检索结果-内蒙古大学图书馆

2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024

作者： Chen, Meiqi Cao, Yixin Zhang, Yan Lu, Chaochao State Key Laboratory of General Artificial Intelligence Peking University Beijing China School of Intelligence Science and Technology Peking University China School of Computer Science Fudan University China Shanghai Artificial Intelligence Laboratory China

ISBN: (纸本)9798891761681

Recent advancements in Large Language Models (LLMs) have facilitated the development of Multimodal LLMs (MLLMs). Despite their impressive capabilities, MLLMs often suffer from over-reliance on unimodal biases (e.g., language bias and vision bias), leading to incorrect answers in complex multimodal tasks. To investigate this issue, we propose a causal framework to interpret the biases in Visual Question Answering (VQA) problems. Within this framework, we conduct an in-depth causal analysis to assess the causal effect of these biases on MLLM predictions. Based on the analysis, we introduce 1) a novel MORE dataset with 12,000 challenging VQA instances requiring multi-hop reasoning and overcoming unimodal biases. 2) a causality-enhanced agent framework CAVE that guides models to comprehensively integrate information from different modalities and mitigate biases. Our experiments show that MLLMs perform poorly on MORE, indicating strong unimodal biases and limited semantic understanding. However, when integrated with our CAVE, promising improvements in reasoning and bias mitigation can be seen. These findings provide important insights for the development of more robust MLLMs and contribute to the broader goal of advancing multimodal AI systems capable of deeper understanding and reasoning. Our project page is at https://***/OpenCausaLab/MORE. © 2024 Association for Computational Linguistics.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

SpikingMiniLM: energy-efficient spiking transformer for natural language understanding

引用

science China(Information sciences) 2024年第10期67卷 115-128页

作者： Jiayu ZHANG Jiangrong SHEN Zeke WANG Qinghai GUO Rui YAN Gang PAN Huajin TANG College of Computer Science and Technology Zhejiang University The State Key Lab of Brain-Machine Intelligence Zhejiang University Collaborative Innovation Center of Artificial Intelligence Zhejiang University Advanced Computing and Storage Laboratory Huawei Technologies Co. Ltd. College of Computer Science and Technology Zhejiang University of Technology MOE Frontier Science Center for Brain Science and Brain-Machine Integration Zhejiang University

In the era of large-scale pretrained models, artificial neural networks（ANNs） have excelled in natural language understanding（NLU） tasks. However, their success often necessitates substantial computational resources and energy consumption. To address this, we explore the potential of spiking neural networks（SNNs） in NLU——a promising avenue with demonstrated advantages, including reduced power consumption and improved efficiency due to their event-driven characteristics. We propose the SpikingMiniLM,a novel spiking Transformer model tailored for natural language understanding. We first introduce a multi-step encoding method to convert text embeddings into spike trains. Subsequently, we redesign the attention mechanism and residual connections to make our model operate on the pure spike-based paradigm without any normalization technique. To facilitate stable and fast convergence, we propose a general parameter initialization method grounded in the stable firing rate principle. Furthermore, we apply an ANN-to-SNN knowledge distillation to overcome the challenges of pretraining SNNs. Our approach achieves a macro-average score of 75.5 on the dev sets of the GLUE benchmark, retaining 98% of the performance exhibited by the teacher model MiniLMv2. Our smaller model also achieves similar performance to BERTMINIwith fewer parameters and much lower energy consumption, underscoring its competitiveness and resource efficiency in NLU tasks.

