Fengbo Ma, Zixin Rao, Xiaoting Li, Zhetao Chen, Hongyue Sun, Yiping Zhao, Xianyan Chen, Zhen Xiang. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Zidi Xiong, Yuping Lin, Wenya Xie, Pengfei He, Zirui Liu, Jiliang Tang, Himabindu Lakkaraju, Zhen Xiang. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Unlike regular tokens derived from existing text corpora, special tokens are artificially created to annotate structured conversations during the fine-tuning process of Large Language Models (LLMs). Serving as metadata of training data, these tokens play a crucial role in instructing LLMs to generate coherent and context-aware responses. We demonstrate that special tokens can be exploited to construct four attack primitives, with which malicious users can reliably bypass the internal safety alignment of online LLM services and circumvent state-of-the-art (SOTA) external content moderation systems simultaneously. Moreover, we found that addressing this threat is challenging, as aggressive defense mechanisms-such as input sanitization by removing special tokens entirely, as suggested in academia-are less effective than anticipated. This is because such defense can be evaded when the special tokens are replaced by regular ones with high semantic similarity within the tokenizer's embedding space. We systemically evaluated our method, named MetaBreak, on both lab environment and commercial LLM platforms. Our approach achieves jailbreak rates comparable to SOTA prompt-engineering-based solutions when no content moderation is deployed. However, when there is content moderation, MetaBreak outperforms SOTA solutions PAP and GPTFuzzer by 11.6% and 34.8%, respectively. Finally, since MetaBreak employs a fundamentally different strategy from prompt engineering, the two approaches can work synergistically. Notably, empowering MetaBreak on PAP and GPTFuzzer boosts jailbreak rates by 24.3% and 20.2%, respectively.

Large Language Model (LLM) agents have become increasingly prevalent across various realworld applications.They enhance decisionmaking by storing private user-agent interactions in the memory module for demonstrations, introducing new privacy risks for LLM agents.In this work, we systematically investigate the vulnerability of LLM agents to our proposed Memory EXTRaction Attack (MEXTRA) under a black-box setting.To extract private information from memory, we propose an effective attacking prompt design and an automated prompt generation method based on different levels of knowledge about the LLM agent.Experiments on two representative agents demonstrate the effectiveness of MEXTRA.Moreover, we explore key factors influencing memory leakage from both the agent designer's and the attacker's perspectives.Our findings highlight the urgent need for effective memory safeguards in LLM agent design and deployment.

Multimodal foundation models (MMFMs) play a crucial role in various applications, including autonomous driving, healthcare, and virtual assistants. However, several studies have revealed vulnerabilities in these models, such as generating unsafe content by text-to-image models. Existing benchmarks on multimodal models either predominantly assess the helpfulness of these models, or only focus on limited perspectives such as fairness and privacy. In this paper, we present the first unified platform, MMDT (Multimodal DecodingTrust), designed to provide a comprehensive safety and trustworthiness evaluation for MMFMs. Our platform assesses models from multiple perspectives, including safety, hallucination, fairness/bias, privacy, adversarial robustness, and out-of-distribution (OOD) generalization. We have designed various evaluation scenarios and red teaming algorithms under different tasks for each perspective to generate challenging data, forming a high-quality benchmark. We evaluate a range of multimodal models using MMDT, and our findings reveal a series of vulnerabilities and areas for improvement across these perspectives. This work introduces the first comprehensive and unique safety and trustworthiness evaluation platform for MMFMs, paving the way for developing safer and more reliable MMFMs and systems. Our platform and benchmark are available at https://mmdecodingtrust.github.io/.

The rapid advancement of large language model (LLM) agents has raised new concerns regarding their safety and security. In this paper, we propose GuardAgent, the first guardrail agent to protect target agents by dynamically checking whether their actions satisfy given safety guard requests. Specifically, GuardAgent first analyzes the safety guard requests to generate a task plan, and then maps this plan into guardrail code for execution. By performing the code execution, GuardAgent can deterministically follow the safety guard request and safeguard target agents. In both steps, an LLM is utilized as the reasoning component, supplemented by in-context demonstrations retrieved from a memory module storing experiences from previous tasks. In addition, we propose two novel benchmarks: EICU-AC benchmark to assess the access control for healthcare agents and Mind2Web-SC benchmark to evaluate the safety policies for web agents. We show that GuardAgent effectively moderates the violation actions for different types of agents on these two benchmarks with over 98% and 83% guardrail accuracies, respectively. Project page: https://guardagent.github.io/

