搜索结果：attacks

Space infrastructures represent an emerging domain that is critical to the global economy and society. However, this domain is vulnerable to attacks, including cyber attacks and other kinds of attacks. To enhance the resilience of this domain, we must understand these attacks that can be waged against it and the defenses that can be employed to mitigate these attacks. The status quo is that there is neither a systematic understanding of these attacks against, nor defenses for, space infrastructures, despite their clear importance in guiding systematic analysis of space security and future research. In this paper, we fill the void by proposing the first systematic taxonomy of attacks against, and defenses for, space infrastructures. We hope this paper will inspire a community effort at refining the taxonomy towards a widely used one.

Large Language Models (LLMs) are vulnerable to attacks like prompt injection, backdoor attacks, and adversarial attacks, which manipulate prompts or models to generate harmful outputs. In this paper, departing from traditional deep learning attack paradigms, we explore their intrinsic relationship and collectively term them Prompt Trigger Attacks (PTA). This raises a key question: Can we determine if a prompt is benign or poisoned? To address this, we propose UniGuardian, the first unified defense mechanism designed to detect prompt injection, backdoor attacks, and adversarial attacks in LLMs. Additionally, we introduce a single-forward strategy to optimize the detection pipeline, enabling simultaneous attack detection and text generation within a single forward pass. Our experiments confirm that UniGuardian accurately and efficiently identifies malicious prompts in LLMs.

The Internet of Things (IoT) has garnered significant interest in both research and industry due to its profound impact on human life. The rapid expansion of IoT technology has ushered in smart healthcare, smart devices, smart cities, and smart grids. However, the security of IoT devices, particularly in healthcare, has become a major concern, with recent attacks revealing serious vulnerabilities. In IoT networks, where connected devices are susceptible to resource-constraint attacks, such as energy consumption attacks, security is paramount. This paper explores the impact of Distributed Denial of Service (DDoS) and Fake Access Points (F-APs) attacks on WiFi-enabled smart healthcare devices. Specifically, it investigates how these attacks can disrupt service on victim devices and Access Points (APs), focusing on device connectivity and energy consumption during attacks. Key findings include identifying the attack rates of DDoS attacks that disrupt services and quantifying the energy consumption impact of Energy Consumption Distributed Denial of Service (EC-DDoS) and F-APs attacks on smart healthcare devices. The study highlights communication protocols, attack rates, payload sizes,

In an era where misinformation spreads freely, fact-checking (FC) plays a crucial role in verifying claims and promoting reliable information. While automated fact-checking (AFC) has advanced significantly, existing systems remain vulnerable to adversarial attacks that manipulate or generate claims, evidence, or claim-evidence pairs. These attacks can distort the truth, mislead decision-makers, and ultimately undermine the reliability of FC models. Despite growing research interest in adversarial attacks against AFC systems, a comprehensive, holistic overview of key challenges remains lacking. These challenges include understanding attack strategies, assessing the resilience of current models, and identifying ways to enhance robustness. This survey provides the first in-depth review of adversarial attacks targeting FC, categorizing existing attack methodologies and evaluating their impact on AFC systems. Additionally, we examine recent advancements in adversary-aware defenses and highlight open research questions that require further exploration. Our findings underscore the urgent need for resilient FC frameworks capable of withstanding adversarial manipulations in pursuit of prese

Large Vision-Language Models (LVLMs) are susceptible to typographic attacks, which are misclassifications caused by an attack text that is added to an image. In this paper, we introduce a multi-image setting for studying typographic attacks, broadening the current emphasis of the literature on attacking individual images. Specifically, our focus is on attacking image sets without repeating the attack query. Such non-repeating attacks are stealthier, as they are more likely to evade a gatekeeper than attacks that repeat the same attack text. We introduce two attack strategies for the multi-image setting, leveraging the difficulty of the target image, the strength of the attack text, and text-image similarity. Our text-image similarity approach improves attack success rates by 21% over random, non-specific methods on the CLIP model using ImageNet while maintaining stealth in a multi-image scenario. An additional experiment demonstrates transferability, i.e., text-image similarity calculated using CLIP transfers when attacking InstructBLIP.