关键词： spiking neural networks natural language understanding spiking Transformer spike-based attention multi-step encoding ANN-to-SNN distillation

来源：评论

学校读者我要写书评

暂无评论

Dual Encoder-Decoder Shifted Window-Based Transformer Network for Polyp Segmentation with Self-Learning Approach

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on artificial intelligence 2024年第7期5卷 3456-3469页

作者： Lijin, P. Ullah, Mohib Vats, Anuja Cheikh, Faouzi Alaya Kumar, Santhosh Nair, Madhu S. Cochin University of Science and Technology Artificial Intelligence & Computer Vision Laboratory Department of Computer Science Kerala Kochi682022 India Norwegian University of Science and Technology Gjovik2815 Norway Norwegian University of Science and Technology Norwegian Colour and Visual Computing Laboratory Gjovik2815 Norway

According to WHO reports, cancer is the leading cause of death worldwide. The second most prevalent cause of cancer-related death in both men and women is colorectal cancer (CRC). One potential approach for reducing the severity of colon cancer is to utilize automatic segmentation and detection of colorectal polyps in colonoscopy videos. This technology can assist endoscopists in quickly identifying colorectal disease, leading to earlier intervention and better patient Quality of Life (QoL). In this article, we propose a self-supervised transformer based dual encoder-decoder architecture named P-SwinNet for polyps segmentation in colonoscopy images. The P-SwinNet adapts the dual encoder-decoder type of model to enhance the feature maps by sharing multiscale information from the encoder to the decoder. The proposed model uses multiple dilated convolutions to enlarge the field of view to gather more information without increasing the computational cost and the loss of spatial information. We also leverage a large-scale unlabeled dataset for training our model using the self-learning strategy of Barlow twins. Additionally, to capture the long-range dependencies in the data, we used a shift window-based approach that computes global attention. We extensively evaluate our model against state-of-the-art algorithms. The quantitative results show that the proposed P-SwinNet achieves a mean dice score of 0.87 and a mean Intersection over Union (IoU) of 0.82 on five datasets used in our study. This performance demonstrates a substantial advancement over existing similar works, highlighting the advantage and novelty of our proposed approach in the field of medical image segmentation. © 2020 IEEE.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Analysis and Classification of Fake News Using Sequential Pattern Mining

引用

Big Data Mining and Analytics 2024年第3期7卷 942-963页

作者： M.Zohaib Nawaz M.Saqib Nawaz Philippe Fournier-Viger Yulin He College of Computer Science and Software Engineering Shenzhen UniversityShenzhen 518060 Guangdong Laboratory of Artificial Intelligence and Digital Economy(SZ) Shenzhen 518107China Department of Computer Science Faculty of Computing and Information TechnologyUnivesity of SargodhaSargodha 40100Pakistan

Disinformation,often known as fake news,is a major issue that has received a lot of attention *** researchers have proposed effective means of detecting and addressing *** machine and deep learning based methodologies for classification/detection of fake news are content-based,network(propagation)based,or multimodal methods that combine both textual and visual *** introduce here a framework,called FNACSPM,based on sequential pattern mining(SPM),for fake news analysis and *** this framework,six publicly available datasets,containing a diverse range of fake and real news,and their combination,are first transformed into a proper ***,algorithms for SPM are applied to the transformed datasets to extract frequent patterns(and rules)of words,phrases,or linguistic *** obtained patterns capture distinctive characteristics associated with fake or real news content,providing valuable insights into the underlying structures and commonalities of ***,the discovered frequent patterns are used as features for fake news *** framework is evaluated with eight classifiers,and their performance is assessed with various *** experiments were performed and obtained results show that FNACSPM outperformed other state-of-the-art approaches for fake news classification,and that it expedites the classification task with high accuracy.

关键词： disinformation fake news sequential pattern mining(SPM) frequent patterns classification

来源：评论

学校读者我要写书评

暂无评论

BatStyler: Advancing Multi-category Style Generation for Source-free Domain Generalization

引用

IEEE Transactions on Circuits and Systems for Video Technology 2025年第6期35卷 5419-5430页

作者： Xu, Xiusheng Qi, Lei Zhou, Jingyang Geng, Xin Southeast University School of Computer Science and Engineering China Interdisciplinary Applications Southeast University Key Laboratory of New Generation Artificial Intelligence Technology Ministry of Education 211189 China

