Attack Efficacy

Attack efficacy research focuses on evaluating the effectiveness of various methods to compromise machine learning models, encompassing data poisoning, model manipulation, and prompt injection attacks. Current research investigates attack strategies across diverse model architectures, including federated learning, recommender systems, vision-language models, and large language models, often employing techniques like adversarial examples, logits poisoning, and profile injection. Understanding attack efficacy is crucial for improving the robustness and security of machine learning systems, impacting the reliability of AI applications across numerous domains. This necessitates the development of robust evaluation metrics and defense mechanisms.

Papers

October 16, 2024

Long-Tailed Backdoor Attack Using Dynamic Data Augmentation Operations
Lu Pang, Tao Sun, Weimin Lyu, Haibin Ling, Chao Chen
Data Augmentation Backdoor Attack Attack Efficacy

September 8, 2024

Influence-based Attributions can be Manipulated
Chhavi Yadav, Ruihan Wu, Kamalika Chaudhuri
Influence Function Adversarial Scenario Attack Efficacy Influence Analysis

August 22, 2024

Understanding Data Reconstruction Leakage in Federated Learning from a Theoretical Perspective
Zifan Wang, Binghui Zhang, Meng Pang, Yuan Hong, Binghui Wang
Data Reconstruction Attack Attack Efficacy

July 25, 2024

Peak-Controlled Logits Poisoning Attack in Federated Distillation
Yuhan Tang, Aoxu Zhang, Zhiyuan Wu, Bo Gao, Tian Wen, Yuwei Wang, Sheng Sun
Knowledge Distillation Poisoning Attack Federated Distillation Attack Efficacy One Stage Self Distillation

March 5, 2024

ImgTrojan: Jailbreaking Vision-Language Models with ONE Image
Xijia Tao, Shuai Zhong, Lei Li, Qi Liu, Lingpeng Kong
Large Language Model Vision Language Model Jailbreak Attack Attack Efficacy

February 28, 2024

Making Them Ask and Answer: Jailbreaking Large Language Models in Few Queries via Disguise and Reconstruction
Tong Liu, Yingjie Zhang, Zhe Zhao, Yinpeng Dong, Guozhu Meng, Kai Chen
Full State Reconstruction Jailbreak Attack Adversarial Prompt Top Two Answer Reconstruction Attack Digital STEALTH Metric Attack Efficacy

February 14, 2024

Review-Incorporated Model-Agnostic Profile Injection Attacks on Recommender Systems
Shiyi Yang, Lina Yao, Chen Wang, Xiwei Xu, Liming Zhu
Recommender System Novel Attack Attack Framework Attack Efficacy

January 17, 2024

AttackEval: How to Evaluate the Effectiveness of Jailbreak Attacking on Large Language Models
Dong shu, Mingyu Jin, Suiyuan Zhu, Beichen Wang, Zihao Zhou, Chong Zhang, Yongfeng Zhang
Jailbreak Attack Robustness Evaluation Pseudo Victim AttackER Attack Efficacy

May 18, 2023

Black-Box Targeted Reward Poisoning Attack Against Online Deep Reinforcement Learning
Yinglun Xu, Gagandeep Singh
Unknown Dynamic Attack Efficacy Multi Step Attack Reward Poisoning Attack

February 20, 2023

An Incremental Gray-box Physical Adversarial Attack on Neural Network Training
Rabiah Al-qudah, Moayad Aloqaily, Bassem Ouni, Mohsen Guizani, Thierry Lestable
Neural Network Adversarial Attack Gradient Free Attack Efficacy

October 30, 2022

Benchmarking Adversarial Patch Against Aerial Detection
Jiawei Lian, Shaohui Mei, Shun Zhang, Mingyang Ma
Adversarial Example Adversarial Patch Attack Efficacy Aerial Detection

September 9, 2022

The Space of Adversarial Strategies
Ryan Sheatsley, Blaine Hoak, Eric Pauley, Patrick McDaniel
Adversarial Attack Adversarial Example Adversarial Learning Deep Space Attack Efficacy Ensemble Attack

January 9, 2022

Rethink the Evaluation for Attack Strength of Backdoor Attacks in Natural Language Processing
Lingfeng Shen, Haiyun Jiang, Lemao Liu, Shuming Shi
Natural Language Processing Global Evaluation Backdoor Attack Backdoor Trigger Stealthy Backdoor Attack Attack Efficacy