检索结果-内蒙古大学图书馆

New Intent Discovery with Attracting and Dispersing Prototype

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhang, Shun Yang, Jian Bai, Jiaqi Yan, Chaoran Li, Tongliang Yan, Zhao Li, Zhoujun State Key Lab of Software Development Environment Beihang University Beijing China School of Cyber Science and Technology Beihang University Beijing China Beijing Information Science and Technology University China Tencent Youtu Lab China

New Intent Discovery (NID) aims to recognize known and infer new intent categories with the help of limited labeled and large-scale unlabeled data. The task is addressed as a feature-clustering problem and recent studies augment instance representation. However, existing methods fail to capture cluster-friendly representations, since they show less capability to effectively control and coordinate within-cluster and between-cluster distances. Tailored to the NID problem, we propose a Robust and Adaptive Prototypical learning (RAP) framework for globally distinct decision boundaries for both known and new intent categories. Specifically, a robust prototypical attracting learning (RPAL) method is designed to compel instances to gravitate toward their corresponding prototype, achieving greater within-cluster compactness. To attain larger between-cluster separation, another adaptive prototypical dispersing learning (APDL) method is devised to maximize the between-cluster distance from the prototype-to-prototype perspective. Experimental results evaluated on three challenging benchmarks (CLINC, BANKING, and StackOverflow) of our method with better cluster-friendly representation demonstrate that RAP brings in substantial improvements over the current state-of-the-art methods (even large language model) by a large margin (average +5.5% improvement). Copyright © 2024, The Authors. All rights reserved.

关键词： Learning systems

Phonemic Adversarial Attack against Audio Recognition in Real World

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Wang, Jiakai Chen, Zhendong Yin, Zixin Yang, Qinghong Liu, Xianglong Zhongguancun Laboratory China College of Software Beihang University China State Key Lab of Software Development Environment Beihang University China

Recently, adversarial attacks for audio recognition have attracted much attention. However, most of the existing studies mainly rely on the coarse-grain audio features at the instance level to generate adversarial noises, which leads to expensive generation time costs and weak universal attacking ability. Motivated by the observations that all audio speech consists of fundamental phonemes, this paper proposes a phonemic adversarial attack (PAT) paradigm, which attacks the fine-grain audio features at the phoneme level commonly shared across audio instances, to generate phonemic adversarial noises, enjoying the more general attacking ability with fast generation speed. Specifically, for accelerating the generation, a phoneme density balanced sampling strategy is introduced to sample quantity less but phonemic features abundant audio instances as the training data via estimating the phoneme density, which substantially alleviates the heavy dependency on the large training dataset. Moreover, for promoting universal attacking ability, the phonemic noise is optimized in an asynchronous way with a sliding window, which enhances the phoneme diversity and thus well captures the critical fundamental phonemic patterns. By conducting extensive experiments, we comprehensively investigate the proposed PAT framework and demonstrate that it outperforms the SOTA baselines by large margins (i.e., at least 11× speed up and 78% attacking ability improvement)1 Copyright © 2022, The Authors. All rights reserved.

关键词： Large dataset

Lifeline Consensus: Protect Small-scale Blockchain From Typical Attacks

学校读者我要写书评

暂无评论

Lifeline Consensus: Protect Small-scale Blockchain From Typi...

International Wireless Communications and Mobile Computing Conference, IWCMC

作者： Kundu Chen Jie Luo State Key Laboratory of Software Development Environment School of Computer Science and Engineering Beihang University Haidian District Beijing China

Blockchain is a distributed network system known for its anonymity, security, and decentralized nature. While blockchains with robust networks such as Bitcoin and Ethereum are proving immutable to attacks, many others are vulnerable to security threats. To address this issue, we proposed the Lifeline consensus, which aims to protect small-scale blockchains by delivering safety from a strong leading blockchain to a relatively vulnerable following blockchain using a new type of data called Coil. Coil carries the proof-of-work evidence created by miners from an arbitrary following blockchain, which the leading blockchain then gathers and generates immutable blocks that contained them. Miners from the following blockchain can use the coils in the leading blockchain for validation, thereby protecting small-scale blockchains from typical blockchain threats such as 51% attacks and selfish mining. We implemented our protocol and encountered adversaries in the local network, and analyzed its performance. The results showed that the Lifeline protocol is dependable with a Trusted Leader. These findings suggest that our protocol has the potential to improve the security of small-scale blockchains and enhance the overall performance of the blockchain network.

