大模型对齐与安全 - arXivDaily 专题

2405.12076 2024-09-09 cs.CR eess.SP 50%

GAN-GRID: A Novel Generative Attack on Smart Grid Stability Prediction

Emad Efatinasab, Alessandro Brighente, Mirco Rampazzo, Nahal Azadi, Mauro Conti

专题命中越狱攻击：safety(abstract)

Journal ref European Symposium on Research in Computer Security (ESORICS 2024), 2024, Lecture Notes in Computer Science, vol 14982

2404.05264 2024-08-13 cs.CR cs.CV 50%

Unbridled Icarus: A Survey of the Potential Perils of Image Inputs in Multimodal Large Language Model Security

Yihe Fan, Yuxin Cao, Ziyu Zhao, Ziyao Liu, Shaofeng Li

专题命中越狱攻击：trustworthy(abstract)

Comments 8 pages, 1 figure. Accepted to 2024 IEEE International Conference on Systems, Man, and Cybernetics

2407.13111 2024-07-19 cs.MM cs.CV 50%

PG-Attack: A Precision-Guided Adversarial Attack Framework Against Vision Foundation Models for Autonomous Driving

Jiyuan Fu, Zhaoyu Chen, Kaixun Jiang, Haijing Guo, Shuyong Gao, Wenqiang Zhang

专题命中越狱攻击：safety(abstract)

Comments First-Place in the CVPR 2024 Workshop Challenge: Black-box Adversarial Attacks on Vision Foundation Models

2407.09292 2024-07-18 cs.CR 50%

Counterfactual Explainable Incremental Prompt Attack Analysis on Large Language Models

Dong Shu, Mingyu Jin, Tianle Chen, Chong Zhang, Yongfeng Zhang

专题命中越狱攻击：safety(abstract)

Comments 23 pages, 6 figures

2407.06948 2024-07-10 eess.SY cs.SY 50%

Detection-Triggered Recursive Impact Mitigation against Secondary False Data Injection Attacks in Microgrids

Mengxiang Liu, Xin Zhang, Rui Zhang, Zhuoran Zhou, Zhenyong Zhang, Ruilong Deng

专题命中越狱攻击：trustworthy(abstract)

Comments Submitted to IEEE Transactions on Smart Grid

2302.02213 2024-07-09 cs.CV 50%

CosPGD: an efficient white-box adversarial attack for pixel-wise prediction tasks

Shashank Agnihotri, Steffen Jung, Margret Keuper

专题命中越狱攻击：alignment(abstract)

Comments Accepted at 41st International Conference on Machine Learning (ICML), 2024

2405.19668 2024-05-31 cs.CV 50%

AutoBreach: Universal and Adaptive Jailbreaking with Efficient Wordplay-Guided Optimization

Jiawei Chen, Xiao Yang, Zhengwei Fang, Yu Tian, Yinpeng Dong, Zhaoxia Yin, Hang Su

专题命中越狱攻击：jailbreak(abstract)

Comments Under review

2405.14169 2024-05-24 cs.CV 50%

Towards Transferable Attacks Against Vision-LLMs in Autonomous Driving with Typography

Nhat Chung, Sensen Gao, Tuan-Anh Vu, Jie Zhang, Aishan Liu, Yun Lin, Jin Song Dong, Qing Guo

专题命中越狱攻击：safety(abstract)

Comments 12 pages, 5 tables, 5 figures, work in progress

2404.12916 2024-04-23 cs.CR 50%

Physical Backdoor Attack can Jeopardize Driving with Vision-Large-Language Models

Zhenyang Ni, Rui Ye, Yuxi Wei, Zhen Xiang, Yanfeng Wang, Siheng Chen

专题命中越狱攻击：safety(abstract)

2308.10299 2024-03-26 cs.CV eess.IV 50%

Boosting Adversarial Transferability by Block Shuffle and Rotation

Kunyu Wang, Xuanran He, Wenxuan Wang, Xiaosen Wang

专题命中越狱攻击：trustworthy(abstract)

Comments Accepted by CVPR 2024

2403.12693 2024-03-20 cs.CV 50%

As Firm As Their Foundations: Can open-sourced foundation models be used to create adversarial examples for downstream tasks?

