491 citations · 632 across the 24 of their papers we have counts for
25 papers
Class-Proportional Coreset Selection for Difficulty-Separable Data
Elisa Tsai, Haizhong Zheng, Atul Prakash
High-quality training data is essential for building reliable and efficient machine learning systems. One-shot coreset selection addresses this by pruning the dataset while maintai…
Test-Time Canonicalization by Foundation Models for Robust Perception
Utkarsh Singhal, Ryan Feng, Stella X. Yu +1
Perception in the real world requires robustness to diverse viewing conditions. Existing approaches often rely on specialized architectures or training with predefined data augment…
Taming Data Challenges in ML-based Security Tasks Using Generative AI
Shravya Kanchi, Neal Mangaokar, Aravind Cheruvu +4
Machine learning-based supervised classifiers are widely used for security tasks, and their improvement has been largely focused on algorithmic advancements. We argue that data cha…
What Really is a Member? Discrediting Membership Inference via Poisoning
Neal Mangaokar, Ashish Hooda, Zhuohang Li +5
Membership inference tests aim to determine whether a particular data point was included in a language model's training set. However, recent works have shown that such tests often…
Harmful Terms and Where to Find Them: Measuring and Modeling Unfavorable Financial Terms and Conditions in Shopping Websites at Scale
Elisa Tsai, Neal Mangaokar, Boyuan Zheng +2
Terms and conditions for online shopping websites often contain terms that can have significant financial consequences for customers. Despite their impact, there is currently no co…
ELFS: Label-Free Coreset Selection with Proxy Training Dynamics
Haizhong Zheng, Elisa Tsai, Yifu Lu +4
High-quality human-annotated data is crucial for modern deep learning pipelines, yet the human annotation process is both costly and time-consuming. Given a constrained human label…
PRP: Propagating Universal Perturbations to Attack Large Language Model Guard-Rails
Neal Mangaokar, Ashish Hooda, Jihye Choi +4
Large language models (LLMs) are typically aligned to be harmless to humans. Unfortunately, recent work has shown that such models are susceptible to automated jailbreak attacks th…
Plato: Plan to Efficiently Decode for Large Language Model Inference
Shuowei Jin, Xueshen Liu, Yongji Wu +7
Large language models (LLMs) have achieved remarkable success in natural language tasks, but their inference incurs substantial computational and memory overhead. To improve effici…
Learn To be Efficient: Build Structured Sparsity in Large Language Models
Haizhong Zheng, Xiaoyan Bai, Xueshen Liu +4
Large Language Models (LLMs) have achieved remarkable success with their billion-level parameters, yet they incur high inference overheads. The emergence of activation sparsity in…
Leveraging Hierarchical Feature Sharing for Efficient Dataset Condensation
Haizhong Zheng, Jiachen Sun, Shutong Wu +4
Given a real-world dataset, data condensation (DC) aims to synthesize a small synthetic dataset that captures the knowledge of a natural dataset while being usable for training mod…
Theoretically Principled Trade-off for Stateful Defenses against Query-Based Black-Box Attacks
Ashish Hooda, Neal Mangaokar, Ryan Feng +3
Adversarial examples threaten the integrity of machine learning systems with alarming success rates even under constrained black-box conditions. Stateful defenses have emerged as a…
CALICO: Self-Supervised Camera-LiDAR Contrastive Pre-training for BEV Perception
Jiachen Sun, Haizhong Zheng, Qingzhao Zhang +3
Perception is crucial in the realm of autonomous driving systems, where bird's eye view (BEV)-based architectures have recently reached state-of-the-art performance. The desirabili…
Stateful Defenses for Machine Learning Models Are Not Yet Secure Against Black-box Attacks
Ryan Feng, Ashish Hooda, Neal Mangaokar +3
Recent work has proposed stateful defense models (SDMs) as a compelling strategy to defend against a black-box attacker who only has query access to the model, as is common for onl…
Coverage-centric Coreset Selection for High Pruning Rates
Haizhong Zheng, Rui Liu, Fan Lai +1
One-shot coreset selection aims to select a representative subset of the training data, given a pruning rate, that can later be used to train future models while retaining high acc…
Defending Object Detectors against Patch Attacks with Out-of-Distribution Smoothing
Ryan Feng, Neal Mangaokar, Jihye Choi +2
Patch attacks against object detectors have been of recent interest due to their being physically realizable and more closely aligned with practical systems. In response to this th…
Concept-based Explanations for Out-Of-Distribution Detectors
Jihye Choi, Jayaram Raghuram, Ryan Feng +3
Out-of-distribution (OOD) detection plays a crucial role in ensuring the safe deployment of deep neural network (DNN) classifiers. While a myriad of methods have focused on improvi…
D4: Detection of Adversarial Diffusion Deepfakes Using Disjoint Ensembles
Ashish Hooda, Neal Mangaokar, Ryan Feng +3
Detecting diffusion-generated deepfake images remains an open problem. Current detection methods fail against an adversary who adds imperceptible adversarial perturbations to the d…
Using Anomaly Feature Vectors for Detecting, Classifying and Warning of Outlier Adversarial Examples
Nelson Manohar-Alers, Ryan Feng, Sahib Singh +2
We present DeClaW, a system for detecting, classifying, and warning of adversarial inputs presented to a classification neural network. In contrast to current state-of-the-art meth…
Content-Adaptive Pixel Discretization to Improve Model Robustness
Ryan Feng, Wu-chi Feng, Atul Prakash
Preprocessing defenses such as pixel discretization are appealing to remove adversarial attacks due to their simplicity. However, they have been shown to be ineffective except on s…
GRAPHITE: Generating Automatic Physical Examples for Machine-Learning Attacks on Computer Vision Systems
Ryan Feng, Neal Mangaokar, Jiefeng Chen +3
This paper investigates an adversary's ease of attack in generating adversarial examples for real-world scenarios. We address three key requirements for practical attacks for the r…
Robust Classification using Robust Feature Augmentation
Kevin Eykholt, Swati Gupta, Atul Prakash +3
Existing deep neural networks, say for image classification, have been shown to be vulnerable to adversarial images that can cause a DNN misclassification, without any perceptible…
Physical Adversarial Examples for Object Detectors
Kevin Eykholt, Ivan Evtimov, Earlence Fernandes +6
Deep neural networks (DNNs) are vulnerable to adversarial examples-maliciously crafted inputs that cause DNNs to make incorrect predictions. Recent work has shown that these attack…
Tyche: Risk-Based Permissions for Smart Home Platforms
Amir Rahmati, Earlence Fernandes, Kevin Eykholt +1
Emerging smart home platforms, which interface with a variety of physical devices and support third-party application development, currently use permission models inspired by smart…
Robust Physical-World Attacks on Deep Learning Models
Kevin Eykholt, Ivan Evtimov, Earlence Fernandes +6
Recent studies show that the state-of-the-art deep neural networks (DNNs) are vulnerable to adversarial examples, resulting from small-magnitude perturbations added to the input. G…