检索结果-内蒙古大学图书馆

29th International Conference on Computational Linguistics, COLING 2022

作者： Gui, Xiangyu Zhao, Feng Jin, Langjunqing Jin, Hai National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab School of Computer Science and Technology Huazhong University of Science and Technology China

Knowledge representation learning is a key step required for link prediction tasks with knowledge graphs (KGs). During the learning process, the semantics of each entity are embedded by a vector or a point in a feature space. The distance between these points is a measure of semantic similarity. However, in a KG, while two entities may have similar semantics in some relations, they have different semantics in others. It is ambiguous to assign a fixed distance to depict the variant semantic similarity of entities. To alleviate the semantic ambiguity in KGs, we design a new embedding approach named OpticE, which is derived from the well-known physical phenomenon of optical interference. It is a lightweight and relation-adaptive model based on coherence theory, in which each entity’s semantics vary automatically regarding different relations. In addition, a unique negative sampling method is proposed to combine the multimapping properties and self-adversarial learning during the training process. The experimental results obtained on practical KG benchmarks show that the OpticE model, with elegant structures, can compete with existing link prediction methods. © 2022 Proceedings - International Conference on Computational Linguistics, COLING. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Refuse Whenever You Feel Unsafe: IMPROVING SAFETY IN LLMS VIA DECOUPLED REFUSAL TRAINING

arXiv

引用

arXiv 2024年

作者： Yuan, Youliang Jiao, Wenxiang Wang, Wenxuan Huang, Jen-Tse Xu, Jiahao Liang, Tian He, Pinjia Tu, Zhaopeng School of Data Science The Chinese University of Hong Kong Shenzhen China Tencent AI Lab China The Chinese University of Hong Kong Hong Kong Shenzhen Research Institute of Big Data China

This study addresses a critical gap in safety tuning practices for Large Language Models (LLMs) by identifying and tackling a refusal position bias within safety tuning data, which compromises the models’ ability to appropriately refuse generating unsafe content. We introduce a novel approach, Decoupled Refusal Training (DeRTa), designed to empower LLMs to refuse compliance to harmful prompts at any response position, significantly enhancing their safety capabilities. DeRTa incorporates two novel components: (1) Maximum Likelihood Estimation (MLE) with Harmful Response Prefix, which trains models to recognize and avoid unsafe content by appending a segment of harmful response to the beginning of a safe response, and (2) Reinforced Transition Optimization (RTO), which equips models with the ability to transition from potential harm to safety refusal consistently throughout the harmful response sequence. Our empirical evaluation, conducted using LLaMA3 and Mistral model families across six attack scenarios, demonstrates that our method not only improves model safety without compromising performance but also surpasses well-known models such as GPT-4 in defending against attacks. Importantly, our approach successfully defends recent advanced attack methods (e.g., CodeAttack Ren et al. (2024)) that have jailbroken GPT-4 and LLaMA3-70B-Instruct. Copyright © 2024, The Authors. All rights reserved.

关键词： Maximum likelihood estimation

来源：评论

学校读者我要写书评

暂无评论

Real is not True: Backdoor Attacks Against Deepfake Detection

Real is not True: Backdoor Attacks Against Deepfake Detectio...

引用

International Conference on big data and Information Analytics (bigDIA)

作者： Hong Sun Ziqiang Li Lei Liu Bin Li University of Science and Technology of China Hefei China Laboratory for Big Data and Decision Zhejiang Lab Hangzhou

The proliferation of malicious deepfake applications has ignited substantial public apprehension, casting a shadow of doubt upon the integrity of digital media. Despite the development of proficient deepfake detection mechanisms, they persistently demonstrate pronounced vulnerability to an array of attacks. It is noteworthy that the pre-existing repertoire of attacks predominantly comprises adversarial example attack, predominantly manifesting during the testing phase. In the present study, we introduce a pioneering paradigm denominated as "Bad-Deepfake," which represents a novel foray into the realm of backdoor attacks levied against deepfake detectors. Our approach hinges upon the strategic manipulation of a delimited subset of the training data, enabling us to wield disproportionate influence over the operational characteristics of a trained model. This manipulation leverages inherent frailties inherent to deepfake detectors, affording us the capacity to engineer triggers and judiciously select the most efficacious samples for the construction of the poisoned set. Through the synergistic amalgamation of these sophisticated techniques, we achieve an remarkable performance—a 100% attack success rate (ASR) against extensively employed deepfake detectors.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Sustainable Self-evolution Adversarial Training 24

