检索结果-内蒙古大学图书馆

Diluie: constructing diverse demonstrations of in-context learning with large language model for unified information extraction

引用

Neural Computing and Applications 2024年第22期36卷 13491-13512页

作者： Guo, Qian Guo, Yi Zhao, Jin Department of Computer Science and Engineering East China University of Science and Technology Shanghai200237 China Shanghai Key Laboratory of Data Science School of Computer Science Fudan University Shanghai200433 China

Large language models (LLMs) have demonstrated promising in-context learning capabilities, especially with instructive prompts. However, recent studies have shown that existing large models still face challenges in specific information extraction (IE) tasks. Moreover, it could have more effectively utilized various prompts such as instruction tuning, diverse demonstrations of in-context learning, and long-range token sequences for assisting language modeling in understanding context. In this study, we propose DILUIE, a unified information extraction framework based on in-context learning with diverse demonstration examples. DILUIE is encoded with an EVA attention mechanism and incremental encoding technology. Based on the constructed diverse demonstrations, we expand the size of instances efficiently in both instruction tuning and in-context learning to gain insights into the potential benefits of utilizing diverse information extraction datasets. To deepen the understanding of context, we further design three auxiliary tasks to assist in aligning contextual semantics. Experimental results demonstrate that DILUIE achieves 2.23 and 2.53% improvements in terms of Micor-/Macor-F1 on average relative to the current state-of-the-art baseline, which also significantly outperforms the GPT-3.5-turbo in zero-shot settings, and the average token length of achieving the best performance over tasks is around 15k. Furthermore, we observe that in-context learning shows enhanced performance when provided with more demonstrations during multiple-shot instruction tuning (8 k). Additionally, increasing the length of instructions (10 k) can result in a more substantial improvement in the upper limits of scaling for in-context learning. Code is available on https://***/Phevos75/DILUIE. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Demonstrations

来源：评论

学校读者我要写书评

暂无评论

Applying Kumaraswamy distribution on stick-breaking process: a Dirichlet neural topic model approach

引用

Neural Computing and Applications 2024年第22期36卷 13731-13744页

作者： Ouyang, Jihong Wang, Teng Cao, Jingyue Wang, Yiming College of Computer Science and Technology Jilin University Qianjin Street Jilin Changchun130015 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Qianjin Street Jilin Changchun130015 China

In recent years, neural topic modeling has increasingly raised extensive attention due to its capacity on generating coherent topics and flexible deep neural structures. However, the widely used Dirichlet distribution in shallow topic models is difficult to reparameterize. Therefore, most existing neural topic models assume the Gaussian as the prior of topic proportions for reparameterization. Gaussian distribution does not have the sparsity like Dirichlet distribution, which limits the model’s topic extraction ability. To address this issue, we propose a novel neural topic model approximating the Dirichlet prior with the reparameterizable Kumaraswamy distribution, namely Kumaraswamy Neural Topic Model (KNTM). Specifically, we adopted the stick-breaking process for posterior inference with the Kumaraswamy distribution as the base distribution. Besides, to capture the dependencies among topics, we propose a Kumaraswamy Recurrent Neural Topic Model (KRNTM) based on the recurrent stick-breaking construction to ensure that the model can still generate coherent topical words in high-dimensional topic space. We examined our method on five prevalent benchmark datasets over six Dirichlet-approximating neural topic models, among which KNTM has the lowest perplexity and KRNTM performance best on topic coherence and topic uniqueness. Qualitative analysis of the top topical words verifies that our proposed models can extract more semantically coherent topics compared with state-of-the-art models, further demonstrating our method’s effectiveness. This work contributes to the broader application of VAEs with Dirichlet priors. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Benchmarking

来源：评论

学校读者我要写书评

暂无评论

Dynamic Scaling of Unit Tests for Code Reward Modeling

arXiv

引用

arXiv 2025年

作者： Ma, Zeyao Zhang, Xiaokang Zhang, Jing Yu, Jifan Luo, Sijia Tang, Jie School of Information Renmin University of China China Tsinghua University China Key Laboratory of Data Engineering and Knowledge Engineering Beijing China