Agents powered by large language models (LLMs) have demonstrated strong capabilities in a wide range of complex, real-world applications. However, LLM agents with a compromised memory bank may easily produce harmful outputs when the past records retrieved for demonstration are malicious. In this paper, we propose a novel Memory INJection Attack, MINJA, without assuming that the attacker can directly modify the memory bank of the agent. The attacker injects malicious records into the memory bank by only **interacting with the agent via queries and output observations**. These malicious records are designed to elicit a sequence of malicious reasoning steps corresponding to a different target query during the agent's execution of the victim user's query. Specifically, we introduce a sequence of *bridging steps* to link victim queries to the malicious reasoning steps. During the memory injection, we propose an *indication prompt* that guides the agent to autonomously generate similar bridging steps, with a *progressive shortening strategy* that gradually removes the indication prompt, such that the malicious record will be easily retrieved when processing later victim queries. Our extensive experiments across diverse agents demonstrate the effectiveness of MINJA in compromising agent memory. With minimal requirements for execution, MINJA enables any user to influence agent memory, highlighting the risk.

Fengqing Jiang, Zhangchen Xu, Luyao Niu, Zhen Xiang, Bhaskar Ramasubramanian, Bo Li, Radha Poovendran. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.

Large language models (LLMs) are shown to benefit from chain-of-thought (COT) prompting, particularly when tackling tasks that require systematic reasoning processes. On the other hand, COT prompting also poses new vulnerabilities in the form of backdoor attacks, wherein the model will output unintended malicious content under specific backdoor-triggered conditions during inference. Traditional methods for launching backdoor attacks involve either contaminating the training dataset with backdoored instances or directly manipulating the model parameters during deployment. However, these approaches are not practical for commercial LLMs that typically operate via API access. In this paper, we propose BadChain, the first backdoor attack against LLMs employing COT prompting, which does not require access to the training dataset or model parameters and imposes low computational overhead. BadChain leverages the inherent reasoning capabilities of LLMs by inserting a backdoor reasoning step into the sequence of reasoning steps of the model output, thereby altering the final response when a backdoor trigger is embedded in the query prompt. In particular, a subset of demonstrations will be manipulated to incorporate a backdoor reasoning step in COT prompting. Consequently, given any query prompt containing the backdoor trigger, the LLM will be misled to output unintended content. Empirically, we show the effectiveness of BadChain for two COT strategies across four LLMs (Llama2, GPT-3.5, PaLM2, and GPT-4) and six complex benchmark tasks encompassing arithmetic, commonsense, and symbolic reasoning. We show that the baseline backdoor attacks designed for simpler tasks such as semantic classification will fail on these complicated tasks. In addition, our findings reveal that LLMs endowed with stronger reasoning capabilities exhibit higher susceptibility to BadChain, exemplified by a high average attack success rate of 97.0\% across the six benchmark tasks on GPT-4. We also demonstrate the interpretability of BadChain by showing that the relationship between the trigger and the backdoor reasoning step can be well-explained based on the output of the backdoored model. Finally, we propose two defenses based on shuffling and demonstrate their overall ineffectiveness against BadChain. Therefore, BadChain remains a severe threat to LLMs, underscoring the urgency for the development of robust and effective future defenses.

Recently, multi-agent collaborative (MAC) perception has been proposed and outperformed the traditional single-agent perception in many applications, such as autonomous driving. However, MAC perception is more vulnerable to adversarial attacks than single-agent perception due to the information exchange. The attacker can easily degrade the performance of a victim agent by sending harmful information from a malicious agent nearby. In this paper, we propose Malicious Agent Detection (MADE), a reactive defense specific to MAC perception that can be deployed by an agent to accurately detect and then remove any potential malicious agent in its local collaboration network. In particular, MADE inspects each agent in the network independently using a semi-supervised anomaly detector based on a double-hypothesis test with the Benjamini-Hochberg procedure for false positive control. For the two hypothesis tests, we propose a match loss statistic and a collaborative reconstruction loss statistic, respectively, both based on the consistency between the agent to be inspected and the ego agent deployed with our detector. We comprehensively evaluate MADE on a benchmark 3D dataset, V2X-sim, and a real-road dataset, DAIR-V2X, comparing it to baseline defenses. Notably, with the protection of MADE, the drops in the average precision compared with the best-case ‘Oracle’ defender are merely 1.27% and 0.28%, respectively.