Sustainable Self-evolution Adversarial Training

引用

32nd ACM International Conference on Multimedia, MM 2024

作者： Wang, Wenxuan Wang, Chenglei Qi, Huihui Ye, Menghao Qian, Xuelin Wang, Peng Zhang, Yanning School of Computer Science Northwestern Polytechnical University China Natl. Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology Shaanxi Xi'an China School of Automation Northwestern Polytechnical University The BRain and Artificial INtelligence Lab Shaanxi Xi'an China

ISBN: (纸本)9798400706868

With the wide application of deep neural network models in various computer vision tasks, there has been a proliferation of adversarial example generation strategies aimed at deeply exploring model security. However, existing adversarial training defense models, which rely on single or limited types of attacks under a one-time learning process, struggle to adapt to the dynamic and evolving nature of attack methods. Therefore, to achieve defense performance improvements for models in long-term applications, we propose a novel Sustainable Self-Evolution Adversarial Training (SSEAT) framework. Specifically, we introduce a continual adversarial defense pipeline to realize learning from various kinds of adversarial examples across multiple stages. Additionally, to address the issue of model catastrophic forgetting caused by continual learning from ongoing novel attacks, we propose an adversarial data replay module to better select more diverse and key relearning data. Furthermore, we design a consistency regularization strategy to encourage current defense models to learn more from previously trained ones, guiding them to retain more past knowledge and maintain accuracy on clean samples. Extensive experiments have been conducted to verify the efficacy of the proposed SSEAT defense method, which demonstrates superior defense performance and classification accuracy compared to competitors. © 2024 ACM.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

AFaVS: Accurate Yet Fast Version Switching for Graph Processing Systems 39

AFaVS: Accurate Yet Fast Version Switching for Graph Process...

引用

39th IEEE International Conference on data Engineering, ICDE 2023

作者： Zheng, Long Ye, Xiangyu Liu, Haifeng Wang, Qinggang Huang, Yu Gui, Chuangyi Yao, Pengcheng Liao, Xiaofei Jin, Hai Xue, Jingling Huazhong University of Science and Technology National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Laboratory Wuhan430074 China Zhejiang Lab Hangzhou311121 China Unsw School of Computer Science and Engineering Sydney Australia

ISBN: (纸本)9798350322279

Multi-version graph processing has been widely used to solve many real-world problems. The process of the multi-version graph processing typically includes: (1) a history graph version switching at a specific time and (2) graph processing on this history graph. Existing multi-version graph systems assume ideally that every request for a particular graph version at a particular time will have a corresponding snapshot available. However, in most cases, this is not true. Then existing solutions usually have to settle with an "approximating"version as a substitute, leading to unexpected results for the underlying graph algorithm and thus reducing the practicality of a multi-version graph system for many application scenarios *** this paper, we observe that only a few graph updates have a great impact on the final results. We therefore present AFaVS, a novel multi-version graph system that can improve accuracy effectively in both time- and memory-efficient manners. The cornerstone of AFaVS lies in a novel concept "value"that characterizes the importance of graph updates. AFaVS proposes differential management of updates based on their values and achieves higher accuracy while preserving processing and memory efficiency. AFaVS is also equipped with value-guided version switching and locality-aware optimizations to boost its overall efficiency. Our results on a variety of real-world datasets show that AFaVS outperforms four state-of-the-art multi-version graph systems by 74.35%~95.72% in terms of accuracy improvement and 57.03%~90.44% in terms of memory reduction while introducing less than 2.96% extra computing time. We have deployed AFaVS in a disaster recovery system on the production cluster of Alibaba, achieving 78.8%~90.1% fewer error rates than advanced systems at a comparable efficiency. © 2023 IEEE.

关键词： Efficiency

来源：评论

学校读者我要写书评

暂无评论

Efficient Trigger Word Insertion

Efficient Trigger Word Insertion

引用

International Conference on big data and Information Analytics (bigDIA)

作者： Yueqi Zeng Ziqiang Li Pengfei Xia Lei Liu Bin Li University of Science and Technology of China Hefei China Laboratory for Big Data and Decision Zhejiang Lab Hangzhou

With the rapid advancements in the natural language processing (NLP) domain in recent years, the emergence of backdoor attacks presents substantial threats to deep neural network models. However, prior research has often overlooked the influence of the poisoning rate. This paper aims to address this gap by prioritizing the reduction of poisoned samples while still attaining a comparable Attack Success Rate (ASR) in the context of text backdoor attacks. Our primary focus revolves around introducing an efficient strategy for trigger word insertion, encompassing both trigger word optimization and poisoned sample selection. To achieve our objectives, extensive experiments were conducted across diverse datasets and models, showcasing the significant enhancements brought forth by our proposed methodology in the realm of text classification tasks. Remarkable outcomes include an ASR surpassing 90%, utilizing a mere 10 poisoned samples in the dirty-label setting, and delivering compelling performance with only 1.5% of the training data in the clean-label setting.

