检索结果-内蒙古大学图书馆

HiSMatch: Historical Structure Matching based Temporal Knowledge Graph Reasoning

学校读者我要写书评

暂无评论

HiSMatch: Historical Structure Matching based Temporal Knowl...

2022 Findings of the Association for Computational Linguistics: EMNLP 2022

作者： Li, Zixuan Hou, Zhongni Guan, Saiping Jin, Xiaolong Peng, Weihua Bai, Long Lyu, Yajuan Li, Wei Guo, Jiafeng Cheng, Xueqi School of Computer Science and Technology University of Chinese Academy of Sciences China CAS Key Laboratory of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences China Baidu Inc

A Temporal Knowledge Graph (TKG) is a sequence of KGs with respective timestamps, which adopts quadruples in the form of (subject, relation, object, timestamp) to describe dynamic facts. TKG reasoning has facilitated many real-world applications via answering such queries as (query entity, query relation, ?, future timestamp) about future. This is actually a matching task between a query and candidate entities based on their historical structures, which reflect behavioral trends of the entities at different timestamps. In addition, recent KGs provide background knowledge of all the entities, which is also helpful for the matching. Thus, in this paper, we propose the Historical Structure Matching (HiSMatch) model. It applies two structure encoders to capture the semantic information contained in the historical structures of the query and candidate entities. Besides, it adopts another encoder to integrate the background knowledge into the model. TKG reasoning experiments on six benchmark datasets demonstrate the significant improvement of the proposed HiSMatch model, with up to 5.6% performance improvement in MRR, compared to the state-of-the-art baselines. © 2022 Association for Computational Linguistics.

关键词： Knowledge graph

MORE: Multi-mOdal REtrieval Augmented Generative Commonsense Reasoning

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Cui, Wanqing Bi, Keping Guo, Jiafeng Cheng, Xueqi CAS Key Lab of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences Beijing China University of Chinese Academy of Sciences Beijing China

Since commonsense information has been recorded significantly less frequently than its existence, language models pre-trained by text generation have difficulty to learn sufficient commonsense knowledge. Several studies have leveraged text retrieval to augment the models' commonsense ability. Unlike text, images capture commonsense information inherently but little effort has been paid to effectively utilize them. In this work, we propose a novel Multi-mOdal REtrieval (MORE) augmentation framework, to leverage both text and images to enhance the commonsense ability of language models. Extensive experiments on the Common-Gen task have demonstrated the efficacy of MORE based on the pre-trained models of both single and multiple modalities. © 2024, CC BY-NC-SA.

关键词： Computational linguistics

A Secure Device Management Scheme with Audio-Based Location Distinction in IoT

学校读者我要写书评

暂无评论

Computer Modeling in Engineering & sciences 2024年第1期138卷 939-956页

作者： Haifeng Lin Xiangfeng Liu Chen Chen Zhibo Liu Dexin Zhao Yiwen Zhang Weizhuang Li Mingsheng Cao College of Economics and Management Nanjing University of Aeronautics and AstronauticsNanjing211106China Chengdu Aircraft Industrial(Group)Co. Ltd.Chengdu610073China Academy of Military Sciences of PLA Beijing100091China School of Information and Software Engineering University of Electronic Science and Technology of ChinaChengdu610054China Ningbo Web King Technology Joint Stock Co. Ltd.Ningbo315000China The Network and Data Security Key Laboratory of Sichuan Province University of Electronic Science and Technology of ChinaChengdu610054China

Identifying a device and detecting a change in its position is critical for secure devices management in the Internet of Things(IoT).In this paper,a device management system is proposed to track the devices by using audio-based location distinction *** the proposed scheme,traditional cryptographic techniques,such as symmetric encryption algorithm,RSA-based signcryption scheme,and audio-based secure transmission,are utilized to provide authentication,non-repudiation,and confidentiality in the information interaction of the management ***,an audio-based location distinction method is designed to detect the position change of the ***,the audio frequency response(AFR)of several frequency points is utilized as a device *** device signature has the features as follows.(1)Hardware Signature:different pairs of speaker and microphone have different signatures;(2)Distance Signature:in the same direction,the signatures are different at different distances;and(3)Direction Signature:at the same distance,the signatures are different in different *** on the features above,amovement detection algorithmfor device identification and location distinction is ***,a secure communication protocol is also proposed by using traditional cryptographic techniques to provide integrity,authentication,and non-repudiation in the process of information interaction between devices,Access Points(APs),and *** experiments are conducted to evaluate the performance of the proposed *** experimental results show that the proposedmethod has a good performance in accuracy and energy consumption.

