检索结果-内蒙古大学图书馆

Hierarchical Contrastive Learning for Semantic Segmentation

IEEE Transactions on Neural Networks and Learning Systems 2024年第6期36卷 11202-11214页

作者： Jie Jiang Xingjian He Weining Wang Hanqing Lu Jing Liu School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China National Laboratory of Pattern Recognition Institute of Automation Chinese Academy of Sciences Beijing China

Recently, pixel-to-pixel contrastive learning in single-scale feature space has been widely studied in semantic segmentation to learn a unified feature expression for pixels of the same category. However, the unified representation is too extreme, and the receptive field of each single-scale pixel is limited, which is insufficient to reflect the representative features of the category. To address these problems, this article extends the single-scale feature space to that of multiscale and proposes a hierarchical contrastive learning (Hi-CL) method to explore pixel-to-component semantic relationships. First, we generate multiscale candidate samples by applying several pooling windows with different sizes on a feature map, where different windows may represent different parts of the objects in the image. Then, we prune the sample set through threshold-based criteria to select appropriate samples for feature representation learning. Finally, Hi-CL is performed to learn the pixel-to-component consistency with the pruned samples. Our method is easy to be applied on existing semantic segmentation models and obtains consistent improvement. Furthermore, we achieve state-of-the-art results on three popular benchmarks, including Cityscapes, ADE20K, and COCO Stuff datasets.

关键词： Semantic segmentation Contrastive learning Semantics Transformers Feature extraction Computer vision Visualization Representation learning Neural networks Convolution

来源：评论

学校读者我要写书评

暂无评论

TOWARDS CODABLE WATERMARKING FOR INJECTING MULTI-BITS INFORMATION TO LLMS 12

TOWARDS CODABLE WATERMARKING FOR INJECTING MULTI-BITS INFORM...

引用

12th International Conference on Learning Representations, ICLR 2024

作者： Wang, Lean Yang, Wenkai Chen, Deli Zhou, Hao Lin, Yankai Meng, Fandong Zhou, Jie Sun, Xu National Key Laboratory for Multimedia Information Processing School of Computer Science Peking University China Gaoling School of Artificial Intelligence Renmin University of China China Pattern Recognition Center WeChat AI Tencent Inc. China DeepSeek AI China

As large language models (LLMs) generate texts with increasing fluency and realism, there is a growing need to identify the source of texts to prevent the abuse of LLMs. Text watermarking techniques have proven reliable in distinguishing whether a text is generated by LLMs by injecting hidden patterns. However, we argue that existing LLM watermarking methods are encoding-inefficient and cannot flexibly meet the diverse information encoding needs (such as encoding model version, generation time, user id, etc.). In this work, we conduct the first systematic study on the topic of Codable Text Watermarking for LLMs (CTWL) that allows text watermarks to carry multi-bit customizable information. First of all, we study the taxonomy of LLM watermarking technologies and give a mathematical formulation for CTWL. Additionally, we provide a comprehensive evaluation system for CTWL: (1) watermarking success rate, (2) robustness against various corruptions, (3) coding rate of payload information, (4) encoding and decoding efficiency, (5) impacts on the quality of the generated text. To meet the requirements of these non-Pareto-improving metrics, we follow the most prominent vocabulary partition-based watermarking direction, and devise an advanced CTWL method named Balance-Marking. The core idea of our method is to use a proxy language model to split the vocabulary into probability-balanced parts, thereby effectively maintaining the quality of the watermarked text. Extensive experimental results show that our method outperforms the baseline under comprehensive evaluation. Our code is available at https://***/lancopku/codable-watermarking-for-llm. © 2024 12th International Conference on Learning Representations, ICLR 2024. All rights reserved.

关键词： Encoding (symbols)

来源：评论

学校读者我要写书评

暂无评论

3D PARTICLE PICKING IN CRYO-ELECTRON TOMOGRAMS USING INSTANCE SEGMENTATION 29

3D PARTICLE PICKING IN CRYO-ELECTRON TOMOGRAMS USING INSTANC...