Current large language models (LLMs) often struggle to produce accurate solutions on the first attempt for code generation. Prior research tackles this challenge by generating multiple candidate solutions and validating them with LLM-generated unit tests. The execution results of unit tests serve as reward signals to identify correct solutions. As LLMs always confidently make mistakes, these unit tests are not reliable, thereby diminishing the quality of reward signals. Motivated by the observation that scaling the number of solutions improves LLM performance, we explore the impact of scaling unit tests to enhance reward signal quality. Our pioneer experiment reveals a positive correlation between the number of unit tests and reward signal quality, with greater benefits observed in more challenging problems. Based on these insights, we propose CodeRM-8B, a lightweight yet effective unit test generator that enables efficient and high-quality unit test scaling. Additionally, we implement a dynamic scaling mechanism that adapts the number of unit tests based on problem difficulty, further improving efficiency. Experimental results show that our approach significantly improves performance across various models on three benchmarks (e.g., with gains of 18.43% for Llama3-8B and 3.42% for GPT-4o-mini on HumanEval Plus). © 2025, CC BY-SA.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A method based on improved ant colony algorithm feature selection combined with GWO-SVR model for predicting chlorophyll-a concentration in Wuliangsu Lake

引用

Water Science and Technology 2024年第1期89卷 20-37页

作者： Wu, Chenhao Fu, Xueliang Li, Honghui Hu, Hua Li, Xue Zhang, Liqian College of Computer and Information Engineering Inner Mongolia Agricultural University Hohhot China Inner Mongolia Autonomous Region Key Laboratory of Big Data Research Application of Agriculture and Animal Husbandry Hohhot China

Chlorophyll-a (Chl-a) is an important parameter in water bodies. Due to the complexity of optics in water bodies, it is difficult to accurately predict Chl-a concentrations in water bodies by current traditional methods. In this paper, using Sentinel-2 remote sensing images as the data source combined with measured data, taking Wuliangsu Lake as the study area, a new intelligent algorithm is proposed for prediction of Chl-a concentration, which uses the adaptive ant colony exhaustive optimization algorithm (A-ACEO) for feature selection and the gray wolf optimization algorithm (GWO) to optimize support vector regression (SVR) to achieve Chl-a concentration prediction. The ant colony optimization algorithm is improved to select remote sensing feature bands for Chl-a concentration by introducing relevant optimization strategies. The GWO-SVR model is built by optimizing SVR using GWO with the selected feature bands as input and comparing it with the traditional SVR model. The results show that the usage of feature bands selected by the presented A-ACEO algorithm as inputs can effectively reduce complexity and improve the prediction performance of the model, under the condition of the same model, which can provide valuable references for monitoring the Chl-a concentration in Wuliangsu Lake. © 2024 The Authors.

关键词： Remote sensing

来源：评论

学校读者我要写书评

暂无评论

Accurate crop row recognition of maize at the seedling stage using lightweight network

引用

International Journal of Agricultural and Biological engineering 2024年第1期17卷 189-198,I0001页

作者： Jian Wei Mengfan Zhang Caicong Wu Qin Ma Weitao Wang Chuanfeng Wan College of Information and Electrical Engineering China Agricultural UniversityBeijing 100083China Key Laboratory of Agricultural Machinery Monitoring and Big Data Applications Ministry of Agriculture and Rural AffairsBeijing 100083China

Accurate extraction of crop row is very important for automation of agricultural *** rows are required for accurate machine guidance in agricultural production such as fertilization,plant protection,weeding and *** this study,an efficient crop row detection algorithm called Crop-BiSeNet V2 was proposed,which combined BiSeNet V2 with a spatial convolutional neural *** proposed Crop-BiSeNet V2 detected crop rows in color images without the use of threshold and other pre-information such as number of rows.A data set had 2697 maize crop images was constructed in challenging field trial conditions such as variable light,shadows,presence of weeds,and irregular crop *** proposed system was experimentally determined to overcome the interference of different complex *** it can be applied to crop rows of different numbers,straight lines and *** analyses were performed to check the robustness of the *** this algorithm with the Fully Convolutional Networks(FCN)algorithm,it exhibited superior performance and saved 84.85 *** accuracy rate reached 0.9811,and the detection speed reached 65.54 ms/*** Crop-BiSeNet V2 algorithm proposed in this study show strong generalization performance for seedling crop row *** provides high-reliability technical support for crop row detection research and assists in the study of intelligent field operation machinery navigation.

