Related papers: No More, No Less: Least-Privilege Language Models

In enterprise settings, organizational data is segregated, siloed and carefully protected by elaborate access control frameworks. These access control structures can completely break down if an LLM fine-tuned on the siloed data serves…

Safety-aligned language models are trained to refuse harmful requests, yet refusal behavior can be suppressed by steering their internal representations. Existing methods do so by ablating a refusal direction from model activations, aiming…

Authentication, authorization, and trust verification are central parts of an access control system. The conditions for granting access in such a system are collected in access policies. Since access conditions are often complex, dedicated…

Driven by the goal of eradicating language barriers on a global scale, machine translation has solidified itself as a key focus of artificial intelligence research today. However, such efforts have coalesced around a small subset of…

Large language models (LLMs) are increasingly deployed in enterprise settings where they handle sensitive documents and user context, raising acute concerns over security and controllability. Conventional access control regulates whether…

Language-based information flow methods offer a principled way to enforce strong security properties, but enforcing noninterference is too inflexible for realistic applications. Security-typed languages have therefore introduced…

Secure Multi-Party Computation (MPC) is an important enabling technology for data privacy in modern distributed applications. Currently, proof methods for low-level MPC protocols are primarily manual and thus tedious and error-prone, and…

Despite the impressive capabilities of large language models, their substantial computational costs, latency, and privacy risks hinder their widespread deployment in real-world applications. Small Language Models (SLMs) with fewer than 10…

To design a discretionary access control policy, a technique is proposed that uses the principle of analogies and is based on both the properties of objects and the properties of subjects. As attributes characterizing these properties, the…

The Model Context Protocol (MCP) has emerged as a widely adopted mechanism for connecting large language models to external tools and resources. While MCP promises seamless extensibility and rich integrations, it also introduces a…

As large language models (LLMs) become increasingly capable, it is prudent to assess whether safety measures remain effective even if LLMs intentionally try to bypass them. Previous work introduced control evaluations, an adversarial…

HyperLTL, the extension of Linear Temporal Logic by trace quantifiers, is a uniform framework for expressing information flow policies by relating multiple traces of a security-critical system. HyperLTL has been successfully applied to…

The widespread adoption of Language Models (LMs) across industries is driving interest in deploying these services across the computing continuum, from the cloud to the network edge. This shift aims to reduce costs, lower latency, and…

We study admission control mechanisms for wireless access networks where (i) each user has a minimum service requirement, (ii) the capacity of the access network is limited, and (iii) the access point is not allowed to use monetary…

As language models are increasingly deployed as autonomous agents in high-stakes settings, ensuring that they reliably follow user-defined rules has become a critical safety concern. To this end, we study whether language models exhibit…

Large-scale language models have achieved tremendous success across various natural language processing (NLP) applications. Nevertheless, language models are vulnerable to backdoor attacks, which inject stealthy triggers into models for…

Designing and implementing secure software is inarguably more important than ever. However, despite years of research into privilege separating programs, it remains difficult to actually do so and such efforts can take years of…

Permissions are highly sensitive in Internet-of-Things (IoT) applications, as IoT devices collect our personal data and control the safety of our environment. Rather than simply granting permissions, further constraints shall be imposed on…

Large language models increasingly operate as autonomous agents that select and invoke tools from large registries. We identify a critical gap: when unauthorized tools are visible in an agent's context, models select them in adversarial…

Organizations often lay down rules or guidelines called Natural Language Access Control Policies (NLACPs) for specifying who gets access to which information and when. However, these cannot be directly used in a target access control model…