Anjun Hu, Jindong Gu, Francesco Pinto, Konstantinos Kamnitsas, Philip Torr

专题命中越狱攻击：safety(abstract)

2403.08701 2024-03-20 cs.CR 50%

Review of Generative AI Methods in Cybersecurity

Yagmur Yigit, William J Buchanan, Madjid G Tehrani, Leandros Maglaras

专题命中越狱攻击：prompt injection(abstract)

Comments 40 pages

2403.07654 2024-03-13 cs.IR 50%

Analyzing Adversarial Attacks on Sequence-to-Sequence Relevance Models

Andrew Parry, Maik Fröbe, Sean MacAvaney, Martin Potthast, Matthias Hagen

专题命中越狱攻击：prompt injection(abstract)

Comments 13 pages, 3 figures, Accepted at ECIR 2024 as a Full Paper

2402.15617 2024-02-27 cs.CR cs.SY eess.SY 50%

Reinforcement Learning-Based Approaches for Enhancing Security and Resilience in Smart Control: A Survey on Attack and Defense Methods

Zheyu Zhang

专题命中越狱攻击：safety(abstract)

2401.00994 2024-01-03 cs.CR 50%

Detection and Defense Against Prominent Attacks on Preconditioned LLM-Integrated Virtual Assistants

Chun Fai Chan, Daniel Wankit Yip, Aysan Esmradi

专题命中越狱攻击：safety(abstract)

Comments Accepted to be published in the Proceedings of the 10th IEEE CSDE 2023, the Asia-Pacific Conference on Computer Science and Data Engineering 2023

2308.15736 2023-12-29 cs.CR cs.SY eess.SY 50%

Vulnerability of Machine Learning Approaches Applied in IoT-based Smart Grid: A Review

Zhenyong Zhang, Mengxiang Liu, Mingyang Sun, Ruilong Deng, Peng Cheng, Dusit Niyato, Mo-Yuen Chow, Jiming Chen

专题命中越狱攻击：safety(abstract)

2312.04403 2023-12-08 cs.CV 50%

OT-Attack: Enhancing Adversarial Transferability of Vision-Language Models via Optimal Transport Optimization

Dongchen Han, Xiaojun Jia, Yang Bai, Jindong Gu, Yang Liu, Xiaochun Cao

专题命中越狱攻击：alignment(abstract)

2311.18350 2023-12-01 cs.DC 50%

Unveiling Backdoor Risks Brought by Foundation Models in Heterogeneous Federated Learning

Xi Li, Chen Wu, Jiaqi Wang

专题命中越狱攻击：safety(abstract)

Comments Jiaqi Wang is the corresponding author. arXiv admin note: text overlap with arXiv:2311.00144

2306.12685 2023-11-22 cs.CV 50%

Rethinking the Backward Propagation for Adversarial Transferability

Xiaosen Wang, Kangheng Tong, Kun He

专题命中越狱攻击：trustworthy(abstract)

Comments Accepted by NeurIPS 2023

2310.13345 2023-10-23 cs.CR 50%

An LLM can Fool Itself: A Prompt-Based Adversarial Attack

Xilie Xu, Keyi Kong, Ning Liu, Lizhen Cui, Di Wang, Jingfeng Zhang, Mohan Kankanhalli

专题命中越狱攻击：safety(abstract)

2304.10136 2023-09-18 cs.CV 50%

Diversifying the High-level Features for better Adversarial Transferability

Zhiyuan Wang, Zeliang Zhang, Siyuan Liang, Xiaosen Wang

专题命中越狱攻击：trustworthy(abstract)

Comments Accepted by BMVC 2023 (Oral)

2308.15674 2023-08-31 cs.CR 50%

Predict And Prevent DDOS Attacks Using Machine Learning and Statistical Algorithms

Azadeh Golduzian

专题命中越狱攻击：trustworthy(abstract)

2308.11894 2023-08-24 cs.CR cs.CV 50%

Does Physical Adversarial Example Really Matter to Autonomous Driving? Towards System-Level Effect of Adversarial Object Evasion Attack

Ningfei Wang, Yunpeng Luo, Takami Sato, Kaidi Xu, Qi Alfred Chen

专题命中越狱攻击：safety(abstract)

Comments Accepted by ICCV 2023

2308.07673 2023-08-16 cs.CV stat.CO stat.ML 50%

A Review of Adversarial Attacks in Computer Vision

Yutong Zhang, Yao Li, Yin Li, Zhichang Guo

专题命中越狱攻击：safety(abstract)

2304.08224 2023-07-18 cs.CR 50%

Uncharted Territory: Energy Attacks in the Battery-less Internet of Things

Luca Mottola, Arslan Hameed, Thiemo Voigt

专题命中越狱攻击：safety(abstract)