引用

29th IEEE International Conference on Image Processing, ICIP 2022

作者： Liu, Guole Luo, Yaoru Yang, Ge School of Artificial Intelligence University of Chinese Academy of Sciences China National Key Laboratory of Pattern Recognition Institute of Automation Chinese Academy of Sciences China

ISBN: (数字)9781665496209

ISBN: (纸本)9781665496209

To identify and localize macromolecules of interest in crowded intracellular environment, the low signal-to-noise ratio and missing imaging wedge of cryo-electron tomography (cryo-ET) data pose substantial technical challenges. Currently, mainstream approaches of 3D particle picking in cryo-ET either follow the 'segment-then-cluster' strategy, or extract potential structural regions as sub-tomograms and then perform classification. Different from these two-step methods, we solve the problem using a one-step instance segmentation approach, termed 3D-SOLOv2. Specifically, the category and mask of each 3D particle are predicted according to the particle's location and size. To solve the lack of real masks for 3D particles in cryo-ET, a Gaussian-shaped mask is proposed to approximate real masks. When tested on simulated datasets of SHREC2020 challenge, our model achieves the fastest inference speed and the state-of-the-art performance for both localization and classification tasks. When tested on real cryo-ET dataset of EMPIAR-10045, our model also achieves better performance than other methods. © 2022 IEEE.

关键词： Signal to noise ratio

来源：评论

学校读者我要写书评

暂无评论

Sheet Music Transformer: End-To-End Optical Music recognition Beyond Monophonic Transcription

arXiv

引用

arXiv 2024年

作者： Ríos-Vila, Antonio Calvo-Zaragoza, Jorge Paquet, Thierry Pattern Recognition and Artificial Intelligence Group University of Alicante Spain LITIS Laboratory - EA 4108 Rouen University France

State-of-the-art end-to-end Optical Music recognition (OMR) has, to date, primarily been carried out using monophonic transcription techniques to handle complex score layouts, such as polyphony, often by resorting to simplifications or specific adaptations. Despite their efficacy, these approaches imply challenges related to scalability and limitations. This paper presents the Sheet Music Transformer (SMT), the first end-to-end OMR model designed to transcribe complex musical scores without relying solely on monophonic strategies. Our model employs a Transformer-based image-to-sequence framework that predicts score transcriptions in a standard digital music encoding format from input images. Our model has been tested on two polyphonic music datasets and has proven capable of handling these intricate music structures effectively. The experimental outcomes not only indicate the competence of the model, but also show that it is better than the state-of-the-art methods, thus contributing to advancements in end-to-end OMR transcription. © 2024, CC BY.

关键词： Music

来源：评论

学校读者我要写书评

暂无评论

VLKEB: a large vision-language model knowledge editing benchmark 24

VLKEB: a large vision-language model knowledge editing bench...

引用

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

作者： Han Huang Haitian Zhong Tao Yu Qiang Liu Shu Wu Liang Wang Tieniu Tan University of Chinese Academy of Sciences (UCAS) and New Laboratory of Pattern Recognition (NLPR) State Key Laboratory of Multimodal Artificial Intelligence Systems (MAIS) Institute of Automation Chinese Academy of Sciences (CASIA) New Laboratory of Pattern Recognition (NLPR) State Key Laboratory of Multimodal Artificial Intelligence Systems (MAIS) Institute of Automation Chinese Academy of Sciences (CASIA) New Laboratory of Pattern Recognition (NLPR) State Key Laboratory of Multimodal Artificial Intelligence Systems (MAIS) Institute of Automation Chinese Academy of Sciences (CASIA) and Nanjing University

