two-pass refinement reduces it further to 0.710%, outperforming existing methods.

关键词： Measurement Adaptation models Quantization (signal) Image coding Quality control Distortion Prediction algorithms Feature extraction Encoding Robustness

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：