检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Yu, Haiyang Yi, Siyang Niu, Ke Zhuo, Minghan Li, Bin Shanghai Key Laboratory of Intelligent Information Processing School of Computer Science Fudan University China

With the rapid advancement of deep learning, particularly in the field of medical image analysis, an increasing number of Vision-Language Models (VLMs) are being widely applied to solve complex health and biomedical challenges. However, existing research has primarily focused on specific tasks or single modalities, which limits their applicability and generalization across diverse medical scenarios. To address this challenge, we propose UMIT, a unified multi-modal, multi-task VLM designed specifically for medical imaging tasks. UMIT is able to solve various tasks, including visual question answering, disease detection, and medical report generation. In addition, it is applicable to multiple imaging modalities (e.g., X-ray, CT and PET), covering a wide range of applications from basic diagnostics to complex lesion analysis. Moreover, UMIT supports both English and Chinese, expanding its applicability globally and ensuring accessibility to healthcare services in different linguistic contexts. To enhance the model’s adaptability and task-handling capability, we design a unique two-stage training strategy and fine-tune UMIT with designed instruction templates. Through extensive empirical evaluation, UMIT outperforms previous methods in five tasks across multiple datasets. The performance of UMIT indicates that it can significantly enhance diagnostic accuracy and workflow efficiency, thus providing effective solutions for medical imaging applications. The code and adopted datasets are available at https://***/dzosamu/UMIT/tree/main. © 2025, CC0.

关键词： Visual languages

来源：评论

学校读者我要写书评

暂无评论

Boosting semi-supervised federated learning by effectively exploiting server-side knowledge and client-side unconfident samples

引用

Neural Networks 2025年 188卷 107440页

作者： Liu, Hongquan Mi, Yuxi Tang, Yateng Guan, Jihong Zhou, Shuigeng Shanghai Key Lab of Intelligent Information Processing and School of Computer Science Fudan University Shanghai China School of Computer Science and Technology Tongji University Shanghai China Wechat Tencent Inc. Shenzhen China

Semi-supervised federated learning (SSFL) has emerged as a promising paradigm to reduce the need for fully labeled data in training federated learning (FL) models. This paper focuses on the label-at-server scenario, where clients’ data are entirely unlabeled and the server possesses only a limited amount of labeled data. In this setting, the non-independent and identically distributed (non-IID) local data and the incorrect pseudo-labels will possibly introduce bias into the model during local training. Prior works try to alleviate the bias by fine-tuning the global model with clean labeled data, ignoring explicitly leveraging server-side knowledge to guide local training. Additionally, existing methods typically discard samples with unconfident pseudo-labels, resulting in many samples being not used, consequently suboptimal performance and slow convergence. This paper introduces a novel method to enhance SSFL performance by effectively exploiting server-side clean knowledge and client-side unconfident samples. Specifically, we propose a representation alignment module that mitigates the influence of non-IID data by aligning local features with the class proxies of the server labeled data. Furthermore, we employ a shrink loss to reduce the risk associated with unreliable pseudo-labels, ensuring the exploitation of valuable information contained in the entire unlabeled dataset. Extensive experiments on five benchmark datasets under various settings demonstrate the effectiveness and generality of the proposed method, which not only outperforms existing methods but also reduces the communication cost required to achieve the target performance. © 2025 Elsevier Ltd

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

PHT-CAD: Efficient CAD Parametric Primitive Analysis with Progressive Hierarchical Tuning

arXiv

引用

arXiv 2025年

作者： Niu, Ke Chen, Yuwen Yu, Haiyang Chen, Zhuofan Que, Xianghui Li, Bin Xue, Xiangyang Shanghai Key Laboratory of Intelligent Information Processing School of Computer Science Fudan University China

