169 citations · 685 across the 42 of their papers we have counts for
45 papers
Mitigating Private Data Leakage in LLMs with Whiteout
Anna Yoo Jeong Ha, Ronik Bhaskar, Haitao Zheng +1
Modern large language models (LLMs) are trained on massive, largely unfiltered datasets, including content scraped from nearly every accessible website and user inputs. As a result…
On the Feasibility of Poisoning Text-to-Image AI Models via Adversarial Mislabeling
Stanley Wu, Ronik Bhaskar, Anna Yoo Jeong Ha +3
Today's text-to-image generative models are trained on millions of images sourced from the Internet, each paired with a detailed caption produced by Vision-Language Models (VLMs).…
Understanding Implosion in Text-to-Image Generative Models
Wenxin Ding, Cathy Y. Li, Shawn Shan +2
Recent works show that text-to-image generative models are surprisingly vulnerable to a variety of poisoning attacks. Empirical results find that these models can be corrupted by a…
Disrupting Style Mimicry Attacks on Video Imagery
Josephine Passananti, Stanley Wu, Shawn Shan +2
Generative AI models are often used to perform mimicry attacks, where a pretrained model is fine-tuned on a small sample of images to learn to mimic a specific artist of interest.…
Inception Attacks: Immersive Hijacking in Virtual Reality Systems
Zhuolin Yang, Cathy Yuanchen Li, Arman Bhalla +2
Today's virtual reality (VR) systems provide immersive interactions that seamlessly connect users with online services and one another. However, these immersive interfaces also int…
Organic or Diffused: Can We Distinguish Human Art from AI-generated Images?
Anna Yoo Jeong Ha, Josephine Passananti, Ronik Bhaskar +4
The advent of generative AI images has completely disrupted the art world. Distinguishing AI generated images from human art is a challenging problem whose impact is growing over t…
Towards Scalable and Robust Model Versioning
Wenxin Ding, Arjun Nitin Bhagoji, Ben Y. Zhao +1
As the deployment of deep learning models continues to expand across industries, the threat of malicious incursions aimed at gaining access to these deployed models is on the rise.…
A Response to Glaze Purification via IMPRESS
Shawn Shan, Stanley Wu, Haitao Zheng +1
Recent work proposed a new mechanism to remove protective perturbation added by Glaze in order to again enable mimicry of art styles from images protected by Glaze. Despite promisi…
Can Virtual Reality Protect Users from Keystroke Inference Attacks?
Zhuolin Yang, Zain Sarwar, Iris Hwang +3
Virtual Reality (VR) has gained popularity by providing immersive and interactive experiences without geographical limitations. It also provides a sense of personal privacy through…
Nightshade: Prompt-Specific Poisoning Attacks on Text-to-Image Generative Models
Shawn Shan, Wenxin Ding, Josephine Passananti +3
Data poisoning attacks manipulate training data to introduce unexpected behaviors into machine learning models at training time. For text-to-image generative models with massive tr…
Characterizing the Optimal 0-1 Loss for Multi-class Classification with a Test-time Attacker
Sihui Dai, Wenxin Ding, Arjun Nitin Bhagoji +4
Finding classifiers robust to adversarial examples is critical for their safe deployment. Determining the robustness of the best possible classifier under a given threat model for…
Glaze: Protecting Artists from Style Mimicry by Text-to-Image Models
Shawn Shan, Jenna Cryan, Emily Wenger +3
Recent text-to-image diffusion models such as MidJourney and Stable Diffusion threaten to displace many in the professional artist community. In particular, models can learn to mim…
Natural Backdoor Datasets
Emily Wenger, Roma Bhattacharjee, Arjun Nitin Bhagoji +4
Extensive literature on backdoor poison attacks has studied attacks and defenses for backdoors using "digital trigger patterns." In contrast, "physical backdoors" use physical obje…
Understanding Robust Learning through the Lens of Representation Similarities
Christian Cianfarani, Arjun Nitin Bhagoji, Vikash Sehwag +3
Representation learning, i.e. the generation of representations useful for downstream applications, is a task of fundamental importance that underlies much of the success of deep n…
On the Permanence of Backdoors in Evolving Models
Huiying Li, Arjun Nitin Bhagoji, Yuxin Chen +2
Existing research on training-time attacks for deep neural networks (DNNs), such as backdoors, largely assume that models are static once trained, and hidden backdoors trained into…
Post-breach Recovery: Protection against White-box Adversarial Examples for Leaked DNN Models
Shawn Shan, Wenxin Ding, Emily Wenger +2
Server breaches are an unfortunate reality on today's Internet. In the context of deep neural network (DNN) models, they are particularly harmful, because a leaked model gives an a…
Assessing Privacy Risks from Feature Vector Reconstruction Attacks
Emily Wenger, Francesca Falzon, Josephine Passananti +2
In deep neural networks for facial recognition, feature vectors are numerical representations that capture the unique features of a given face. While it is known that a version of…
SoK: Anti-Facial Recognition Technology
Emily Wenger, Shawn Shan, Haitao Zheng +1
The rapid adoption of facial recognition (FR) technology by both government and commercial entities in recent years has raised concerns about civil liberties and privacy. In respon…
Poison Forensics: Traceback of Data Poisoning Attacks in Neural Networks
Shawn Shan, Arjun Nitin Bhagoji, Haitao Zheng +1
In adversarial machine learning, new defenses against attacks on deep learning systems are routinely broken soon after their release by more powerful attacks. In this context, fore…
"Hello, It's Me": Deep Learning-based Speech Synthesis Attacks in the Real World
Emily Wenger, Max Bronckers, Christian Cianfarani +4
Advances in deep learning have introduced a new wave of voice synthesis tools, capable of producing audio that sounds as if spoken by a target speaker. If successful, such tools in…
Towards Performance Clarity of Edge Video Analytics
Zhujun Xiao, Zhengxu Xia, Haitao Zheng +2
Edge video analytics is becoming the solution to many safety and management tasks. Its wide deployment, however, must first address the tension between inference accuracy and resou…
Understanding the Effect of Bias in Deep Anomaly Detection
Ziyu Ye, Yuxin Chen, Haitao Zheng
Anomaly detection presents a unique challenge in machine learning, due to the scarcity of labeled anomaly data. Recent work attempts to mitigate such problems by augmenting trainin…
A Real-time Defense against Website Fingerprinting Attacks
Shawn Shan, Arjun Nitin Bhagoji, Haitao Zheng +1
Anonymity systems like Tor are vulnerable to Website Fingerprinting (WF) attacks, where a local passive eavesdropper infers the victim's activity. Current WF attacks based on deep…
Backdoor Attacks Against Deep Learning Systems in the Physical World
Emily Wenger, Josephine Passananti, Arjun Bhagoji +3
Backdoor attacks embed hidden malicious behaviors into deep learning models, which only activate and cause misclassifications on model inputs containing a specific trigger. Existin…