Related papers: On the Computation of Worst Attacks: a LP Framewor…
Large language models (LLMs) are increasingly trained on massive, heterogeneous text corpora, raising serious concerns about the unauthorised use of proprietary or personal data during model training. In this work, we address the problem of…
We propose an active-learning method for nonlinear minimax regression. Given a nonlinear function that can be arbitrarily evaluated over a compact set, we fit a surrogate model, such as a feedforward neural network, by minimizing the…
Mixtures of classifiers (a.k.a. randomized ensembles) have been proposed as a way to improve robustness against adversarial attacks. However, it has been shown that existing attacks are not well suited for this kind of classifiers. In this…
While there are various methods to detect application layer attacks or intrusion attempts on an individual end host, it is not efficient to provide all end hosts in the network with heavy-duty defense systems or software firewalls. In this…
We consider an unconstrained continuous optimization problem where, in each iteration, gradient estimates may be arbitrarily corrupted with a probability greater than 1/2. Additionally, function value estimates may exhibit heavy-tailed…
We consider the problem of discounted optimal state-feedback regulation for general unknown deterministic discrete-time systems. It is well known that open-loop instability of systems, non-quadratic cost functions and complex nonlinear…
It has been shown that natural language processing (NLP) models are vulnerable to a kind of security threat called the Backdoor Attack, which utilizes a `backdoor trigger' paradigm to mislead the models. The most threatening backdoor attack…
Many problems of interest for cyber-physical network systems can be formulated as Mixed Integer Linear Programs in which the constraints are distributed among the agents. In this paper we propose a distributed algorithm to solve this class…
Large language models (LLMs) are increasingly used to help security analysts manage the surge of cyber threats, automating tasks from vulnerability assessment to incident response. Yet in operational CTI workflows, reliability gaps remain…
Large Language Models (LLMs) aligned with human feedback have recently garnered significant attention. However, it remains vulnerable to jailbreak attacks, where adversaries manipulate prompts to induce harmful outputs. Exploring jailbreak…
To explore the vulnerability of deep neural networks (DNNs), many attack paradigms have been well studied, such as the poisoning-based backdoor attack in the training stage and the adversarial attack in the inference stage. In this paper,…
The advent of Large Language Models (LLMs) has marked significant achievements in language processing and reasoning capabilities. Despite their advancements, LLMs face vulnerabilities to data poisoning attacks, where the adversary inserts…
Stealth attacks pose potential risks to cyber-physical systems because they are difficult to detect. Assessing the risk of systems under stealth attacks remains an open challenge, especially in nonlinear systems. To comprehensively quantify…
Leading language model (LM) providers like OpenAI and Anthropic allow customers to fine-tune frontier LMs for specific use cases. To prevent abuse, these providers apply filters to block fine-tuning on overtly harmful data. In this setting,…
Despite the great achievements of deep neural networks (DNNs), the vulnerability of state-of-the-art DNNs raises security concerns of DNNs in many application domains requiring high reliability.We propose the fault sneaking attack on DNNs,…
While convenient, relying on LLM-powered code assistants in day-to-day work gives rise to severe attacks. For instance, the assistant might introduce subtle flaws and suggest vulnerable code to the user. These adversarial code-suggestions…
We introduce a novel generalization of entropy and conditional entropy from which most definitions from the literature can be derived as particular cases. Within this general framework, we investigate the problem of designing…
Adversarial attacks by malicious users that threaten the safety of large language models (LLMs) can be viewed as attempts to infer a target property $T$ that is unknown when an instruction is issued, and becomes knowable only after the…
Control policies, trained using the Deep Reinforcement Learning, have been recently shown to be vulnerable to adversarial attacks introducing even very small perturbations to the policy input. The attacks proposed so far have been designed…
We introduce a novel data-driven method to mitigate the risk of cascading failures in delayed discrete-time Linear Time-Invariant (LTI) systems. Our approach involves formulating a distributionally robust finite-horizon optimal control…