Source-Free Domain Generalization (SFDG) aims to develop a model that performs on unseen domains without relying on any source domains. However, the implementation remains constrained due to the unavailability of training data. Research on SFDG focus on knowledge transfer of multi-modal models and style synthesis based on joint space of multiple modalities, thus eliminating the dependency on source domain images. However, existing works primarily work for multi-domain and less-category configuration, but performance on multi-domain and multi-category configuration is relatively poor. In addition, the efficiency of style synthesis also deteriorates in multi-category scenarios. How to efficiently synthesize sufficiently diverse data and apply it to multi-category configuration is a direction with greater practical value. In this paper, we propose a method called BatStyler, which is utilized to improve the capability of style synthesis in multi-category scenarios. BatStyler consists of two modules: Coarse Semantic Generation and Uniform Style Generation modules. The Coarse Semantic Generation module extracts coarse-grained semantics to prevent the compression of space for style diversity learning in multi-category configuration, while the Uniform Style Generation module provides a template of styles that are uniformly distributed in space and implements parallel training. Extensive experiments demonstrate that our method exhibits comparable performance on less-category datasets, while surpassing state-of-the-art methods on multi-category datasets. Code is available here. © 1991-2012 IEEE.

关键词： Knowledge transfer

来源：评论

学校读者我要写书评

暂无评论

Selective arguments representation with dual relation-aware network for video situation recognition

引用

Neural Computing and Applications 2024年第17期36卷 9945-9961页

作者： Liu, Wei He, Qing Wang, Chao Peng, Yan Xie, Shaorong School of Computer Engineering and Science Shanghai University Shanghai200444 China Shanghai Artificial Intelligence Laboratory Shanghai201114 China School of Future Technology Shanghai University Shanghai200444 China Institute of Artificial Intelligence Shanghai University Shanghai200444 China

Argument visual states are helpful for detecting structured components of events in videos, and existing methods tend to use object detectors to generate their candidates. However, directly leveraging object features captured by bounding boxes overlooks a deep understanding of object relations and differences between them and real arguments. In this work, we propose a novel framework to generate selective contextual representations of videos, thereby reducing the interference of useless or incorrect object features. Firstly, we construct grid-based object features as graphs based on the internal grid connection and then use graph convolutional network to execute feature aggregation. Secondly, a weighted geometric attention module is designed to obtain the contextual representation of objects, which explicitly combines visual similarity and geometric correlation with different importance proportions. Then, we propose a dual relation-aware selection module for further feature selection. Finally, we utilize labels as the ladder to bridge the gap between object features and semantic roles, while considering the proximity in the semantic space. Experimental results and extensive ablation studies on the VidSitu indicate that our method effectively obtains a deep understanding of events in videos and outperforms state-of-the-art models. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Federated Multiarmed Bandits Under Byzantine Attacks

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on artificial intelligence 2025年第6期6卷 1488-1501页

作者： Saday, Artun Demirel, Ilker Yildirim, Yigit Tekin, Cem Bilkent University Department of Electrical and Electronics Engineering Ankara06800 Turkey Massachusetts Institute of Technology Computer Science & Artificial Intelligence Laboratory CambridgeMA02139 United States

Multiarmed bandits (MAB) is a sequential decision-making model in which the learner controls the trade-off between exploration and exploitation to maximize its cumulative reward. Federated multiarmed bandits (FMAB) is an emerging framework where a cohort of learners with heterogeneous local models play an MAB game and communicate their aggregated feedback to a server to learn a globally optimal arm. Two key hurdles in FMAB are communication-efficient learning and resilience to adversarial attacks. To address these issues, we study the FMAB problem in the presence of Byzantine clients who can send false model updates threatening the learning process. We analyze the sample complexity and the regret of β-optimal arm identification. We borrow tools from robust statistics and propose a median-of-means (MoM)-based online algorithm, Fed-MoM-UCB, to cope with Byzantine clients. In particular, we show that if the Byzantine clients constitute less than half of the cohort, the cumulative regret with respect to \beta-optimal arms is bounded over time with high probability, showcasing both communication efficiency and Byzantine resilience. We analyze the interplay between the algorithm parameters, a discernibility margin, regret, communication cost, and the arms’ suboptimality gaps. We demonstrate Fed-MoM-UCB's effectiveness against the baselines in the presence of Byzantine attacks via experiments. © 2020 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Simultaneous Learning Intensity and Optical Flow from High-speed Spike Stream

