808 citations · 1.1k across the 21 of their papers we have counts for
22 papers
Throttling Web Agents Using Reasoning Gates
Abhinav Kumar, Jaechul Roh, Ali Naseh +2
AI web agents use Internet resources at far greater speed, scale, and complexity -- changing how users and services interact. Deployed maliciously or erroneously, these agents coul…
Network-Level Prompt and Trait Leakage in Local Research Agents
Hyejun Jeong, Mohammadreza Teymoorianfard, Abhinav Kumar +2
We show that Web and Research Agents (WRAs) -- language-model-based systems that investigate complex topics on the Internet -- are vulnerable to inference attacks by passive networ…
Privacy Reasoning in Ambiguous Contexts
Ren Yi, Octavian Suciu, Adria Gascon +3
We study the ability of language models to reason about appropriate information disclosure - a central aspect of the evolving field of agentic privacy. Whereas previous works have…
Can Large Language Models Really Recognize Your Name?
Dzung Pham, Peter Kairouz, Niloofar Mireshghallah +3
Large language models (LLMs) are increasingly being used in privacy pipelines to detect and remedy sensitive data leakage. These solutions often rely on the premise that LLMs can r…
OverThink: Slowdown Attacks on Reasoning LLMs
Abhinav Kumar, Jaechul Roh, Ali Naseh +4
A reasoning language model (RLM) generates costly reasoning tokens, often hidden from the users, that help it excel at many tasks. Our Overthink attack targets RLM-based applicatio…
Firewalls to Secure Dynamic LLM Agentic Networks
Sahar Abdelnabi, Amr Gomaa, Eugene Bagdasarian +2
The emergence of agent-to-agent communication protocols mirrors the early internet: powerful connectivity with minimal security infrastructure. When AI agents communicate on behalf…
Trusted Machine Learning Models Unlock Private Inference for Problems Currently Infeasible with Cryptography
Ilia Shumailov, Daniel Ramage, Sarah Meiklejohn +4
We often interact with untrusted parties. Prioritization of privacy can limit the effectiveness of these interactions, as achieving certain goals necessitates sharing private data.…
CI-Bench: Benchmarking Contextual Integrity of AI Assistants on Synthetic Data
Zhao Cheng, Diane Wan, Matthew Abueg +6
Advances in generative AI point towards a new era of personalized applications that perform diverse tasks on behalf of users. While general AI assistants have yet to fully emerge,…
Backdooring Bias () into Stable Diffusion Models
Ali Naseh, Jaechul Roh, Eugene Bagdasarian +1
Recent advances in large text-conditional diffusion models have revolutionized image generation by enabling users to create realistic, high-quality images from textual prompts, sig…
AirGapAgent: Protecting Privacy-Conscious Conversational Agents
Eugene Bagdasarian, Ren Yi, Sahra Ghalebikesabi +5
The growing use of large language model (LLM)-based conversational agents to manage sensitive user data raises significant privacy concerns. While these agents excel at understandi…
Synthesizing Physical Backdoor Datasets: An Automated Framework Leveraging Deep Generative Models
Sze Jue Yang, Chinh D. La, Quang H. Nguyen +4
Backdoor attacks, representing an emerging threat to the integrity of deep neural networks, have garnered significant attention due to their ability to compromise deep learning sys…
Adversarial Illusions in Multi-Modal Embeddings
Tingwei Zhang, Rishi Jha, Eugene Bagdasaryan +1
Multi-modal embeddings encode texts, images, thermal images, sounds, and videos into a single embedding space, aligning representations across different modalities (e.g., associate…
Abusing Images and Sounds for Indirect Instruction Injection in Multi-Modal LLMs
Eugene Bagdasaryan, Tsung-Yin Hsieh, Ben Nassi +1
We demonstrate how images and sounds can be used for indirect prompt and instruction injection in multi-modal LLMs. An attacker generates an adversarial perturbation corresponding…
Mithridates: Auditing and Boosting Backdoor Resistance of Machine Learning Pipelines
Eugene Bagdasaryan, Vitaly Shmatikov
Machine learning (ML) models trained on data from potentially untrusted sources are vulnerable to poisoning. A small, maliciously crafted subset of the training inputs can cause th…
Training a Tokenizer for Free with Private Federated Learning
Eugene Bagdasaryan, Congzheng Song, Rogier van Dalen +2
Federated learning with differential privacy, i.e. private federated learning (PFL), makes it possible to train models on private data distributed across users' devices without har…
Spinning Language Models: Risks of Propaganda-As-A-Service and Countermeasures
Eugene Bagdasaryan, Vitaly Shmatikov
We investigate a new threat to neural sequence-to-sequence (seq2seq) models: training-time attacks that cause models to "spin" their outputs so as to support an adversary-chosen se…
Towards Sparse Federated Analytics: Location Heatmaps under Distributed Differential Privacy with Secure Aggregation
Eugene Bagdasaryan, Peter Kairouz, Stefan Mellem +4
We design a scalable algorithm to privately generate location heatmaps over decentralized data from millions of user devices. It aims to ensure differential privacy before data bec…
Blind Backdoors in Deep Learning Models
Eugene Bagdasaryan, Vitaly Shmatikov
We investigate a new method for injecting backdoors into machine learning models, based on compromising the loss-value computation in the model-training code. We use it to demonstr…
Policy-Based Federated Learning
Kleomenis Katevas, Eugene Bagdasaryan, Jason Waterman +4
In this paper we present PoliFL, a decentralized, edge-based framework that supports heterogeneous privacy policies for federated learning. We evaluate our system on three use case…
Salvaging Federated Learning by Local Adaptation
Tao Yu, Eugene Bagdasaryan, Vitaly Shmatikov
Federated learning (FL) is a heavily promoted approach for training ML models on sensitive data, e.g., text typed by users on their smartphones. FL is expressly designed for traini…
Differential Privacy Has Disparate Impact on Model Accuracy
Eugene Bagdasaryan, Vitaly Shmatikov
Differential privacy (DP) is a popular mechanism for training machine learning models with bounded leakage about the presence of specific points in the training data. The cost of d…
How To Backdoor Federated Learning
Eugene Bagdasaryan, Andreas Veit, Yiqing Hua +2
Federated learning enables thousands of participants to construct a deep learning model without sharing their private training data with each other. For example, multiple smartphon…