Backdoor attacks aim to inject a backdoor into a classifier such that it predicts any input with an attacker-chosen backdoor trigger as an attacker-chosen target class. Existing backdoor attacks require either retraining the classifier with some clean data or modifying the model's architecture. As a result, they are 1) not applicable when clean data is unavailable, 2) less efficient when the model is large, and 3) less stealthy due to architecture changes. In this work, we propose DFBA, a novel retraining-free and data-free backdoor attack without changing the model architecture. Technically, our proposed method modifies a few parameters of a classifier to inject a backdoor. Through theoretical analysis, we verify that our injected backdoor is provably undetectable and unremovable by various state-of-the-art defenses under mild assumptions. Our evaluation on multiple datasets further demonstrates that our injected backdoor: 1) incurs negligible classification loss, 2) achieves 100\% attack success rates, and 3) bypasses six existing state-of-the-art defenses. Moreover, our comparison with a state-of-the-art non-data-free backdoor attack shows our attack is more stealthy and effective against various defenses while achieving less classification accuracy loss. We will release our code upon paper acceptance.

LLM agents have demonstrated remarkable performance across various applications, primarily due to their advanced capabilities in reasoning, utilizing external knowledge and tools, calling APIs, and executing actions to interact with environments. Current agents typically utilize a memory module or a retrieval-augmented generation (RAG) mechanism, retrieving past knowledge and instances with similar embeddings from knowledge bases to inform task planning and execution. However, the reliance on unverified knowledge bases raises significant concerns about their safety and trustworthiness. To uncover such vulnerabilities, we propose a novel red teaming approach AgentPoison, the first backdoor attack targeting generic and RAG-based LLM agents by poisoning their long-term memory or RAG knowledge base. In particular, we form the trigger generation process as a constrained optimization to optimize backdoor triggers by mapping the triggered instances to a unique embedding space, so as to ensure that whenever a user instruction contains the optimized backdoor trigger, the malicious demonstrations are retrieved from the poisoned memory or knowledge base with high probability. In the meantime, benign instructions without the trigger will still maintain normal performance. Unlike conventional backdoor attacks, AgentPoison requires no additional model training or fine-tuning, and the optimized backdoor trigger exhibits superior transferability, resilience, and stealthiness. Extensive experiments demonstrate AgentPoison's effectiveness in attacking three types of real-world LLM agents: RAG-based autonomous driving agent, knowledge-intensive QA agent, and healthcare EHRAgent. We inject the poisoning instances into the RAG knowledge base and long-term memories of these agents, respectively, demonstrating the generalization of AgentPoison. On each agent, AgentPoison achieves an average attack success rate of $\ge$ 80% with minimal impact on benign performance ($\le$ 1%) with a poison rate < 0.1%. The code and data is available at https://github.com/BillChan226/AgentPoison.

Backdoor attacks are an important type of adversarial threat against deep neural network classifiers, wherein test samples from one or more source classes will be (mis)classified to the attacker’s target class when a backdoor pattern is embedded. In this paper, we focus on the post-training backdoor defense scenario commonly considered in the literature, where the defender aims to detect whether a trained classifier was backdoor-attacked without any access to the training set. Many post-training detectors are designed to detect attacks that use either one or a few specific backdoor embedding functions (e.g., patch-replacement or additive attacks). These detectors may fail when the backdoor embedding function used by the attacker (unknown to the defender) is different from the backdoor embedding function assumed by the defender. In contrast, we propose a post-training defense that detects backdoor attacks with arbitrary types of backdoor embeddings, without making any assumptions about the backdoor embedding type. Our detector leverages the influence of the backdoor attack, independent of the backdoor embedding mechanism, on the landscape of the classifier’s outputs prior to the softmax layer. For each class, a maximum margin statistic is estimated. Detection inference is then performed by applying an unsupervised anomaly detector to these statistics. Thus, our detector does not need any legitimate clean samples, and can efficiently detect backdoor attacks with arbitrary numbers of source classes. These advantages over several state-of-the-art methods are demonstrated on four datasets, for three different types of backdoor patterns, and for a variety of attack configurations. Finally, we propose a novel, general approach for backdoor mitigation once a detection is made. The mitigation approach was the runner-up at the first IEEE Trojan Removal Competition. The code is online available.