computer-Aided Design (CAD) plays a pivotal role in industrial manufacturing, yet 2D Parametric Primitive Analysis (PPA) remains underexplored due to two key challenges: structural constraint reasoning and advanced semantic understanding. To tackle these challenges, we first propose an Efficient Hybrid Parametrization (EHP) for better representing 2D engineering drawings. EHP contains four types of atomic component (i.e., point, line, circle, and arc). Additionally, we propose PHT-CAD, a novel 2D PPA framework that harnesses the modality alignment and reasoning capabilities of Vision-Language Models (VLMs) for precise engineering drawing analysis. In PHT-CAD, we introduce four dedicated regression heads to predict corresponding atomic components. To train PHT-CAD, a three-stage training paradigm Progressive Hierarchical Tuning (PHT) is proposed to progressively enhance PHT-CAD’s capability to perceive individual primitives, infer structural constraints, and align annotation layers with their corresponding geometric representations. Considering that existing datasets lack complete annotation layers and real-world engineering drawings, we introduce ParaCAD, the first large-scale benchmark that explicitly integrates both the geometric and annotation layers. ParaCAD comprises over 10 million annotated drawings for training and 3,000 real-world industrial drawings with complex topological structures and physical constraints for test. Extensive experiments demonstrate the effectiveness of PHT-CAD and highlight the practical significance of ParaCAD in advancing 2D PPA research. The code and adopted datasets are available at https://***/yuwen-chen616/ PHT-CAD. Copyright © 2025, 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

Learning Visual-Semantic Hierarchical Attribute Space for Interpretable Open-Set Recognition

Learning Visual-Semantic Hierarchical Attribute Space for In...

引用

2025 IEEE/CVF Winter Conference on Applications of computer Vision, WACV 2025

作者： Xu, Zhuo Xiang, Xiang School of Artificial Intelligence and Automation Huazhong University of Science and Technology National Key Lab of Multi-Spectral Information Intelligent Processing Technology Wuhan China Peng Cheng Lab China

ISBN: (纸本)9798331510831

In the field of open-set recognition, conventional models often focus on addressing challenges within a single hierarchical category, and these methods frequently lack inter-pretability. In this paper, we propose a novel solution that utilizes attributes and hierarchical relationships to achieve interpretable open-set recognition. Our method is centered around the visual-semantic attribute space. By leveraging hierarchy division, we can decompose the attributes into more granular components, thereby yielding additional performance improvements. When confronted with an unfamiliar object, our method not only classifies it as an unknown category but also provides insights into the broader category and its associated attributes. This capability enhances interpretability by offering valuable information regarding the potential category and characteristics of the object. Experimental results demonstrate great performance improvements compared to existing methods. © 2025 IEEE.

关键词： Hierarchical systems

来源：评论

学校读者我要写书评

暂无评论

RoleScan: Enhancing Social Bot Detection Using Social Role Vector 19th

RoleScan: Enhancing Social Bot Detection Using Social Role V...

引用

19th CCF Conference on computer Supported Cooperative Work and Social Computing, ChineseCSCW 2024

作者： Wen, Wen Gao, Min Gong, Qingyuan Wang, Xin Chen, Yang Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University Shanghai China Research Institute of Intelligent Complex Systems Fudan University Shanghai China

ISBN: (纸本)9789819623723

The increase in social media users has led to the rise of social bots, which can disrupt online environments and threaten user privacy. This study presents RoleScan, a novel framework that integrates Social Role Vector (SRV) and Graph Attention Network v2 (GATv2) to detect social bots more accurately. We are the first to formulate the social bot detection problem by considering users’ properties and social influence. RoleScan extracts user features and text features from user profiles and tweets, and uses SRV to capture the social features. SRV is based on node centrality measurement, structural hole theory, and social capital theory, offering a comprehensive view of the influence of a user within the network. By leveraging an attention mechanism, RoleScan focuses on pivotal features and connections, enhancing its ability to differentiate between bots and humans. The experimental results show that RoleScan outperforms existing models, achieving an AUC of 0.887. An ablation study further highlights the critical role of incorporating social features into the detection process. Our study not only enhances the social bot detection performance, but also offers new insights into the users’ diverse roles within social networks. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Bot (Internet)

来源：评论

学校读者我要写书评

暂无评论

CLIMS++: Cross Language Image Matching with Automatic Context Discovery for Weakly Supervised Semantic Segmentation