关键词： Acoustic hardware fingerprinting device management IoT location distinction

Robust Information Retrieval

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Liu, Yu-An Zhang, Ruqing Guo, Jiafeng de Rijke, Maarten CAS Key Lab of Network Data Science and Technology ICT CAS University of Chinese Academy of Sciences Beijing China University of Amsterdam Amsterdam Netherlands

Beyond effectiveness, the robustness of an information retrieval (IR) system is increasingly attracting attention. When deployed, a critical technology such as IR should not only deliver strong performance on average but also have the ability to handle a variety of exceptional situations. In recent years, research into the robustness of IR has seen significant growth, with numerous researchers offering extensive analyses and proposing myriad strategies to address robustness challenges. In this tutorial, we first provide background information covering the basics and a taxonomy of robustness in IR. Then, we examine adversarial robustness and out-of-distribution (OOD) robustness within IR-specific contexts, extensively reviewing recent progress in methods to enhance robustness. The tutorial concludes with a discussion on the robustness of IR in the context of large language models (LLMs), highlighting ongoing challenges and promising directions for future research. This tutorial aims to generate broader attention to robustness issues in IR, facilitate an understanding of the relevant literature, and lower the barrier to entry for interested researchers and practitioners. © 2024, CC BY.

关键词： Search engines

Research on Zenith Tropospheric Delay Model Based on TCN Improving HGPT2 Model 14th

学校读者我要写书评

暂无评论

Research on Zenith Tropospheric Delay Model Based on TCN Imp...

14th China Satellite Navigation Conference, CSNC 2024

作者： Li, Dengao Shi, Danyang Zhao, Jumin Wu, Fanming Yan, Liangquan Feng, Ran Zhang, Xinfang Zhao, Jinhua College of Data Science Taiyuan University of Technology Taiyuan030024 China College of Information and Computer Taiyuan University of Technology Taiyuan030024 China Key Laboratory of Big Data Fusion Analysis and Application of Shanxi Province Taiyuan030024 China Intelligent Perception Engineering Technology Center of Shanxi Taiyuan030024 China Shanxi Province Engineering Technology Research Center of Spatial Information Network Taiyuan030024 China

ISBN: (纸本)9789819969272

The zenith tropospheric delay (ZTD) obtained by global navigation satellite system (GNSS) atmospheric sounding is a pivotal data source for water vapor monitoring. Meteorological changes in Antarctica play an important role in analysis of the global climate, but factors such as complex climatic conditions can limit the collection of meteorological data needed for ZTD retrieval. Therefore, it is necessary to establish high-precision ZTD models that do not rely on measured meteorological data. The existing global pressure and temperature 3 (GPT3) model has limited ability to capture complex weather variations and cannot obtain high-precision GPT3_ZTD. To address the above issues, this research proposes a high-precision ZTD model through using temporal convolutional network (TCN) for improving the hourly global pressure and temperature 2 (HGPT2) model. The HGTP2 model based on Fourier analysis and the time-segmentation concept can consider the linear trend of climate variations, simultaneously the TCN is introduced to simulate the long-term temporal dependence of HGPT2_ZTD, so as to achieve the goal of obtaining high-precision ZTD without relying on measured meteorological data. The experimental results show that the precision of TCN_ZTD obtained from the proposed model is higher than GPT3_ZTD and HGPT2_ZTD when using the GNSS_ZTD observations obtained from 35 stations in Antarctica as a reference, and the improvement of the model is obvious. © 2024, Aerospace Information Research Institute.

关键词： Fourier analysis

Prediction of Ionospheric TEC Based on BLS-LSTM-GRU Hybrid Model 14th

学校读者我要写书评

暂无评论

Prediction of Ionospheric TEC Based on BLS-LSTM-GRU Hybrid M...

14th China Satellite Navigation Conference, CSNC 2024

作者： Li, Dengao Zhang, Xinfang Zhao, Jumin Wu, Fanming Feng, Ran Zhao, Jinhua Shi, Danyang College of Data Science Taiyuan University of Technology Taiyuan030024 China College of Information and Computer Taiyuan University of Technology Taiyuan030024 China Key Laboratory of Big Data Fusion Analysis and Application of Shanxi Province Taiyuan030024 China Intelligent Perception Engineering Technology Center of Shanxi Taiyuan030024 China Shanxi Province Engineering Technology Research Center of Spatial Information Network Taiyuan030024 China

ISBN: (纸本)9789819969272

In satellite communications, global navigation satellite systems(GNSS) and other important space activities, the value of total electron content (TEC) in the ionosphere directly affects the size of ionospheric delay. Achieving accurate prediction of ionospheric total electron content through modeling can effectively improve the reliability and accuracy of GNSS positioning. Considering the high prediction accuracy of long short-term memory neural network (LSTM) and the fast training speed of gated recurrent unit network (GRU), but it is easy to fall into the local optimum, while the width learning system(BLS) can effectively avoid this drawback. Therefore, this paper proposes a model for predicting the total electron content of the ionosphere, which consists of a broad learning system, an LSTM network and a GRU networks. Six parameters of TEC, geomagnetic index Dst, Kp, Ap, solar activity index F10.7, and hour (HD) are selected as input features from the observed data to predict the variation of the total ionospheric electron content. The results generate prediction maps of the total ionospheric electron content and are compared with empirical models and conventional neural network models at different times and solar activity conditions. The experimental results show that the root mean square error (RMSE) of the prediction results of this BLS-LSTM-GRU model is decreased by 23.07%, 17.48% and 9.46% compared with the IRI-2016, NeQuick and CNN-LSTM prediction models, respectively, with higher prediction accuracy. © 2024, Aerospace Information Research Institute.