关键词：

来源：评论

学校读者我要写书评

暂无评论

On the Performance of Deep Learning Models for Time Series Classification in Streaming 15th

On the Performance of Deep Learning Models for Time Series C...

引用

15th International Conference on Soft Computing Models in Industrial and Environmental Applications, SOCO 2020

作者： Lara-Benítez, Pedro Carranza-García, Manuel Martínez-Álvarez, Francisco Riquelme, José C. Division of Computer Science University of Sevilla Seville41012 Spain Data Science & Big Data Lab Pablo de Olavide University Seville41013 Spain

ISBN: (纸本)9783030578015

Processing data streams arriving at high speed requires the development of models that can provide fast and accurate predictions. Although deep neural networks are the state-of-the-art for many machine learning tasks, their performance in real-time data streaming scenarios is a research area that has not yet been fully addressed. Nevertheless, there have been recent efforts to adapt complex deep learning models for streaming tasks by reducing their processing rate. The design of the asynchronous dual-pipeline deep learning framework allows to predict over incoming instances and update the model simultaneously using two separate layers. The aim of this work is to assess the performance of different types of deep architectures for data streaming classification using this framework. We evaluate models such as multi-layer perceptrons, recurrent, convolutional and temporal convolutional neural networks over several time-series datasets that are simulated as streams. The obtained results indicate that convolutional architectures achieve a higher performance in terms of accuracy and efficiency. © 2021, The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

AN ENSEMBLE MODEL FOR DISTORTED IMAGES IN REAL SCENARIOS

arXiv

引用

arXiv 2023年

作者： Ji, Boyuan Huang, Jianchang Huang, Wenzhuo He, Shuke Zhejiang University of Science and Technology Edge Intelligence Security Lab School of Big Data Science Zhejiang Hangzhou China

Image acquisition conditions and environments can significantly affect high-level tasks in computer vision, and the performance of most computer vision algorithms will be limited when trained on distortion-free datasets. Even with updates in hardware such as sensors and deep learning methods, it will still not work in the face of variable conditions in real-world applications. In this paper, we apply the object detector YOLOv7 to detect distorted images from the dataset CDCOCO. Through carefully designed optimizations including data enhancement, detection box ensemble, denoiser ensemble, super-resolution models, and transfer learning, our model achieves excellent performance on the CDCOCO test set. Our denoising detection model can denoise and repair distorted images, making the model useful in a variety of real-world scenarios and environments. Copyright © 2023, The Authors. All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

HIERARCHICAL EMOTION PREDICTION AND CONTROL IN TEXT-TO-SPEECH SYNTHESIS

arXiv

引用

arXiv 2024年

作者： Inoue, Sho Zhou, Kun Wang, Shuai Li, Haizhou School of Data Science The Chinese University of Hong Kong Shenzhen [CUHK-Shenzhen China Shenzhen Research Institute of Big Data Shenzhen China Speech Lab of DAMO Academy Alibaba Group Singapore

It remains a challenge to effectively control the emotion rendering in text-to-speech (TTS) synthesis. Prior studies have primarily focused on learning a global prosodic representation at the utterance level, which strongly correlates with linguistic prosody. Our goal is to construct a hierarchical emotion distribution (ED) that effectively encapsulates intensity variations of emotions at various levels of granularity, encompassing phonemes, words, and utterances. During TTS training, the hierarchical ED is extracted from the ground-truth audio and guides the predictor to establish a connection between emotional and linguistic prosody. At run-time inference, the TTS model generates emotional speech and, at the same time, provides quantitative control of emotion over the speech constituents. Both objective and subjective evaluations validate the effectiveness of the proposed framework in terms of emotion prediction and control. © 2024, CC BY.

关键词： Speech synthesis

来源：评论

学校读者我要写书评

暂无评论

Hierarchical Emotion Prediction and Control in Text-to-Speech Synthesis

Hierarchical Emotion Prediction and Control in Text-to-Speec...

引用

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

作者： Sho Inoue Kun Zhou Shuai Wang Haizhou Li School of Data Science The Chinese University of Hong Kong Shenzhen (CUHK-Shenzhen) China Shenzhen Research Institute of Big Data Shenzhen China Alibaba Group Speech Lab of DAMO Academy Singapore

关键词：

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：