Related papers: Recipient Profiling: Predicting Characteristics fr…
Online Social Networking Sites attracted a massive number of users over the past decade but also raised privacy concerns with the amount of personal information disclosed. Studies have shown that 25% of the users are not aware of privacy…
The increasing popularity of online social network brings huge privacy threat for the end users. While existing work focus on inferring sensitive attributes from the social network such as age, location and gender, little has been done on…
We examine a large dialog corpus obtained from the conversation history of a single individual with 104 conversation partners. The corpus consists of half a million instant messages, across several messaging platforms. We focus our analyses…
Generative machine learning models are being increasingly viewed as a way to share sensitive data between institutions. While there has been work on developing differentially private generative modeling approaches, these approaches…
This article deals with adversarial attacks towards deep learning systems for Natural Language Processing (NLP), in the context of privacy protection. We study a specific type of attack: an attacker eavesdrops on the hidden representations…
Audio signals can reveal intimate details about a person's life, including their conversations, health status, emotions, location, and personal preferences. Unauthorized access or misuse of this information can have profound personal and…
Text embedding is becoming an increasingly popular AI methodology, especially among businesses, yet the potential of text embedding models to be biased is not well understood. This paper examines the degree to which a selection of popular…
It is difficult for individuals and organizations to protect personal information without a fundamental understanding of relative privacy risks. By analyzing over 5,000 empirical identity theft and fraud cases, this research identifies…
The interactive use of large language models (LLMs) in AI assistants (at work, home, etc.) introduces a new set of inference-time privacy risks: LLMs are fed different types of information from multiple sources in their inputs and are…
Practitioners commonly download pretrained machine learning models from open repositories and finetune them to fit specific applications. We show that this practice introduces a new risk of privacy backdoors. By tampering with a pretrained…
Personal photos of individuals when shared online, apart from exhibiting a myriad of memorable details, also reveals a wide range of private information and potentially entails privacy risks (e.g., online harassment, tracking). To mitigate…
The following work presents a new algorithm for character recognition from obfuscated images. The presented method is an example of a potential threat to current postal services. This paper both analyzes the efficiency of the given…
As text-to-image systems continue to grow in popularity with the general public, questions have arisen about bias and diversity in the generated images. Here, we investigate properties of images generated in response to prompts which are…
In collaborative learning, clients keep their data private and communicate only the computed gradients of the deep neural network being trained on their local data. Several recent attacks show that one can still extract private information…
High-latency anonymous communication systems prevent passive eavesdroppers from inferring communicating partners with certainty. However, disclosure attacks allow an adversary to recover users' behavioral profiles when communications are…
Pre-trained large language models, such as GPT\nobreakdash-2 and BERT, are often fine-tuned to achieve state-of-the-art performance on a downstream task. One natural example is the ``Smart Reply'' application where a pre-trained model is…
Authorship identification has proven unsettlingly effective in inferring the identity of the author of an unsigned document, even when sensitive personal information has been carefully omitted. In the digital era, individuals leave a…
A central tenet of Federated learning (FL), which trains models without centralizing user data, is privacy. However, previous work has shown that the gradient updates used in FL can leak user information. While the most industrial uses of…
Users of social networks display diversified behavior and online habits. For instance, a user's tendency to reply to a post can depend on the user and the person posting. For convenience, we group users into aggregated behavioral patterns,…
Social Media is a cyber-security risk for every business. What do people share on the Internet? Almost everything about oneself is shared: friendship, demographics, family, activities, and work-related information. This could become a…