ISBN: (纸本)9798331314385

Recently, knowledge editing on large language models (LLMs) has received considerable attention. Compared to this, editing Large Vision-Language Models (LVLMs) faces extra challenges from diverse data modalities and complicated model components, and data for LVLMs editing are limited. The existing LVLM editing benchmark, which comprises three metrics (Reliability, Locality, and Generality), falls short in the quality of synthesized evaluation images and cannot assess whether models apply edited knowledge in relevant content. Therefore, we employ more reliable data collection methods to construct a new Large Vision-Language Model Knowledge Editing Benchmark, VLKEB, and extend the Portability metric for more comprehensive evaluation. Leveraging a multi-modal knowledge graph, our image data are bound with knowledge entities. This can be further used to extract entity-related knowledge, which constitutes the base of editing data. We conduct experiments of different editing methods on five LVLMs, and thoroughly analyze how do they impact the models. The results reveal strengths and deficiencies of these methods and hopefully provide insights for future research. The codes and dataset are available at: https://***/VLKEB/VLKEB.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Parallel High-Utility Itemset Mining Algorithm Based on Hadoop

引用

Complex System Modeling and Simulation 2023年第1期3卷 47-58页

作者： Zaihe Cheng Wei Shen Wei Fang Jerry Chun-Wei Lin School of Internet of Things Wuxi Institute of TechnologyWuxi 214121China School of Artificial Intelligence and Computer Science Jiangnan UniversityWuxi 214122China Jiangsu Provincial Engineering Laboratory of Pattern Recognition and Computational Intelligence Jiangnan UniversityWuxi 214122China Department of Computer Science Electrical Engineering and Mathematical SciencesWestern Norway University of Applied SciencesBergen 5020Norway

High-utility itemset mining(HUIM)can consider not only the profit factor but also the profitable factor,which is an essential task in data ***,most HUIM algorithms are mainly developed on a single machine,which is inefficient for big data since limited memory and processing capacities are available.A parallel efficient high-utility itemset mining(P-EFIM)algorithm is proposed based on the Hadoop platform to solve this problem in this *** P-EFIM,the transaction-weighted utilization values are calculated and ordered for the itemsets with the MapReduce *** the ordered itemsets are renumbered,and the low-utility itemsets are pruned to improve the dataset *** the Map phase,the P-EFIM algorithm divides the task into multiple independent *** uses the proposed S-style distribution strategy to distribute the subtasks evenly across all nodes to ensure ***,the P-EFIM uses the EFIM algorithm to mine each subtask dataset to enhance the performance in the Reduce *** are performed on eight datasets,and the results show that the runtime performance of P-EFIM is significantly higher than that of the PHUI-Growth,which is also HUIM algorithm based on the Hadoop framework.

关键词： pattern mining data mining Hadoop parallel high-utility itemset mining big data

来源：评论

学校读者我要写书评

暂无评论

Sustainable Mining in the Era of artificial intelligence

引用

IEEE/CAA Journal of Automatica Sinica 2024年第1期11卷 1-4页

作者： Long Chen Yuting Xie Yutong Wang Shirong Ge Fei-Yue Wang IEEE the State Key Laboratory for Management and Control of Complex Systems at the Institute of Automation Chinese Academy of SciencesBeijing 100190China the National Laboratory of Pattern Recognition at the Institute of Automation and Waytous Ltd.China the School of Computer Science and Engineering Sun Yat-Sen UniversityGuangzhou 510275GuangdongChina the School of Mechanical Electronic and Information Engineering China University of Mining and Technology BeijingBejing 100083China the School of Artificial Intelligence University of Chinese Academy of SciencesBeijing 100190China

The mining sector historically drove the global economy but at the expense of severe environmental and health repercussions,posing sustainability challenges[1]-[3].Recent advancements on artificial intelligence(AI)are revolutionizing mining through robotic and data-driven innovations[4]-[7].While AI offers mining industry advantages,it is crucial to acknowledge the potential risks associated with its widespread ***-reliance on AI may lead to a loss of human control over mining operations in the future,resulting in unpredictable consequences.

关键词： Sustainable mining consequences

来源：评论

学校读者我要写书评

暂无评论

Accurate Extrinsic Calibration of LiDAR and Camera with Refined Vertex Features 12

Accurate Extrinsic Calibration of LiDAR and Camera with Refi...

