Related papers: Auditing Privacy in Multi-Tenant RAG under Account…

The rapid growth of Decentralized Finance (DeFi) has been accompanied by substantial financial losses due to smart contract vulnerabilities, underscoring the critical need for effective security auditing. With attacks becoming more…

Regulatory audits of AI systems increasingly rely on differential privacy (DP) to protect training data and model internals. We study audit design when the audited developer can strategically respond to the privacy-constrained audit…

Secure aggregation (SecAgg) is a commonly-used privacy-enhancing mechanism in federated learning, affording the server access only to the aggregate of model updates while safeguarding the confidentiality of individual updates. Despite…

The potential risk of privacy leakage prevents users from sharing their honest opinions on social platforms. This paper addresses the problem of privacy preservation if the query returns the histogram of rankings. The framework of…

In spite that Federated Learning (FL) is well known for its privacy protection when training machine learning models among distributed clients collaboratively, recent studies have pointed out that the naive FL is susceptible to gradient…

The continued promise of Large Language Models (LLMs), particularly in their natural language understanding and generation capabilities, has driven a rapidly increasing interest in identifying and developing LLM use cases. In an effort to…

This paper studies the problem of multi-agent computation under the differential privacy requirement of the agents' local datasets against eavesdroppers having node-to-node communications. We first propose for the network equipped with…

Multi-agent LLM systems introduce a security risk in which sensitive information accessed by one agent can propagate through shared context and reappear in downstream outputs, even without explicit adversarial intent. We formalise this…

Privacy amplification exploits randomness in data selection to provide tighter differential privacy (DP) guarantees. This analysis is key to DP-SGD's success in machine learning, but, is not readily applicable to the newer state-of-the-art…

Individual Differential Privacy (iDP) promises users control over their privacy, but this promise can be broken in practice. We reveal a previously overlooked vulnerability in sampling-based iDP mechanisms: while conforming to the iDP…

Consider the problem of Private Information Retrieval (PIR), where a user wishes to retrieve a single message from $N$ non-communicating and non-colluding databases (servers). All servers store the same set of $M$ messages and they respond…

Multi-modal retrieval-augmented generation (MRAG) systems retrieve visual evidence from large image corpora to ground the responses of large multi-modal models, yet the retrieved images frequently contain human faces whose identities…

Retrieval-augmented generation (RAG) grounds answers in retrieved passages, but retrieval is not verification: a passage can be topical and still fail to justify the answer. We frame this gap as evidence sufficiency verification for…

Multi-agent Large Language Model (LLM) systems create privacy risks that current benchmarks cannot measure. When agents coordinate on tasks, sensitive data passes through inter-agent messages, shared memory, and tool arguments, all pathways…

Gradient leakage attacks are considered one of the wickedest privacy threats in deep learning as attackers covertly spy gradient updates during iterative training without compromising model training quality, and yet secretly reconstruct…

State-of-the-art Differentially Private (DP) synthetic data generators such as MST and AIM are widely used, yet tightly auditing their privacy guarantees remains challenging. We introduce a Gaussian Differential Privacy (GDP)-based auditing…

Distributed aggregative optimization underpins many cooperative optimization and multi-agent control systems, where each agent's objective function depends both on its local optimization variable and an aggregate of all agents' optimization…

Prior work on differential privacy analysis of randomized SGD algorithms relies on composition theorems, where the implicit (unrealistic) assumption is that the internal state of the iterative algorithm is revealed to the adversary. As a…

Computing the principal component (PC) of the adjacency matrix of an undirected graph has several applications ranging from identifying key vertices for influence maximization and controlling diffusion processes, to discovering densely…

Retrieval-augmented generation (RAG) has rapidly emerged as a transformative approach for integrating large language models into clinical and biomedical workflows. However, privacy risks, such as protected health information (PHI) exposure,…