引用

International Journal of computer Vision 2025年 1-20页

作者： Xie, Jinheng Deng, Songhe Hou, Xianxu Luo, Zhaochuan Shen, Linlin Huang, Yawen Zheng, Yefeng Shou, Mike Zheng Computer Vision Institute School of Computer Science and Software Engineering Shenzhen University Shenzhen518060 China Department of Computer Science Wenzhou-Kean University Wenzhou325060 China Guangdong Provincial Key Laboratory of Intelligent Information Processing Shenzhen University Shenzhen China Show Lab National University of Singapore Singapore639798 Singapore School of AI and Advanced Computing Xi’an Jiaotong-Liverpool University Suzhou215400 China Jarvis Research Center Tencent YouTu Lab Shenzhen518703 China

While promising results have been achieved in weakly-supervised semantic segmentation (WSSS), limited supervision from image-level tags inevitably induces discriminative reliance and spurious relations between target classes and background regions. Thus, Class Activation Map (CAM) usually tends to activate discriminative object regions and falsely includes lots of class-related backgrounds. Without pixel-level supervisions, it could be very difficult to enlarge the foreground activation and suppress those false activation of background regions. In this paper, we propose a novel framework of Cross Language Image Matching with Automatic Context Discovery (CLIMS++), based on the recently introduced Contrastive Language-Image Pre-training (CLIP) model, for WSSS. The core idea of our framework is to introduce natural language supervision to activate more complete object regions and suppress class-related background regions in CAM. In particular, we design object, background region, and text label matching losses to guide the model to excite more reasonable object regions of each category. In addition, we propose to automatically find spurious relations between foreground categories and backgrounds, through which a background suppression loss is designed to suppress the activation of class-related backgrounds. The above designs enable the proposed CLIMS++ to generate a more complete and compact activation map for the target objects. Extensive experiments on PASCAL VOC 2012 and MS COCO 2014 datasets show that our CLIMS++ significantly outperforms the previous state-of-the-art methods. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Retrieval Augmented Recipe Generation

Retrieval Augmented Recipe Generation

引用

IEEE Workshop on Applications of computer Vision (WACV)

作者： Guoshan Liu Hailong Yin Bin Zhu Jingjing Chen Chong-Wah Ngo Yu-Gang Jiang Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University Shanghai Collaborative Innovation Center on Intelligent Visual Computing Singapore Management University

ISBN: (数字)9798331510831

ISBN: (纸本)9798331510848

The growing interest in generating recipes from food images has drawn substantial research attention in recent years. Existing works for recipe generation primarily utilize a two-stage training method—first predicting ingredients from a food image and then generating instructions from both the image and ingredients. Large Multi-modal Models (LMMs), which have achieved notable success across a variety of vision and language tasks, shed light on generating both ingredients and instructions directly from images. Nevertheless, LMMs still face the common issue of hallu-cinations during recipe generation, leading to suboptimal performance. To tackle this issue, we propose a retrieval augmented large multimodal model for recipe generation. We first introduce Stochastic Diversified Retrieval Augmentation (SDRA) to retrieve recipes semantically related to the image from an existing datastore as a supplement, integrating them into the prompt to add diverse and rich context to the input image. Additionally, Self-Consistency Ensemble Voting mechanism is proposed to determine the most confident prediction recipes as the final output. It calculates the consistency among generated recipe candidates, which use different retrieval recipes as context for generation. Extensive experiments validate the effectiveness of our proposed method, which demonstrates state-of-the-art (SOTA) performance in recipe generation on the Recipe1M dataset.

关键词： Training computer vision Accuracy Computational modeling Stochastic processes Predictive models Reliability Faces

来源：评论

学校读者我要写书评

暂无评论

BeatKAN: An Efficient and Drum-Attuned Beat Tracking Method Using Kolmogorov-Arnold Networks

BeatKAN: An Efficient and Drum-Attuned Beat Tracking Method ...

引用

International Conference on Acoustics, Speech, and Signal processing (ICASSP)

作者： Zhicheng Zhang Ganghui Ru Wei Li School of Computer Science and Technology Fudan University Shanghai China School of Computer Science and Technology Shanghai Key Laboratory of Intelligent Information Processing Fudan University Shanghai China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

In this paper, we propose an efficient and drum-attuned beat tracking method based on Kolmogorov-Arnold networks (KAN). Traditional MLP-based frameworks struggle with complex musical signals due to limited capacity in modeling intricate patterns. Inspired by KAN’s efficient ability to capture complex time-frequency relationships, we leverage it to enhance our model by employing learnable nonlinear activation functions on convolutional kernels. Additionally, we utilize music source separation techniques to extract drum tracks from the original audio, thereby expanding the existing beat datasets and simulating the human perception of beats through drum sounds. Experimental results demonstrate that our approach significantly reduces the parameter count while maintaining high accuracy in beat tracking.