2212.09360 2023-07-14 cs.CV 50%

AI Security for Geoscience and Remote Sensing: Challenges and Future Trends

Yonghao Xu, Tao Bai, Weikang Yu, Shizhen Chang, Peter M. Atkinson, Pedram Ghamisi

专题命中越狱攻击：safety(abstract)

Journal ref IEEE Geoscience and Remote Sensing Magazine, Volume 11, Issue 2, Pages 60-85, 2023

2306.14782 2023-06-27 cs.CR 50%

On the Resilience of Machine Learning-Based IDS for Automotive Networks

Ivo Zenden, Han Wang, Alfonso Iacovazzi, Arash Vahidi, Rolf Blom, Shahid Raza

专题命中越狱攻击：safety(abstract)

Journal ref 2023 IEEE Vehicular Networking Conference (VNC), Istanbul, Turkiye, 2023, pp. 239-246

2211.01112 2022-11-29 cs.CR 50%

Adversarial Attack on Radar-based Environment Perception Systems

Amira Guesmi, Ihsen Alouani

专题命中越狱攻击：safety(abstract)

2206.10110 2022-06-22 cs.SE 50%

ProML: A Decentralised Platform for Provenance Management of Machine Learning Software Systems

Nguyen Khoi Tran, Bushra Sabir, M. Ali Babar, Nini Cui, Mehran Abolhasan, Justin Lipman

专题命中越狱攻击：safety(abstract)

Comments Accepted as full paper in ECSA 2022 conference. To be presented

2012.02675 2021-08-25 cs.CR 50%

Resilience-by-design in Adaptive Multi-Agent Traffic Control Systems

Ranwa Al Mallah, Talal Halabi, Bilal Farooq

专题命中越狱攻击：safety(abstract)

1. 越狱攻击 1721 篇

GAN-GRID: A Novel Generative Attack on Smart Grid Stability Prediction

Unbridled Icarus: A Survey of the Potential Perils of Image Inputs in Multimodal Large Language Model Security

PG-Attack: A Precision-Guided Adversarial Attack Framework Against Vision Foundation Models for Autonomous Driving

Counterfactual Explainable Incremental Prompt Attack Analysis on Large Language Models

Detection-Triggered Recursive Impact Mitigation against Secondary False Data Injection Attacks in Microgrids

CosPGD: an efficient white-box adversarial attack for pixel-wise prediction tasks

AutoBreach: Universal and Adaptive Jailbreaking with Efficient Wordplay-Guided Optimization

Towards Transferable Attacks Against Vision-LLMs in Autonomous Driving with Typography

Physical Backdoor Attack can Jeopardize Driving with Vision-Large-Language Models

Boosting Adversarial Transferability by Block Shuffle and Rotation

As Firm As Their Foundations: Can open-sourced foundation models be used to create adversarial examples for downstream tasks?

Review of Generative AI Methods in Cybersecurity

Analyzing Adversarial Attacks on Sequence-to-Sequence Relevance Models

Reinforcement Learning-Based Approaches for Enhancing Security and Resilience in Smart Control: A Survey on Attack and Defense Methods

Detection and Defense Against Prominent Attacks on Preconditioned LLM-Integrated Virtual Assistants

Vulnerability of Machine Learning Approaches Applied in IoT-based Smart Grid: A Review

OT-Attack: Enhancing Adversarial Transferability of Vision-Language Models via Optimal Transport Optimization

Unveiling Backdoor Risks Brought by Foundation Models in Heterogeneous Federated Learning

Rethinking the Backward Propagation for Adversarial Transferability

An LLM can Fool Itself: A Prompt-Based Adversarial Attack

Diversifying the High-level Features for better Adversarial Transferability

Predict And Prevent DDOS Attacks Using Machine Learning and Statistical Algorithms

Does Physical Adversarial Example Really Matter to Autonomous Driving? Towards System-Level Effect of Adversarial Object Evasion Attack

A Review of Adversarial Attacks in Computer Vision

Uncharted Territory: Energy Attacks in the Battery-less Internet of Things

AI Security for Geoscience and Remote Sensing: Challenges and Future Trends

On the Resilience of Machine Learning-Based IDS for Automotive Networks

Adversarial Attack on Radar-based Environment Perception Systems

ProML: A Decentralised Platform for Provenance Management of Machine Learning Software Systems

Resilience-by-design in Adaptive Multi-Agent Traffic Control Systems