关键词： Long short-term memory

CAME: Competitively Learning a Mixture-of-Experts Model for First-stage Retrieval

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Cai, Yinqiong Fan, Yixing Bi, Keping Guo, Jiafeng Chen, Wei Zhang, Ruqing Cheng, Xueqi CAS Key Lab of Network Data Science and Technology ICT CAS University of Chinese Academy of Sciences Beijing China

The first-stage retrieval aims to retrieve a subset of candidate documents from a huge collection both effectively and efficiently. Since various matching patterns can exist between queries and relevant documents, previous work tries to combine multiple retrieval models to find as many relevant results as possible. The constructed ensembles, whether learned independently or jointly, do not care which component model is more suitable to an instance during training. Thus, they cannot fully exploit the capabilities of different types of retrieval models in identifying diverse relevance patterns. Motivated by this observation, in this paper, we propose a Mixture-of-Experts (MoE) model consisting of representative matching experts and a novel competitive learning mechanism to let the experts develop and enhance their expertise during training. Specifically, our MoE model shares the bottom layers to learn common semantic representations and uses differently structured upper layers to represent various types of retrieval experts. Our competitive learning mechanism has two stages: (1) a standardized learning stage to train the experts equally to develop their capabilities to conduct relevance matching;(2) a specialized learning stage where the experts compete with each other on every training instance and get rewards and updates according to their performance to enhance their expertise on certain types of samples. Experimental results on three retrieval benchmark datasets show that our method significantly outperforms the state-of-the-art baselines. Copyright © 2023, The Authors. All rights reserved.

关键词： Semantics

Inducing Causal Structure for Abstractive Text Summarization

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Chen, Lu Zhang, Ruqing Huang, Wei Chen, Wei Guo, Jiafeng Cheng, Xueqi CAS Key Lab of Network Data Science and Technology ICT CAS University of Chinese Academy of Sciences Beijing China

The mainstream of data-driven abstractive summarization models tends to explore the correlations rather than the causal relationships. Among such correlations, there can be spurious ones which suffer from the language prior learned from the training corpus and therefore undermine the overall effectiveness of the learned model. To tackle this issue, we introduce a Structural Causal Model (SCM) to induce the underlying causal structure of the summarization data. We assume several latent causal factors and non-causal factors, representing the content and style of the document and summary. Theoretically, we prove that the latent factors in our SCM can be identified by fitting the observed training data under certain conditions. On the basis of this, we propose a Causality Inspired Sequence-to-Sequence model (CI-Seq2Seq) to learn the causal representations that can mimic the causal factors, guiding us to pursue causal information for summary generation. The key idea is to reformulate the Variational Auto-encoder (VAE) to fit the joint distribution of the document and summary variables from the training corpus. Experimental results on two widely used text summarization datasets demonstrate the advantages of our approach. © 2023, CC BY-NC-ND.

关键词： Text processing

Multi-granular Adversarial Attacks against Black-box Neural Ranking Models

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Liu, Yu-An Zhang, Ruqing Guo, Jiafeng de Rijke, Maarten Fan, Yixing Cheng, Xueqi CAS Key Lab of Network Data Science and Technology ICT CAS University of Chinese Academy of Sciences Beijing China University of Amsterdam Amsterdam Netherlands

Adversarial ranking attacks have gained increasing attention due to their success in probing vulnerabilities, and, hence, enhancing the robustness, of neural ranking models. Conventional attack methods employ perturbations at a single granularity, e.g., word or sentence level, to target documents. However, limiting perturbations to a single level of granularity may reduce the flexibility of adversarial examples, thereby diminishing the potential threat of the attack. Therefore, we focus on generating high-quality adversarial examples by incorporating multi-granular perturbations. Achieving this objective involves tackling a combinatorial explosion problem, which requires identifying an optimal combination of perturbations across all possible levels of granularity, positions, and textual pieces. To address this challenge, we transform the multi-granular adversarial attack into a sequential decision-making process, where perturbations in the next attack step build on the perturbed document in the current attack step. Since the attack process can only access the final state without direct intermediate signals, we use reinforcement learning to perform multi-granular attacks. During the reinforcement learning process, two agents work cooperatively to identify multi-granular vulnerabilities as attack targets and organize perturbation candidates into a final perturbation sequence. Experimental results show that our attack method surpasses prevailing baselines in both attack effectiveness and imperceptibility. Copyright © 2024, The Authors. All rights reserved.

关键词： Reinforcement learning