Imperceptible adversarial attacks aim to fool DNNs by adding imperceptible perturbation to the input data. Previous methods typically improve the imperceptibility of attacks by integrating common attack paradigms with specifically designed perception-based losses or the capabilities of generative models. In this paper, we propose Adversarial Attacks in Diffusion (AdvAD), a novel modeling framework distinct from existing attack paradigms. AdvAD innovatively conceptualizes attacking as a non-parametric diffusion process by theoretically exploring basic modeling approach rather than using the denoising or generation abilities of regular diffusion models requiring neural networks. At each step, much subtler yet effective adversarial guidance is crafted using only the attacked model without any additional network, which gradually leads the end of diffusion process from the original image to a desired imperceptible adversarial example. Grounded in a solid theoretical foundation of the proposed non-parametric diffusion process, AdvAD achieves high attack efficacy and imperceptibility with intrinsically lower overall perturbation strength. Additionally, an enhanced version AdvAD-X is propo

The run-time electromagnetic (EM) emanation of microprocessors presents a side-channel that leaks the confidentiality of the applications running on them. Many recent works have demonstrated successful attacks leveraging such side-channels to extract the confidentiality of diverse applications, such as the key of cryptographic algorithms and the hyperparameter of neural network models. This paper proposes ShuffleV, a microarchitecture defense strategy against EM Side-Channel Attacks (SCAs). ShuffleV adopts the moving target defense (MTD) philosophy, by integrating hardware units to randomly shuffle the execution order of program instructions and optionally insert dummy instructions, to nullify the statistical observation by attackers across repetitive runs. We build ShuffleV on the open-source RISC-V core and provide six design options, to suit different application scenarios. To enable rapid evaluation, we develop a ShuffleV simulator that can help users to (1) simulate the performance overhead for each design option and (2) generate an execution trace to validate the randomness of execution on their workload. We implement ShuffleV on a Xilinx PYNQ-Z2 FPGA and validate its perform

Quantum conditional entropies play a fundamental role in quantum information theory. In quantum key distribution, they are exploited to obtain reliable lower bounds on the secret-key rates in the finite-size regime, against collective attacks and coherent attacks under suitable assumptions. Here we consider continuous-variable communication protocols, where the sender Alice encodes information using a discrete modulation of phase-shifted coherent states, and the receiver Bob decodes by homodyne or heterodyne detection. We compute the Petz-Rényi and sandwiched Rényi conditional entropies associated with these setups, assuming either a passive eavesdropper or one that injects thermal photons into the channel, who gathers the quantum information leaked through a lossy communication line of known or bounded transmittance. Whereas our results do not directly provide reliable key-rate estimates, they do represent useful ball-park figures. We obtain analytical or semi-analytical expressions that do not require intensive numerical calculations. These expressions serve as bounds on the key rates that may be tight in certain scenarios. We compare different estimates, including known bounds t

In this work, we reexamine the vulnerability of Payment Channel Networks (PCNs) to bribing attacks, where an adversary incentivizes blockchain miners to deliberately ignore a specific transaction to undermine the punishment mechanism of PCNs. While previous studies have posited a prohibitive cost for such attacks, we show that this cost may be dramatically reduced (to approximately \$125), thereby increasing the likelihood of these attacks. To this end, we introduce Bribe & Fork, a modified bribing attack that leverages the threat of a so-called feather fork which we analyze with a novel formal model for the mining game with forking. We empirically analyze historical data of some real-world blockchain implementations to evaluate the scale of this cost reduction. Our findings shed more light on the potential vulnerability of PCNs and highlight the need for robust solutions.

