Algorithms, 2022 Universal adversarial attacks, which hinder most deep neural network (DNN) tasks using only a single perturbation called universal adversarial perturbation (UAP), are a realistic security threat to the practical application of a DNN for medical imaging ...
Kazuhiro Takemoto
exaly +3 more sources
Back in Black: A Comparative Evaluation of Recent State-Of-The-Art Black-Box Attacks
IEEE Access, 2022 The field of adversarial machine learning has experienced a near exponential growth in the amount of papers being produced since 2018. This massive information output has yet to be properly processed and categorized.
Kaleel Mahmood +2 more
exaly +3 more sources
DeDiAttack: Enhancing Transferability of Unrestricted Adversarial Examples via Deformation-Constrained Diffusion [PDF]
SensorsDNNs are highly vulnerable to adversarial examples (AEs). To achieve high transferability, traditional AEs often introduce unnatural artifacts that are easily perceptible to the human eye.
Bin Qu, Anjie Peng, Shijie Zhao
doaj +2 more sources
AB jailbreaking - a novel hybrid framework for exploitation of adversarial vulnerabilities in LLMs [PDF]
Scientific ReportsLarge language models (LLMs) have advanced rapidly but remain vulnerable to adversarial “jailbreaking” attacks that elicit harmful or disallowed outputs.
Abrar Ahmad +3 more
doaj +2 more sources
The strength of Nesterov's accelerated gradient in boosting transferability of stealthy adversarial attacks. [PDF]
PLoS ONEDeep neural networks have been shown to be highly vulnerable to adversarial examples-inputs crafted to mislead models by adding subtle, human-imperceptible perturbations. Transferability and stealthiness are two crucial metrics for evaluating adversarial
Chen Lin, Sheng Long
doaj +2 more sources
RIB-Guard: A Risk-Aware Information Bottleneck Defense for Black-Box Large Language Models [PDF]
EntropyLarge language models (LLMs) remain vulnerable to jailbreak attacks, especially in black-box settings where target-model gradients and internal tokenization are inaccessible.
Muen Cai, Yuan Shen, Xiong Luo, Jian Hu
doaj +2 more sources
Momentum-Based Adversarial Attacks and Multi-Level Denoising Defenses in Deep Learning-Based Wind Power Forecasting [PDF]
SensorsDeep learning (DL) techniques have significantly advanced wind power forecasting by enhancing accuracy. However, these DL models are vulnerable to adversarial attacks, which can lead to severely inaccurate forecasts.
Yangming Min +4 more
doaj +2 more sources
Black-box attacks and defense for DNN-based power quality classification in smart grid
Energy Reports, 2022 Machine learning (ML) models are widely used in smart grid, but they are vulnerable to adversarial examples that are maliciously crafted using the input data. Therefore, the use of these models in smart grid can cause significant damage.
Aiping Pang +2 more
exaly +3 more sources
Black Box Adversarial Attack Starting Point Promotion Method Based on Mobility Between Models [PDF]
Jisuanji gongcheng, 2021 In order to efficiently find the adversarial samples under the decision-based black box attacks, a method using the mobility between models is proposed to enhance the adversarial starting point. The mobility is used to circularly superimpose interference
CHEN Xiaonan, HU Jianmin, ZHANG Benjun, CHEN Ailing
doaj +1 more source
Adversarial Examples Generation Method Based on Image Color Random Transformation [PDF]
Jisuanji kexue, 2023 Although deep neural networks(DNNs) have good performance in most classification tasks,they are vulnerable to adversarial examples,making the security of DNNs questionable.Research designs to generate strongly aggressive adversarial examples can help ...
BAI Zhixu, WANG Hengjun, GUO Kexiang
doaj +1 more source