Data Poisoning (DP) is an effective attack that causes trained classifiers to misclassify their inputs. DP attacks significantly degrade a classifier's accuracy by covertly injecting attack samples into the training set. Broadly applicable to different classifier structures, without strong assumptions about the attacker, anunsupervisedBayesian Information Criterion (BIC)-based mixture model defense against “error generic” DP attacks is herein proposed that: 1) addresses the most challengingembeddedDP scenario wherein, if DP is present, the poisoned samples are ana prioriunknown subset of the training set, and with no clean validation set available; 2) applies a mixture model both to well-fit potentially multi-modal class distributions and to capture poisoned samples within a small subset of the mixture components; 3) jointly identifies poisoned components and samples by minimizing the BIC cost defined over the whole training set, with the identified poisoned data removed prior to classifier training. Our experimental results, for various classifier structures and benchmark datasets, demonstrate the effectiveness of our defense under strong DP attacks, as well as its superiority over other DP defenses.

A backdoor or Trojan attack is an important type of data poisoning attack against deep neural network (DNN) classifiers, wherein the training dataset is poisoned with a small number of samples that each possess the backdoor pattern (usually a pattern that is either imperceptible or innocuous) and which are mislabeled to the attacker’s target class. When trained on a backdoor-poisoned dataset, a DNN behaves normally on most benign test samples but makes incorrect predictions to the target class when the test sample has the backdoor pattern incorporated (i.e., contains a backdoor trigger). Here we focus on image classification tasks and show that supervised training may build stronger association between the backdoor pattern and the associated target class than that between normal features and the true class of origin. By contrast, self-supervised representation learning ignores the labels of samples and learns a feature embedding based on images’ semantic content. Using a feature embedding found by self-supervised representation learning, a data cleansing method, which combines sample filtering and relabeling, is developed. Experiments on CIFAR-10 benchmark datasets show that our method achieves state-of-the-art performance in mitigating backdoor attacks.

Backdoor (Trojan) attack is a common threat to deep neural networks, where samples from one or more source classes embedded with a backdoor trigger will be misclassified to adversarial target classes. Existing methods for detecting whether a classifier is backdoor attacked are mostly designed for attacks with a single adversarial target (e.g., all-to-one attack). To the best of our knowledge, without supervision, no existing methods can effectively address the more general X2X attack with an arbitrary number of source classes, each paired with an arbitrary target class. In this paper, we propose UMD, the first Unsupervised Model Detection method that effectively detects X2X backdoor attacks via a joint inference of the adversarial (source, target) class pairs. In particular, we first define a novel transferability statistic to measure and select a subset of putative backdoor class pairs based on a proposed clustering approach. Then, these selected class pairs are jointly assessed based on an aggregation of their reverse-engineered trigger size for detection inference, using a robust and unsupervised anomaly detector we proposed. We conduct comprehensive evaluations on CIFAR-10, GTSRB, and Imagenette dataset, and show that our unsupervised UMD outperforms SOTA detectors (even with supervision) by 17%, 4%, and 8%, respectively, in terms of the detection accuracy against diverse X2X attacks. We also show the strong detection performance of UMD against several strong adaptive attacks.

Backdoor attack is a common threat to deep neural networks. During testing, samples embedded with a backdoor trigger will be misclassified as an adversarial target by a backdoored model, while samples without the backdoor trigger will be correctly classified. In this paper, we present the first certified backdoor detector (CBD), which is based on a novel, adjustable conformal prediction scheme based on our proposed statistic local dominant probability. For any classifier under inspection, CBD provides 1) a detection inference, 2) the condition under which the attacks are guaranteed to be detectable for the same classification domain, and 3) a probabilistic upper bound for the false positive rate. Our theoretical results show that attacks with triggers that are more resilient to test-time noise and have smaller perturbation magnitudes are more likely to be detected with guarantees. Moreover, we conduct extensive experiments on four benchmark datasets considering various backdoor types, such as BadNet, CB, and Blend. CBD achieves comparable or even higher detection accuracy than state-of-the-art detectors, and it in addition provides detection certification. Notably, for backdoor attacks with random perturbation triggers bounded by $\ell_2\leq0.75$ which achieves more than 90\% attack success rate, CBD achieves 100\% (98\%), 100\% (84\%), 98\% (98\%), and 72\% (40\%) empirical (certified) detection true positive rates on the four benchmark datasets GTSRB, SVHN, CIFAR-10, and TinyImageNet, respectively, with low false positive rates.