The delicate equilibrium between user privacy and the ability to unleash the potential of distributed data is an important concern. Federated learning, which enables the training of collaborative models without sharing of data, has emerged as a privacy-centric solution. This approach brings forth security challenges, notably poisoning and backdoor attacks where malicious entities inject corrupted data into the training process, as well as evasion attacks that aim to induce misclassifications at test time. Our research investigates the intersection of adversarial training, a common defense method against evasion attacks, and backdoor attacks within federated learning. We introduce Adversarial Robustness Unhardening (ARU), which is employed by a subset of adversarial clients to intentionally undermine model robustness during federated training, rendering models susceptible to a broader range of evasion attacks. We present extensive experiments evaluating ARU's impact on adversarial training and existing robust aggregation defenses against poisoning and backdoor attacks. Our results show that ARU can substantially undermine adversarial training's ability to harden models against test-

We provide a game-theoretic analysis of the problem of front-running attacks. We use it to distinguish attacks from legitimate competition among honest users for having their transactions included earlier in the block. We also use it to introduce an intuitive notion of the severity of front-running attacks. We then study a simple commit-reveal protocol and discuss its properties. This protocol has costs because it requires two messages and imposes a delay. However, we show that it prevents the most severe front-running attacks while preserving legitimate competition between users, guaranteeing that the earliest transaction in a block belongs to the honest user who values it the most. When the protocol does not fully eliminate attacks, it nonetheless benefits honest users because it reduces competition among attackers (and overall expenditure by attackers).

This chapter introduces the concept of adversarial attacks on image classification models built on convolutional neural networks (CNN). CNNs are very popular deep-learning models which are used in image classification tasks. However, very powerful and pre-trained CNN models working very accurately on image datasets for image classification tasks may perform disastrously when the networks are under adversarial attacks. In this work, two very well-known adversarial attacks are discussed and their impact on the performance of image classifiers is analyzed. These two adversarial attacks are the fast gradient sign method (FGSM) and adversarial patch attack. These attacks are launched on three powerful pre-trained image classifier architectures, ResNet-34, GoogleNet, and DenseNet-161. The classification accuracy of the models in the absence and presence of the two attacks are computed on images from the publicly accessible ImageNet dataset. The results are analyzed to evaluate the impact of the attacks on the image classification task.

The widespread use of publicly available datasets for training machine learning models raises significant concerns about data misuse. Availability attacks have emerged as a means for data owners to safeguard their data by designing imperceptible perturbations that degrade model performance when incorporated into training datasets. However, existing availability attacks are ineffective when only a portion of the data can be perturbed. To address this challenge, we propose a novel availability attack approach termed Parameter Matching Attack (PMA). PMA is the first availability attack capable of causing more than a 30\% performance drop when only a portion of data can be perturbed. PMA optimizes perturbations so that when the model is trained on a mixture of clean and perturbed data, the resulting model will approach a model designed to perform poorly. Experimental results across four datasets demonstrate that PMA outperforms existing methods, achieving significant model performance degradation when a part of the training data is perturbed. Our code is available in the supplementary materials.

Watermarking approaches are widely used to identify if images being circulated are authentic or AI-generated. Determining the robustness of image watermarking methods in the ``no-box'' setting, where the attacker is assumed to have no knowledge about the watermarking model, is an interesting problem. Our main finding is that evading the no-box setting is challenging: the success of optimization-based transfer attacks (involving training surrogate models) proposed in prior work~\cite{hu2024transfer} depends on impractical assumptions, including (i) aligning the architecture and training configurations of both the victim and attacker's surrogate watermarking models, as well as (ii) a large number of surrogate models with potentially large computational requirements. Relaxing these assumptions i.e., moving to a more pragmatic threat model results in a failed attack, with an evasion rate at most $21.1\%$. We show that when the configuration is mostly aligned, a simple non-optimization attack we propose, OFT, with one single surrogate model can already exceed the success of optimization-based efforts. Under the same $\ell_\infty$ norm perturbation budget of $0.25$, prior work~\citet{hu2

