Attack Success Rate

Attack success rate (ASR) quantifies the effectiveness of adversarial attacks against machine learning models, focusing on compromising their security and reliability. Current research investigates ASR across various model types, including large language models (LLMs), federated learning systems, and text-to-image generators, employing diverse attack methods like gradient-based optimization, backdoor insertion, and prompt engineering. Understanding and improving ASR is crucial for developing robust and secure AI systems, impacting both the theoretical foundations of machine learning and the practical deployment of AI in sensitive applications. The field is actively exploring both improved attack strategies and more effective defenses.

Papers

May 31, 2024

Enhancing Jailbreak Attack Against Large Language Models through Silent Tokens
Jiahao Yu, Haozheng Luo, Jerry Yao-Chieh Hu, Wenbo Guo, Han Liu, Xinyu Xing
Language Model Jailbreak Attack Attack Success Rate

May 29, 2024

DiveR-CT: Diversity-enhanced Red Teaming with Relaxing Constraints
Andrew Zhao, Quentin Xu, Matthieu Lin, Shenzhi Wang, Yong-jin Liu, Zilong Zheng, Gao Huang
Red Teaming Attack Success Rate DIVeR Identification Semantic Diversity Semantic Reward Constraint Relaxation

May 24, 2024

TrojanForge: Generating Adversarial Hardware Trojan Examples with Reinforcement Learning
Amin Sarihi, Peter Jamieson, Ahmad Patooghy, Abdel-Hameed A. Badawy
Reinforcement Learning Generative Adversarial Network Adversarial Example Attack Success Rate Hardware Trojan Adversarial Malware

May 19, 2024

An Invisible Backdoor Attack Based On Semantic Feature
Yangming Chen
Deep Neural Network Backdoor Attack Backdoor Defense Semantic Feature Attack Success Rate Invisible Backdoor Attack

May 10, 2024

Improving Transferable Targeted Adversarial Attack via Normalized Logit Calibration and Truncated Feature Mixing
Juanjuan Weng, Zhiming Luo, Shaozi Li
Task Transferability Adversarial Sample Attack Success Rate Transferable Adversarial Attack Logit Calibration Feature Mixing Source Training

April 26, 2024

Beyond Traditional Threats: A Persistent Backdoor Attack on Federated Learning
Tao Liu, Yuhang Zhang, Zhu Feng, Zhiqin Yang, Chen Xu, Dapeng Man, Wu Yang
Backdoor Attack Attack Success Rate Backdoor Effect

April 22, 2024

Typos that Broke the RAG's Back: Genetic Attack on RAG Pipeline by Simulating Documents in the Wild via Low-level Perturbations
Sukmin Cho, Soyeong Jeong, Jeongyeon Seo, Taeho Hwang, Jong C. Park
Native Robustness Retrieval Augmented Generation Attack Success Rate Recent Large Language Model Kg Rag RAG Based Document Generation Typo Squatting

April 18, 2024

Uncovering Safety Risks of Large Language Models through Concept Activation Vector
Zhihao Xu, Ruixuan Huang, Changyu Chen, Xiting Wang
Attack Success Rate Attack Method Concept Activation Vector Invisible Attack

April 15, 2024

On the Efficiency of Privacy Attacks in Federated Learning
Nawrin Tabassum, Ka-Ho Chow, Xuyu Wang, Wenbin Zhang, Yanzhao Wu
High Efficiency Privacy Attack Attack Success Rate Gradient Leakage Attack

April 3, 2024

Exploring Backdoor Vulnerabilities of Chat Models
Yunzhuo Hao, Wenkai Yang, Yankai Lin
Backdoor Attack Backdoor Trigger Attack Success Rate Chat Model

March 18, 2024

Impart: An Imperceptible and Effective Label-Specific Backdoor Attack
Jingke Zhao, Zan Wang, Yongwei Wang, Lanjun Wang
Backdoor Attack Attack Success Rate Clean Label Backdoor Attack Imperceptible Pattern Label Attack

February 23, 2024

Fast Adversarial Attacks on Language Models In One GPU Minute
Vinu Sankar Sadasivan, Shoumik Saha, Gaurang Sriramanan, Priyatham Kattakinda, Atoosa Chegini, Soheil Feizi
Language Model Membership Inference Attack Adversarial Prompt Attack Success Rate Adversarial Search

February 12, 2024

OrderBkd: Textual backdoor attack through repositioning
Irina Alekseevskaia, Konstantin Arkhipenko
Attack Success Rate Textual Backdoor Attack

January 11, 2024

Universal Vulnerabilities in Large Language Models: Backdoor Attacks for In-context Learning
Shuai Zhao, Meihuizi Jia, Luu Anh Tuan, Fengjun Pan, Jinming Wen
Large Language Model Language Model Context Learning Backdoor Attack Attack Success Rate Inherent Vulnerability

December 22, 2023

Asymmetric Bias in Text-to-Image Generation with Adversarial Attacks
Haz Sameen Shahgir, Xianghao Kong, Greg Ver Steeg, Yue Dong
Adversarial Attack Text to Image Generation Attack Success Rate Adversarial Suffix

December 19, 2023

Bypassing the Safety Training of Open-Source LLMs with Priming Attacks
Jason Vega, Isha Chaudhary, Changming Xu, Gagandeep Singh
Attack Success Rate Open Source LLM Text Attack

November 23, 2023

November 15, 2023

Defending Large Language Models Against Jailbreaking Attacks Through Goal Prioritization
Zhexin Zhang, Junxiao Yang, Pei Ke, Fei Mi, Hongning Wang, Minlie Huang
Jailbreak Attack New Attack Attack Success Rate Priority Based Inference Stage

November 14, 2023

$DA^3$: A Distribution-Aware Adversarial Attack against Language Models
Yibo Wang, Xiangjue Dong, James Caverlee, Philip S. Yu
Language Model Adversarial Attack Adversarial Example Attack Success Rate LoRA Based