关键词： computer vision crop row detection precision agriculture semantic segmentation

来源：评论

学校读者我要写书评

暂无评论

Tooth number abnormality:from bench to bedside

引用

International Journal of Oral Science 2023年第1期15卷 1-15页

作者： Han Zhang Xuyan Gong Xiaoqiao Xu Xiaogang Wang Yao Sun Department of Implantology Stomatological Hospital and Dental School of Tongji UniversityShanghai Engineering Research Center of Tooth Restoration and RegenerationShanghaiChina Key Laboratory of Big Data-Based Precision Medicine School of Engineering MedicineBeihang UniversityBeijingChina

Tooth number abnormality is one of the most common dental developmental diseases,which includes both tooth agenesis and supernumerary *** development is regulated by numerous developmental signals,such as the well-known Wnt,BMP,FGF,Shh and Eda pathways,which mediate the ongoing complex interactions between epithelium and *** expression of these crutial signalling during this process may eventually lead to the development of anomalies in tooth number;however,the underlying mechanisms remain *** this review,we summarized the major process of tooth development,the latest progress of mechanism studies and newly reported clinical investigations of tooth number *** addition,potential treatment approaches for tooth number abnormality based on developmental biology are also *** review not only provides a reference for the diagnosis and treatment of tooth number abnormality in clinical practice but also facilitates the translation of basic research to the clinical application.

关键词： abnormality diagnosis epithelium

来源：评论

学校读者我要写书评

暂无评论

A quinolinyl analog of resveratrol improves neuronal damage after ischemic stroke by promoting Parkin-mediated mitophagy

引用

Chinese Journal of Natural Medicines 2025年第2期23卷 214-224页

作者： Qingqi Meng Yan Mi Libin Xu Yeshu Liu Dong Liang Yongping Wang Yan Wang Yueyang Liu Guoliang Chen Yue Hou Key Laboratory of Bioresource Research and Development of Liaoning Province College of Life and Health SciencesNational Frontiers Science Center for Industrial Intelligence and Systems OptimizationKey Laboratory of Data Analytics and Optimization for Smart IndustryMinistry of EducationNortheastern UniversityShenyang 110167China State Key Laboratory for Chemistry and Molecular Engineering of Medicinal Resources School of Chemistry and Pharmaceutical SciencesGuangxi Normal UniversityGuilin 541004China Shenyang Key Laboratory of Vascular Biology Science and Research CenterDepartment of PharmacologyShenyang Medical CollegeShenyang 110034China Key Laboratory of Structure-Based Drug Design&Discovery of Ministry of Education School of Pharmaceutical EngineeringShenyang Pharmaceutical UniversityShenyang 110016China

Ischemic stroke(IS)is a prevalent neurological disorder often resulting in significant disability or ***,extracted from Polygonum cuspidatum *** Zucc.(commonly known as Japanese knotweed),has been recognized for its potent neuroprotective ***,the neuroprotective efficacy of its derivative,(E)-4-(3,5-dimethoxystyryl)quinoline(RV02),against ischemic stroke remains inadequately *** study aimed to evaluate the protective effects of RV02 on neuronal ischemia-reperfusion injury both in vitro and in *** research utilized an animal model of middle cerebral artery occlusion/reperfusion and SH-SY5Y cells subjected to oxygen-glucose deprivation and reperfusion to simulate ischemic *** findings demonstrate that RV02 attenuates neuronal mitochondrial damage and scavenges reactive oxygen species(ROS)through mitophagy ***,Parkin knockdown was found to abolish RV02's ability to activate mitophagy and neuroprotection in *** results suggest that RV02 shows promise as a neuroprotective agent,with the activation of Parkin-mediated mitophagy potentially serving as the primary mechanism underlying its neuroprotective effects.

关键词： (E)-4-(3 5-dimethoxystyryl)quinoline Resveratrol Ischemic stroke Mitophagy Parkin

来源：评论

学校读者我要写书评

暂无评论

Low frequency error analysis and calibration for multiple star sensors system of GaoFen7 satellite

引用

地球空间信息科学学报（英文版） 2024年第1期27卷 82-94页

作者： Yanli Wang Mi Wang Ying Zhu Xiaoxiang Long The College of Geodesy and Geomatics Shandong University of Science and TechnologyQingdaoChina The State Key Laboratory of Information Engineering in Surveying Mappingand Remote SensingWuhan UniversityWuhanChina School of Electrical and Information Engineering Wuhan Institute of TechnologyWuhanChina Value-added Service Department China Center for Resources Satellite Data and ApplicationBeijingChina