We show that we can easily design a single adversarial perturbation $P$ that changes the class of $n$ images $X_1,X_2,\dots,X_n$ from their original, unperturbed classes $c_1, c_2,\dots,c_n$ to desired (not necessarily all the same) classes $c^*_1,c^*_2,\dots,c^*_n$ for up to hundreds of images and target classes at once. We call these \textit{multi-attacks}. Characterizing the maximum $n$ we can achieve under different conditions such as image resolution, we estimate the number of regions of high class confidence around a particular image in the space of pixels to be around $10^{\mathcal{O}(100)}$, posing a significant problem for exhaustive defense strategies. We show several immediate consequences of this: adversarial attacks that change the resulting class based on their intensity, and scale-independent adversarial examples. To demonstrate the redundancy and richness of class decision boundaries in the pixel space, we look for its two-dimensional sections that trace images and spell words using particular classes. We also show that ensembling reduces susceptibility to multi-attacks, and that classifiers trained on random labels are more susceptible. Our code is available on Git

Signature-based malware detectors have proven to be insufficient as even a small change in malignant executable code can bypass these signature-based detectors. Many machine learning-based models have been proposed to efficiently detect a wide variety of malware. Many of these models are found to be susceptible to adversarial attacks - attacks that work by generating intentionally designed inputs that can force these models to misclassify. Our work aims to explore vulnerabilities in the current state of the art malware detectors to adversarial attacks. We train a Transformers-based malware detector, carry out adversarial attacks resulting in a misclassification rate of 23.9% and propose defenses that reduce this misclassification rate to half. An implementation of our work can be found at https://github.com/yashjakhotiya/Adversarial-Attacks-On-Transformers.

Unlearnable example attacks are data poisoning techniques that can be used to safeguard public data against unauthorized use for training deep learning models. These methods add stealthy perturbations to the original image, thereby making it difficult for deep learning models to learn from these training data effectively. Current research suggests that adversarial training can, to a certain degree, mitigate the impact of unlearnable example attacks, while common data augmentation methods are not effective against such poisons. Adversarial training, however, demands considerable computational resources and can result in non-trivial accuracy loss. In this paper, we introduce the UEraser method, which outperforms current defenses against different types of state-of-the-art unlearnable example attacks through a combination of effective data augmentation policies and loss-maximizing adversarial augmentations. In stark contrast to the current SOTA adversarial training methods, UEraser uses adversarial augmentations, which extends beyond the confines of $ \ell_p $ perturbation budget assumed by current unlearning attacks and defenses. It also helps to improve the model's generalization ab

We apply formal methods to lay and streamline theoretical foundations to reason about Cyber-Physical Systems (CPSs) and physics-based attacks, i.e., attacks targeting physical devices. We focus on a formal treatment of both integrity and denial of service attacks to sensors and actuators of CPSs, and on the timing aspects of these attacks. Our contributions are fourfold. (1)~We define a hybrid process calculus to model both CPSs and physics-based attacks. (2)~We formalise a threat model that specifies MITM attacks that can manipulate sensor readings or control commands in order to drive a CPS into an undesired state, and we provide the means to assess attack tolerance/vulnerability with respect to a given attack. (3)~We formalise how to estimate the impact of a successful attack on a CPS and investigate possible quantifications of the success chances of an attack. (4)~We illustrate our definitions and results by formalising a non-trivial running example in Uppaal SMC, the statistical extension of the Uppaal model checker; we use Uppaal SMC as an automatic tool for carrying out a static security analysis of our running example in isolation and when exposed to three different physics

Quantum Cryptography uses the counter-intuitive properties of Quantum Mechanics for performing cryptographic tasks in a secure and reliable way. The Quantum Key Distribution (QKD) protocol BB84 has been proven secure against several important types of attacks: collective attacks and joint attacks. Here we analyze the security of a modified BB84 protocol, for which information is sent only in the z basis while testing is done in both the z and the x bases, against collective attacks. The proof follows the framework of a previous paper (Boyer, Gelles, and Mor, 2009), but it avoids a classical information-theoretical analysis and proves a fully composable security. We show that this modified BB84 protocol is as secure against collective attacks as the original BB84 protocol, and that it requires more bits for testing.