关键词： Time-frequency analysis Accuracy Source separation Convolution information retrieval Acoustics Kernel Speech processing

来源：评论

学校读者我要写书评

暂无评论

Automated Clinical Summary Generation via Integrating Structured and Unstructured Data 12th

Automated Clinical Summary Generation via Integrating Struct...

引用

12th CCF Conference on BigData, BigData 2024

作者： Fu, Jiaojiao Yang, Bowen Guo, Yi Zhou, Yangfan Wang, Xin School of Information Science and Engineering East China University of Science and Technology Shanghai China Shanghai Key Lab. of Intelligent Information Processing Fudan University Shanghai China School of Computer Science Fudan University Shanghai China

ISBN: (纸本)9789819610235

Automatically generating clinical texts can significantly reduce the time physicians spend on clinical data recording, which is particularly important for developing countries where physicians are extremely busy due to a severe shortage. This work automatically generates discharge summaries as a case to explore the methods and feasibilities of automatic clinical text summarization. Existing work typically uses either structured or unstructured data alone to generate discharge summaries. However, the content generated often has issues such as being overly verbose, lacking focus, or omitting significant information, especially key indicators and medications. This work innovatively proposes a data integration-based clinical text generation approach, using content generated from unstructured clinical data as the basis and supplementing it with text generated from structured clinical data. This study utilizes advanced natural language processing algorithms and models to create clinical texts. It addresses the challenges of lacking datasets suitable for fine-tuning pre-trained models and combining the advantages rather than the disadvantages of both types to produce discharge summaries. Experimental results show that the structured supplementation approach can effectively improve the generation of clinical texts. This work demonstrates that clinical texts generated using existing natural language processing technologies still do not meet the demands of medical practice, pointing out the need to develop further text generation technologies tailored to the characteristics of clinical data. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Data assimilation

来源：评论

学校读者我要写书评

暂无评论

Beyond Boundaries: Learning a Universal Entity Taxonomy across Datasets and Languages for Open Named Entity Recognition 31

Beyond Boundaries: Learning a Universal Entity Taxonomy acro...

引用

31st International Conference on Computational Linguistics, COLING 2025

作者： Yang, Yuming Zhao, Wantong Huang, Caishuang Ye, Junjie Wang, Xiao Zheng, Huiyuan Nan, Yang Wang, Yuran Xu, Xueying Huang, Kaixin Zhang, Yunke Gui, Tao Zhang, Qi Huang, Xuanjing School of Computer Science Fudan University China Honor Device Co. Ltd. Institute of Modern Languages and Linguistics Fudan University China Pengcheng Laboratory China Research Institute of Intelligent Complex Systems Fudan University China Shanghai Key Laboratory of Intelligent Information Processing China

ISBN: (纸本)9798891761964

Open Named Entity Recognition (NER), which involves identifying arbitrary types of entities from arbitrary domains, remains challenging for Large Language Models (LLMs). Recent studies suggest that fine-tuning LLMs on extensive NER data can boost their performance. However, training directly on existing datasets neglects their inconsistent entity definitions and redundant data, limiting LLMs to dataset-specific learning and hindering out-of-domain adaptation. To address this, we present B2NERD, a compact dataset designed to guide LLMs' generalization in Open NER under a universal entity taxonomy. B2NERD is refined from 54 existing English and Chinese datasets using a two-step process. First, we detect inconsistent entity definitions across datasets and clarify them by distinguishable label names to construct a universal taxonomy of 400+ entity types. Second, we address redundancy using a data pruning strategy that selects fewer samples with greater category and semantic diversity. Comprehensive evaluation shows that B2NERD significantly enhances LLMs' Open NER capabilities. Our B2NER models, trained on B2NERD, outperform GPT-4 by 6.8-12.0 F1 points and surpass previous methods in 3 out-of-domain benchmarks across 15 datasets and 6 languages. The data, models, and code are publicly available at https://***/UmeanNever/B2NER. © 2025 Association for Computational Linguistics.

关键词： Taxonomies

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：