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

A rapidly growing number of applications rely on a small set of closed-source language models (LMs). This dependency might introduce novel security risks if LMs develop self-recognition capabilities. Inspired by human identity verification…

Regulators currently govern the AI data economy based on intuition rather than evidence, struggling to choose between inconsistent regimes of informed consent, immunity, and liability. To fill this policy vacuum, this paper develops a novel…

Large language models pick up social biases from the data they are trained on and carry those biases into downstream applications, often reinforcing stereotypes around gender, race, religion, disability, age, and socioeconomic status. The…

Inference-time computation offers a powerful axis for scaling the performance of language models. However, naively increasing computation in techniques like Best-of-N sampling can lead to performance degradation due to reward hacking.…

Large Language Models (LLMs) have become pivotal in advancing natural language processing, yet their potential to perpetuate biases poses significant concerns. This paper introduces a new framework employing Direct Preference Optimization…

Style transfer is the task of rewriting a sentence into a target style while approximately preserving content. While most prior literature assumes access to a large style-labelled corpus, recent work (Riley et al. 2021) has attempted…

We investigate how large language models perform on low-resource languages by benchmarking eight LLMs across five experimental conditions in English, Kazakh, and Mongolian. Using 50 hand-crafted questions spanning factual, reasoning,…

While large language models (LLMs) show impressive decision-making abilities, current methods lack a mechanism for automatic self-improvement from errors during task execution. We propose LEAP, an iterative fine-tuning framework that…

Societal biases present in pre-trained large language models are a critical issue as these models have been shown to propagate biases in countless downstream applications, rendering them unfair towards specific groups of people. Since…

In the field of access control, delegation is an important aspect that is considered as a part of the administration mechanism. Thus, a complete access control must provide a flexible administration model to manage delegation.…

Language models have gained significant interest due to their general-purpose capabilities, which appear to emerge as models are scaled to increasingly larger parameter sizes. However, these large models impose stringent requirements on…

Recent studies have revealed a security threat to natural language processing (NLP) models, called the Backdoor Attack. Victim models can maintain competitive performance on clean samples while behaving abnormally on samples with a specific…

This paper introduces a run-time mechanism for preventing leakage of secure information in distributed systems. We consider a general concurrency language model, where concurrent objects interact by asynchronous method calls and futures.…

We initiate the study of Access Control Encryption (ACE), a novel cryptographic primitive that allows fine-grained access control, by giving different rights to different users not only in terms of which messages they are allowed to…

Recently, physical layer security based approaches have drawn considerable attentions and are envisaged to provide secure communications in the wireless networks. However, most existing literatures only focus on the physical layer. Thus,…

LLM-enabled applications are rapidly reshaping the software ecosystem by using large language models as core reasoning components for complex task execution. This paradigm shift, however, introduces fundamentally new reliability challenges…

We survey recent work on the specification of an access control mechanism in a collaborative environment. The work is presented in the context of the WebdamLog language, an extension of datalog to a distributed context. We discuss a…

As Model Context Protocol adoption grows, securing tool invocations via meaningful user consent has become a critical challenge, as existing methods, broad always allow toggles or opaque LLM-based decisions, fail to account for dangerous…

The increasing number of wireless devices operating in unlicensed spectrum motivates the development of intelligent adaptive approaches to spectrum access. We consider decentralized contention-based medium access for base stations (BSs)…

A key challenge in applying reinforcement learning (RL) to diffusion large language models (dLLMs) lies in the intractability of their likelihood functions, which are essential for the RL objective, necessitating corresponding approximation…