关键词：

Know What I don’t Know: Handling Ambiguous and Unanswerable Questions for Text-to-SQL

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Wang, Bing Gao, Yan Li, Zhoujun Lou, Jian-Guang State Key Lab of Software Development Environment Beihang University Beijing China Microsoft Research Asia China

The task of text-to-SQL aims to convert a natural language question into its corresponding SQL query within the context of relational tables. Existing text-to-SQL parsers generate a "plausible" SQL query for an arbitrary user question, thereby failing to correctly handle problematic user questions. To formalize this problem, we conduct a preliminary study on the observed ambiguous and unanswerable cases in text-to-SQL and summarize them into 6 feature categories. Correspondingly, we identify the causes behind each category and propose requirements for handling ambiguous and unanswerable questions. Following this study, we propose a simple yet effective counterfactual example generation approach that automatically produces ambiguous and unanswerable text-to-SQL examples. Furthermore, we propose a weakly supervised DTE (Detecting-Then-Explaining) model for error detection, localization, and explanation. Experimental results show that our model achieves the best result on both real-world examples and generated examples compared with various baselines. We release our data and code at: https://***/wbbeyourself/DTE. © 2022, CC BY-NC-SA.

关键词： Natural language processing systems

CQR-SQL: Conversational Question Reformulation Enhanced Context-Dependent Text-to-SQL Parsers

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Xiao, Dongling Chai, Linzheng Zhang, Qian-Wen Yan, Zhao Li, Zhoujun Cao, Yunbo Tencent Cloud Xiaowei China State Key Lab of Software Development Environment Beihang University Beijing China

Context-dependent text-to-SQL is the task of translating multi-turn questions into database-related SQL queries. Existing methods typically focus on making full use of history context or previously predicted SQL for currently SQL parsing, while neglecting to explicitly comprehend the schema and conversational dependency, such as co-reference, ellipsis and user focus change. In this paper, we propose CQR-SQL, which uses auxiliary Conversational Question Reformulation (CQR) learning to explicitly exploit schema and decouple contextual dependency for multi-turn SQL parsing. Specifically, we first present a schema enhanced recursive CQR method to produce domain-relevant self-contained questions. Secondly, we train CQR-SQL models to map the semantics of multi-turn questions and auxiliary self-contained questions into the same latent space through schema grounding consistency task and tree-structured SQL parsing consistency task, which enhances the abilities of SQL parsing by adequately contextual understanding. At the time of writing, our CQR-SQL achieves new state-of-the-art results on two context-dependent text-to-SQL benchmarks SPARC and COSQL. Copyright © 2022, The Authors. All rights reserved.

关键词： Semantics

SAT Requires Exhaustive Search

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Xu, Ke Zhou, Guangyan State Key Lab of Software Development Environment Beihang University Beijing100083 China Department of Mathematics and Statistics Beijing Technology and Business University Beijing100048 China

In this paper, by constructing extremely hard examples of CSP (with large domains) and SAT (with long clauses), we prove that such examples cannot be solved without exhaustive search, which is stronger than P ≠ NP. This constructive approach for proving impossibility results is very different (and missing) from those currently used in computational complexity theory, but is similar to that used by Kurt Gödel in proving his famous logical impossibility results. Just as shown by Gödel’s results that proving formal unprovability is feasible in mathematics, the results of this paper show that proving computational hardness is not hard in mathematics. Specifically, proving lower bounds for many problems, such as 3-SAT, can be challenging because these problems have various effective strategies available for avoiding exhaustive search. However, in cases of extremely hard examples, exhaustive search may be the only viable option, and proving its necessity becomes more straightforward. Consequently, it makes the separation between SAT (with long clauses) and 3-SAT much easier than that between 3-SAT and 2-SAT. Finally, the main results of this paper demonstrate that the fundamental difference between the syntax and the semantics revealed by Gödel’s results also exists in CSP and SAT. © 2023, CC BY.

关键词： Computational complexity

Enhancing the Transferability of Adversarial Attacks with Stealth Preservation

学校读者我要写书评

暂无评论

Enhancing the Transferability of Adversarial Attacks with St...

IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)

作者： Xinwei Zhang Tianyuan Zhang Yitong Zhang Shuangcheng Liu School of Computer Science and Engineering Beihang University Beijing China State Key Lab of Software Development Environment Beihang University Beijing China Shen Yuan Honors College Beihang University Beijing China

ISBN: (数字)9798350365474

ISBN: (纸本)9798350365481

Deep neural networks are susceptible to attacks from adversarial examples in recent years. Especially, the black-box attacks cause a more serious threat to practical applications. However, while most existing black-box attacks have achieved a high success rate in deceiving models, they have not focused on the stealthiness of adversarial examples, often exhibiting suspicious visual appearances. To address this issue, this paper proposes the Mask Momentum Iterative Attack (MMIA), which introduces a masking mechanism and adopts an optimal perturbation strategy to identify regions of an image most vulnerable to attacks. This approach effectively ensures the transferability and stealthiness of adversarial examples. Simultaneously, by integrating image enhancement techniques and temporal and spatial momentum terms into the iterative process of the attack, we prevent the attack from getting stuck in local optima, further improving the transferability of adversarial examples. To enhance the success rate of black-box attacks, we apply MMIA to a model ensemble using a joint optimization strategy. We demonstrate that adversarially trained models with a strong defense ability are also susceptible to our black-box attacks. We conduct extensive experiments on classification tasks using common vision models, and our results significantly demonstrate the superiority of our method over state-of-the-art approaches when considering both transferability and stealthiness.

关键词： Visualization Fuses Computational modeling Perturbation methods Closed box Pattern recognition Iterative methods

HILL: Hierarchy-aware Information Lossless Contrastive Learning for Hierarchical Text Classification

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhu, He Wu, Junran Liu, Ruomei Hou, Yue Yuan, Ze Li, Shangzhe Pan, Yicheng Xu, Ke State Key Lab of Software Development Environment Beihang University Beijing100191 China Zhongguancun Laboratory Beijing100094 China School of Statistics and Mathematics Central University of Finance and Economics Beijing100081 China

Existing self-supervised methods in natural language processing (NLP), especially hierarchical text classification (HTC), mainly focus on self-supervised contrastive learning, extremely relying on human-designed augmentation rules to generate contrastive samples, which can potentially corrupt or distort the original information. In this paper, we tend to investigate the feasibility of a contrastive learning scheme in which the semantic and syntactic information inherent in the input sample is adequately reserved in the contrastive samples and fused during the learning process. Specifically, we propose an information lossless contrastive learning strategy for HTC, namely Hierarchy-aware Information Lossless contrastive Learning (HILL), which consists of a text encoder representing the input document, and a structure encoder directly generating the positive sample. The structure encoder takes the document embedding as input, extracts the essential syntactic information inherent in the label hierarchy with the principle of structural entropy minimization, and injects the syntactic information into the text representation via hierarchical representation learning. Experiments on three common datasets are conducted to verify the superiority of HILL. © 2024, CC BY-NC-SA.

关键词： Entropy

基于时空数据的地下空间基础设施智能监测系统

学校读者我要写书评

暂无评论

Engineering 2023年第6期25卷 194-203,M0008页

作者：杜博文叶俊辰朱合华孙磊磊杜彦良 State Key Laboratory of Software Development Environment Beihang UniversityBeijing 100191China Department of Geotechnical Engineering College of Civil EngineeringTongji UniversityShanghai 200092China Institute of Urban Smart Transportation and Safety Maintenance Shenzhen UniversityShenzhen 518060China

基于时空大数据的智能感知、机理认知和劣化预知,不仅促进了基础设施安全的发展,同时也是基础设施建设向智能化转变的基础理论和关键技术。地下空间利用的发展,形成了深、大、聚的三大特征和立体的城市布局。然而,与地上的建筑物和桥梁... 详细信息

基于时空大数据的智能感知、机理认知和劣化预知,不仅促进了基础设施安全的发展,同时也是基础设施建设向智能化转变的基础理论和关键技术。地下空间利用的发展,形成了深、大、聚的三大特征和立体的城市布局。然而,与地上的建筑物和桥梁相比,发生在地下的疾病和退化更为隐蔽,难以识别,在建设和服务期间仍然存在许多挑战。针对这一问题,本文总结了现有的方法,并在现实世界的空间安全管理中评估了它们的长处和短处,并在统一的智能监控系统中,讨论关键科学问题和解决方案。

关键词：基础设施安全智能感知智能监控系统智能监测系统时空数据地下空间利用时空大数据城市布局