引用

IEEE Transactions on Circuits and Systems for Video Technology 2024年第6期35卷 5126-5139页

作者： Zhu, Lin Yan, Weiquan Chang, Yi Tian, Yonghong Huang, Hua Beijing Institute of Technology School of Computer Science and Technology Beijing100081 China Peng Cheng Laboratory Shenzhen518055 China Huazhong University of Science and Technology School of Artificial Intelligence and Automation Wuhan430074 China Peking University School of Computer Science Beijing100081 China Beijing Normal University School of Artificial Intelligence Beijing100875 China

Bio-inspired vision sensors, which emulate the human retina by recording light intensity as binary spikes, have gained increasing interest in recent years. Among them, the spike camera is capable of perceiving fine textures by simulating a small retinal region called the fovea and producing high temporal resolution (20,000 Hz) spatiotemporal spike streams. To bridge the gap between binary spike streams and human vision in high-speed scenes, reconstructing intensity and optical flow from high temporal resolution spikes is particularly important. In this paper, we present a hybrid SNN-ANN network designed for simultaneous intensity and optical flow learning from spike streams. To adaptively extract spatial and temporal features from continuous spike streams, we propose a spiking neuron module with dense connections that efficiently processes both short-term and long-term spike data, while maintaining low power consumption characteristics. Subsequently, we introduce two decoders for optical flow and intensity estimation that complement each other. A temporal-aware warping module, based on flow features, is specifically designed to align the temporal features of the intensity decoder, thereby reducing motion artifacts. Concurrently, improved intensity features contribute to more accurate flow feature predictions, resulting in a mutually beneficial relationship within our network. To evaluate the effectiveness of our proposed network, we conduct experiments on both simulated and real spike datasets. Our network outperforms existing state-of-the-art spike-based reconstruction and optical flow estimation methods, demonstrating its potential for advancing the field of bio-inspired vision sensors. Our code is available at https://***/LinZhu111/SLIO. © 1991-2012 IEEE.

关键词： Optical flows

来源：评论

学校读者我要写书评

暂无评论

End-to-end learning of 3D phase-only holograms for holographic display

引用

Light(science & Applications) 2022年第9期11卷 2176-2193页

作者： Liang Shi Beichen Li Wojciech Matusik Computer Science and Artificial Intelligence Laboratory Massachusetts Institute of Technology32 Vassar StCambridgeMA 02139USA

computer-generated holography(CGH)provides volumetric control of coherent wavefront and is fundamental to applications such as volumetric 3D displays,lithography,neural photostimulation,and optical/acoustic ***,deep learning-based methods emerged as promising computational paradigms for CGH synthesis that overcome the quality-runtime tradeoff in conventional simulation/optimization-based ***,the quality of the predicted hologram is intrinsically bounded by the dataset's *** we introduce a new hologram dataset,MIT-CGH-4K-V2,that uses a layered depth image as a data-efficient volumetric 3D input and a two-stage supervised+unsupervised training protocol for direct synthesis of high-quality 3D phase-only *** proposed system also corrects vision aberration,allowing customization for *** experimentally show photorealistic 3D holographic projections and discuss relevant spatial light modulator calibration *** method runs in real-time on a consumer GPU and 5 FPS on an iPhone 13 Pro,promising drastically enhanced performance for the applications above.

关键词： holographic synthesis overcome

来源：评论

学校读者我要写书评

暂无评论

Software Engineering Research in a World with Generative artificial intelligence 24

Software Engineering Research in a World with Generative Art...

引用

44th ACM/IEEE International Conference on Software Engineering, ICSE 2024

作者： Rinard, Martin Computer Science and Artificial Intelligence Laboratory Massachusetts Institute of Technology Department of Electrical Engineering and Computer Science CambridgeMA United States

Generative artificial intelligence systems such as large language models (LLMs) exhibit powerful capabilities that many see as the kind of flexible and adaptive intelligence that previously only humans could exhibit. ... 详细信息

ISBN: (纸本)9798400702174

关键词： Software engineering

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：