English
Related papers

Related papers: Pre-trained Adversarial Perturbations

200 papers

Prompt-based learning paradigm bridges the gap between pre-training and fine-tuning, and works effectively under the few-shot setting. However, we find that this learning paradigm inherits the vulnerability from the pre-training stage,…

Computation and Language · Computer Science 2022-04-12 Lei Xu , Yangyi Chen , Ganqu Cui , Hongcheng Gao , Zhiyuan Liu

In this paper, we propose novel generative models for creating adversarial examples, slightly perturbed images resembling natural images but maliciously crafted to fool pre-trained models. We present trainable deep neural networks for…

Computer Vision and Pattern Recognition · Computer Science 2018-07-09 Omid Poursaeed , Isay Katsman , Bicheng Gao , Serge Belongie

Most existing works focus on improving robustness against adversarial attacks bounded by a single $l_p$ norm using adversarial training (AT). However, these AT models' multiple-norm robustness (union accuracy) is still low, which is crucial…

Machine Learning · Computer Science 2024-09-24 Enyi Jiang , Gagandeep Singh

Improving the effectiveness and efficiency of large language models (LLMs) simultaneously is a critical yet challenging research goal. In this paper, we find that low-rank pre-training, normally considered as efficient methods that will…

Computation and Language · Computer Science 2024-11-05 Xingtai Lv , Ning Ding , Kaiyan Zhang , Ermo Hua , Ganqu Cui , Bowen Zhou

Latent diffusion models have recently demonstrated superior capabilities in many downstream image synthesis tasks. However, customization of latent diffusion models using unauthorized data can severely compromise the privacy and…

Computer Vision and Pattern Recognition · Computer Science 2025-06-17 Sen Peng , Mingyue Wang , Jianfei He , Jijia Yang , Xiaohua Jia

Large Vision-Language Models (VLMs) have demonstrated remarkable performance across multimodal tasks by integrating vision encoders with large language models (LLMs). However, these models remain vulnerable to adversarial attacks. Among…

Computer Vision and Pattern Recognition · Computer Science 2024-12-20 Hee-Seon Kim , Minbeom Kim , Changick Kim

Pretrained language models (PLMs) perform poorly under adversarial attacks. To improve the adversarial robustness, adversarial data augmentation (ADA) has been widely adopted to cover more search space of adversarial attacks by adding…

Computation and Language · Computer Science 2021-06-08 Chenglei Si , Zhengyan Zhang , Fanchao Qi , Zhiyuan Liu , Yasheng Wang , Qun Liu , Maosong Sun

Pre-trained models (PTMs) have been widely used in various downstream tasks. The parameters of PTMs are distributed on the Internet and may suffer backdoor attacks. In this work, we demonstrate the universal vulnerability of PTMs, where…

Computation and Language · Computer Science 2023-10-23 Zhengyan Zhang , Guangxuan Xiao , Yongwei Li , Tian Lv , Fanchao Qi , Zhiyuan Liu , Yasheng Wang , Xin Jiang , Maosong Sun

Open-weight AI systems offer unique benefits, including enhanced transparency, open research, and decentralized access. However, they are vulnerable to tampering attacks which can efficiently elicit harmful behaviors by modifying weights or…

Attacking deep learning based biometric systems has drawn more and more attention with the wide deployment of fingerprint/face/speaker recognition systems, given the fact that the neural networks are vulnerable to the adversarial examples,…

Audio and Speech Processing · Electrical Eng. & Systems 2020-04-08 Jiguo Li , Xinfeng Zhang , Chuanmin Jia , Jizheng Xu , Li Zhang , Yue Wang , Siwei Ma , Wen Gao

Backdoor attacks, which maliciously control a well-trained model's outputs of the instances with specific triggers, are recently shown to be serious threats to the safety of reusing deep neural networks (DNNs). In this work, we propose an…

Computation and Language · Computer Science 2021-10-18 Wenkai Yang , Yankai Lin , Peng Li , Jie Zhou , Xu Sun

Large language models (LLMs) can often be made to behave in undesirable ways that they are explicitly fine-tuned not to. For example, the LLM red-teaming literature has produced a wide variety of 'jailbreaking' techniques to elicit harmful…

Applying a pre-trained large model to downstream tasks is prohibitive under resource-constrained conditions. Recent dominant approaches for addressing efficiency issues involve adding a few learnable parameters to the fixed backbone model.…

Computer Vision and Pattern Recognition · Computer Science 2023-11-21 Yangyang Guo , Guangzhi Wang , Mohan Kankanhalli

Deep features extracted from certain layers of a pre-trained deep model show superior performance over the conventional hand-crafted features. Compared with fine-tuning or linear probing that can explore diverse augmentations, \eg, random…

Computer Vision and Pattern Recognition · Computer Science 2024-08-27 Qi Qian , Yuanhong Xu , Juhua Hu

Fine-tuning pre-trained language models (LMs) has become the de facto standard in many NLP tasks. Nevertheless, fine-tuned LMs are still prone to robustness issues, such as adversarial robustness and model calibration. Several perspectives…

Computation and Language · Computer Science 2023-12-08 Jaehyung Kim , Yuning Mao , Rui Hou , Hanchao Yu , Davis Liang , Pascale Fung , Qifan Wang , Fuli Feng , Lifu Huang , Madian Khabsa

Adversarial training is an approach of increasing the robustness of models to adversarial attacks by including adversarial examples in the training set. One major challenge of producing adversarial examples is to contain sufficient…

Machine Learning · Computer Science 2021-10-13 Tianjin Huang , Vlado Menkovski , Yulong Pei , Mykola Pechenizkiy

Owing to security implications of adversarial vulnerability, adversarial robustness of deep metric learning models has to be improved. In order to avoid model collapse due to excessively hard examples, the existing defenses dismiss the…

Machine Learning · Computer Science 2022-03-04 Mo Zhou , Vishal M. Patel

Despite their advances and success, real-world deep neural networks are known to be vulnerable to adversarial attacks. Universal adversarial perturbation, an input-agnostic attack, poses a serious threat for them to be deployed in…

Machine Learning · Computer Science 2025-02-11 Bing Sun , Jun Sun , Wei Zhao

Federated Learning (FL) enables collaborative model training while preserving data privacy, but it is highly vulnerable to backdoor attacks. Most existing defense methods in FL have limited effectiveness due to their neglect of the model's…

Cryptography and Security · Computer Science 2025-08-05 Xinhai Yan , Libing Wu , Zhuangzhuang Zhang , Bingyi Liu , Lijuan Huo , Jing Wang

Adversarial images are designed to mislead deep neural networks (DNNs), attracting great attention in recent years. Although several defense strategies achieved encouraging robustness against adversarial samples, most of them fail to…

Machine Learning · Computer Science 2020-02-25 Hang Yu , Aishan Liu , Xianglong Liu , Gengchao Li , Ping Luo , Ran Cheng , Jichen Yang , Chongzhi Zhang