Related papers: Privacy Patterns
Prompt-tuning has received attention as an efficient tuning method in the language domain, i.e., tuning a prompt that is a few tokens long, while keeping the large language model frozen, yet achieving comparable performance with…
Regulations for privacy protection aim to protect individuals from the unauthorized storage, processing, and transfer of their personal data but oftentimes fail in providing helpful support for understanding these regulations. To better…
Federated learning (FL) enhances privacy by keeping user data on local devices. However, emerging attacks have demonstrated that the updates shared by users during training can reveal significant information about their data. This has…
We study a statistical signal processing privacy problem, where an agent observes useful data $Y$ and wants to reveal the information to a user. Since the useful data is correlated with the private data $X$, the agent employs a privacy…
Integrating Foundation Models (FMs) into recommendation systems is an emerging and promising research direction. However, centralized paradigms face growing pressure from privacy concerns and strict regulatory requirements. Federated…
Machine learning (ML) systems are increasingly deployed in high-stakes domains where reliability is paramount. This thesis investigates how uncertainty estimation can enhance the safety and trustworthiness of ML, focusing on selective…
Privacy analysis is critical but also a time-consuming and tedious task. We present a formalization which eases designing and auditing high-level privacy properties of software architectures. It is incorporated into a larger policy analysis…
The design of a statistical signal processing privacy problem is studied where the private data is assumed to be observable. In this work, an agent observes useful data $Y$, which is correlated with private data $X$, and wants to disclose…
We introduce a framework for reasoning about the security of computer systems using modal logic. This framework is sufficiently expressive to capture a variety of known security properties, while also being intuitive and independent of…
We present a practical framework to deploy privacy-preserving machine learning (PPML) applications in untrusted clouds based on a trusted execution environment (TEE). Specifically, we shield unmodified PyTorch ML applications by running…
Large language models (LLMs), especially those based on the Transformer architecture, have had a profound impact on various aspects of daily life, such as natural language processing, content generation, research methodologies, and more.…
In information systems, a system is analyzed using a modeling tool. Analysis is an important phase prior to implementation in order to obtain the correct requirements of the system. During the requirements phase, the software requirements…
We study statistical risk minimization problems under a privacy model in which the data is kept confidential even from the learner. In this local privacy framework, we establish sharp upper and lower bounds on the convergence rates of…
The population protocol model introduced by Angluin et al. in 2006 offers a theoretical framework for designing and analyzing distributed algorithms among limited-resource mobile agents. While the original population protocol model…
Service providers of large language model (LLM) applications collect user instructions in the wild and use them in further aligning LLMs with users' intentions. These instructions, which potentially contain sensitive information, are…
It is becoming increasingly clear that users should own and control their data. Utility providers are also becoming more interested in guaranteeing data privacy. As such, users and utility providers should collaborate in data privacy, a…
The rapid deployment of large language models (LLMs) in consumer applications has led to frequent exchanges of personal information. To obtain useful responses, users often share more than necessary, increasing privacy risks via…
We propose a general statistical inference framework to capture the privacy threat incurred by a user that releases data to a passive but curious adversary, given utility constraints. We show that applying this general framework to the…
We present in this paper the preliminary design of a module system based on a notion of components such as they are found in COM. This module system is inspired from that of Standard ML, and features first-class instances of components,…
The performance of modern machine learning systems depends on access to large, high-quality datasets, often sourced from user-generated content or proprietary, domain-specific corpora. However, these rich datasets inherently contain…