引用

12th International Conference on Indoor Positioning and Indoor Navigation - Work-in-Progress Papers, IPIN-WiP 2022

作者： Wang, Shuo Rong, Zheng Zhang, Pengju Wu, Yihong School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China National Laboratory of Pattern Recognition Institute of Automation Chinese Academy of Sciences Beijing China

LiDARs and cameras are widely used in many research fields, due to the complementarity of their data. Calibrating the extrinsic parameters of LiDAR frame and camera frame is essential for fusing the two kinds of data. Calibration methods based on a calibration board extract geometry features from point clouds and images, then build geometry constraints to estimate the extrinsic parameters. In this paper, we exhaustively analyze the measurement characteristics of LiDARs that would introduce notable negative effect on the calibration, including the noise of depth measurement and the divergence of laser beams. Therefore, we propose a refining method for vertex features from LiDAR point clouds using the prior information of the board, which can effectively mitigate the effect of systematic measurement errors and improve the accuracy of the calibration results. Meanwhile, our calibration method reduces the minimal number of calibration datasets to one, which promotes the efficiency of calibration processes. Besides, we propose an objective and independent evaluation method for target-based calibration methods. Extensive experiments and comparisons with state-of-the-art methods show that using refined vertex features can notably improve the accuracy and efficiency of extrinsic parameter calibration. © 2022 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).

关键词： Calibration

来源：评论

学校读者我要写书评

暂无评论

Find Parent then Label Children: A Two-stage Taxonomy Completion Method with Pre-trained Language Model

arXiv

引用

arXiv 2024年

作者： Xia, Fei Weng, Yixuan He, Shizhu Liu, Kang Zhao, Jun National Laboratory of Pattern Recognition Institute of Automation CAS China School of Artificial Intelligence University of Chinese Academy of Sciences China

Taxonomies, which organize domain concepts into hierarchical structures, are crucial for building knowledge systems and downstream applications. As domain knowledge evolves, taxonomies need to be continuously updated to include new concepts. Previous approaches have mainly focused on adding concepts to the leaf nodes of the existing hierarchical tree, which does not fully utilize the taxonomy’s knowledge and is unable to update the original taxonomy structure (usually involving non-leaf nodes). In this paper, we propose a two-stage method called ATTEMPT for taxonomy completion. Our method inserts new concepts into the correct position by finding a parent node and labeling child nodes. Specifically, by combining local nodes with prompts to generate natural sentences, we take advantage of pre-trained language models for hypernym/hyponymy recognition. Experimental results on two public datasets (including six domains) show that ATTEMPT performs best on both taxonomy completion and extension tasks, surpassing existing methods. © 2024, CC BY.

关键词： Taxonomies

来源：评论

学校读者我要写书评

暂无评论

A Detection System and Device for Main Variable Temperature Controller Based on Cloud Platform 8

A Detection System and Device for Main Variable Temperature ...

引用

8th International Conference on Electronic Technology and Information Science, ICETIS 2023

作者： Li, Guiliang Yan, Jinhui Luo, Xinyuan Luo, Yongsheng Ren, Yousong Zhou, Chengjiang Dehong Power Supply Bureau Yunnan Power Grid Corporation Dehong678400 China School of Information Science and Technology Yunnan Normal University Kunming650500 China Laboratory of Pattern Recognition and Artificial Intelligence Kunming650500 China

ISBN: (纸本)9781510666535

The main variable temperature controller has not achieved an automatic testing function so far, which consumes great labor and capital costs. The innovation of this paper is to achieve fully automatic and intelligent detection of the main variable temperature controller. We construct an intelligent testing equipment remote control system, including a remote monitoring and management system for testing equipment backstage, a mobile APP for testing equipment, an APP for receiving terminals, and an application service for the testing report interface, which realizes the function of remote management and viewing of reports. A fully automatic calibration device for the main substation thermostat is designed, including the inspected thermostat, constant temperature oil tank, calibrator, and graphic conversion device, which realizes the device to automatically complete the testing of substation-related devices. By designing the communication serial port of the device and developing the corresponding device on the cloud, the software and hardware are well integrated. The technical solution promotes the turnover rate and utilization rate of testing equipment and enhances the comprehensive management capability of testing equipment. © 2023 SPIE.

关键词： Controllers

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：