The GaoFen7(GF7)optical satellite is the first Chinese civilian sub-meter high-resolution stereo mapping satellite and is equipped with a double linear array camera and laser altimeter to achieve large-scale topographic *** improve the accuracy of attitude determination,an attitude determination system comprised of four star sensors is *** to the measurement accuracy and steady performance,the star sensors 1a and 1b is usually used together for satellite attitude calculation,which is called the conventional mode of attitude ***,the combination of star sensors 2a and 2b is called the unconventional mode of attitude *** by variations in the incident angle of sunlight and solar radiation,thermal deformation occurs in the body and installation structure of the star sensor,which causes Attitude Low-Frequency Error(ALFE)and seriously influences the con-sistency of attitude determination results of different combination modes for multiple star sensors *** study proposes an ALFE analysis and calibration approach for the multiple star sensors system of GF7 satellite to ensure the consistency of attitude determination results of different combination *** on the statistical characteristics of the angles of the three axes,the installation parameters of the four star sensors are first *** analyzing the characteristics of the optical axis angles within 1420 orbit periods over 135 days,the segmented ALFE compensation model between the unconventional and conven-tional modes is proposed based on the Fourier series model and input parameter of *** on the on-orbit installation parameters and the ALFE model,the precise attitude determination results of the unconventional mode are *** results show that the attitude determination consistency after compensation is better than 2".Moreover,the reliable application time range of the compensation model is 30 days to satisfy the re

关键词： Attitude Low Frequency Error(ALFE) multiple star sensors system GaoFen7 attitude determination consistency Fourier series

来源：评论

学校读者我要写书评

暂无评论

Multi-View Graph Embedding Learning for Image Co-Segmentation and Co-Localization

引用

IEEE Transactions on Circuits and Systems for Video Technology 2024年第6期34卷 4942-4956页

作者： Huang, Aiping Li, Lijian Zhang, Le Niu, Yuzhen Zhao, Tiesong Lin, Chia-Wen Fuzhou University Fujian Key Laboratory for Intelligent Processing and Wireless Transmission of Media Information College of Physics and Information Engineering Fuzhou350108 China Fuzhou University Fujian Key Laboratory of Network Computing and Intelligent Information Processing College of Computer and Data Science Fuzhou350108 China University of Electronic Science and Technology of China School of Information and Communication Engineering Chengdu611731 China Fuzhou University Fujian Key Laboratory for Intelligent Processing and Wireless Transmission of Media Information College of Physics and Information Engineering The Fujian Science and Technology Innovation Laboratory for Optoelectronic Information Fuzhou350108 China Institute of Communications Engineering National Tsing Hua University Department of Electrical Engineering Hsinchu30013 Taiwan

Image co-segmentation and co-localization exploit inter-image information to identify and extract foreground objects with a batch mode. However, they remain challenging when confronted with large object variations or complex backgrounds. This paper proposes a multi-view graph embedding (MV-Gem) learning scheme which integrates diversity, robustness and discernibility of object features to alleviate this phenomenon. To encourage the diversity, the deep co-information containing both low-layer general representations and high-layer semantic information is generated to form a multi-view feature pool for comprehensive co-object description. To enhance the robustness, a multi-view adaptive weighted learning is formulated to fuse the deep co-information for feature complementation. To ensure the discernibility, the graph embedding and sparse constraint are embedded into the fusion formulation for feature selection. The former aims to inherit important structures from multiple views, and the latter further selects important features to restrain irrelevant backgrounds. With these techniques, MV-Gem gradually recovers all co-objects through optimization iterations. Extensive experimental results on real-world datasets demonstrate that MV-Gem is capable of locating and delineating co-objects in an image group. © 1991-2012 IEEE.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

P2 Law: Scaling Law for Post-Training After Model Pruning

arXiv

引用

arXiv 2024年

作者： Chen, Xiaodong Hu, Yuxuan Zhang, Xiaokang Wang, Yanling Li, Cuiping Chen, Hong Zhang, Jing School of Information Renmin University of China Beijing China Key Laboratory of Data Engineering and Knowledge Engineering Beijing China Zhongguancun Laboratory China

Pruning has become a widely adopted technique for reducing the hardware requirements of large language models (LLMs). To recover model performance after pruning, post-training is commonly employed to mitigate the resulting performance degradation. While post-training benefits from larger datasets, once the dataset size is already substantial, increasing the training data provides only limited performance gains. To balance post-training cost and model performance, it is necessary to explore the optimal amount of post-training data. Through extensive experiments on the Llama-3 and Qwen-2.5 series models, pruned using various common pruning methods, we uncover the scaling Law for Post-training after model Pruning, referred to as the P2 Law. This law identifies four key factors for predicting the pruned model’s post-training loss: the model size before pruning, the number of post-training tokens, the pruning rate, and the model’s loss before pruning. Moreover, P2 Law can generalize to larger dataset sizes, larger model sizes, and higher pruning rates, offering valuable insights for the post-training of pruned LLMs. © 2024, CC BY.

关键词： Digital elevation model

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

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

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：