Backdoor (Trojan) attacks are emerging threats against deep neural networks (DNN). A DNN being attacked will predict to an attacker-desired target class whenever a test sample from any source class is embedded with a backdoor pattern, while correctly classifying clean (attack-free) test samples. Existing backdoor defenses have shown success in detecting whether a DNN is attacked and in reverse-engineering the backdoor pattern in a "post-training" scenario: the defender has access to the DNN to be inspected and a small, clean dataset collected independently, but has no access to the (possibly poisoned) training set of the DNN. However, these defenses neither catch culprits in the act of triggering the backdoor mapping, nor mitigate the backdoor attack at test-time. In this paper, we propose an "in-flight" unsupervised defense against backdoor attacks on image classification that 1) detects use of a backdoor trigger at test-time; and 2) infers the class of origin (source class) for a detected trigger example. The effectiveness of our defense is demonstrated experimentally for a wide variety of DNN architectures, datasets, and backdoor attack configurations.

Backdoor attacks (BA) are an emerging threat to deep neural network classifiers. A classifier being attacked will predict to the attacker’s target class when a test sample from a source class is embedded with the backdoor pattern (BP). Recently, the first BA against point cloud (PC) classifiers was proposed, creating new threats to many important applications including autonomous driving. Such PC BAs are not detectable by existing BA defenses due to their special BP embedding mechanism. In this paper, we propose a reverse-engineering defense that infers whether a PC classifier is backdoor attacked, without access to its training set or to any clean classifiers for reference. The effectiveness of our defense is demonstrated on the benchmark ModeNet40 dataset for PCs.

This paper is concerned with distributed Nash equilibrium seeking problems involving double-integrator-type players who do not have knowledge on their own velocity signals. Moreover, the control inputs of players are considered to be affected by time-varying disturbances. To solve this problem, an uncertainty and disturbance estimator (UDE) is adopted to construct a distributed robust Nash equilibrium seeker. By our theoretical analysis, it is found that players' actions can be driven to be arbitrarily close to the Nash equilibrium by utilizing the developed method. In addition, for constant and exponentially decaying disturbances, asymptotic results can be derived. Finally, a simulation example is provided to illustrate the presented theoretical results.

With wide deployment of deep neural network (DNN) classifiers, there is great potential for harm from adversarial learning attacks. Recently, a special type of data poisoning (DP) attack, known as a backdoor (or Trojan), was proposed. These attacks do not seek to degrade classification accuracy, but rather to have the classifier learn to classify to a target class$t^{\ast }$whenever the backdoor pattern is present in a test example originally from a source class$s^{\ast }$. Launching backdoor attacks does not require knowledge of the classifier or its training process—only the ability to poison the training set with exemplars containing a backdoor pattern (labeled with the target class). Defenses against backdoors can be deployed before/during training, post-training, or at test time. Here, we address post-training detection in DNN image classifiers, seldom considered in existing works, whereinthe defender does not have access to the poisoned training set, but only to the trained classifier itself, as well as to clean (unpoisoned) examples from the classification domain. This scenario is of great interest because e.g., a classifier may be the basis of a phone app that will be shared with many users. Detection may thus reveal a widespread attack. We propose a purely unsupervised anomaly detection (AD) defense against imperceptible backdoor attacks that: 1) detects whether the trained DNN has been backdoor-attacked; 2) infers the source and target classes in a detected attack; 3) estimates the backdoor pattern itself. Our AD approach involves learning (via suitable cost function minimization) the minimum size/norm perturbation (putative backdoor) required to induce the classifier to misclassify (most) examples from class$s$to class$t$, for all$(s,t)$pairs. Our hypothesis is that nonattacked pairs require large perturbations, while the attacked pair$(s^{\ast }, t^{\ast })$requires much smaller ones. This is convincingly borne out experimentally. We identify a variety of plausible cost functions and devise a novel, robust hypothesis testing approach to perform detection inference. We test our approach, in comparison with the state-of-the-art methods, for several backdoor patterns, attack settings and mechanisms, and data sets and demonstrate its favorability. Our defense essentially requires setting a single hyperparameter (the detection threshold), which can e.g., be chosen to fix the system’s false positive rate.

Backdoor attacks (BAs) are an emerging form of adversarial attack typically against deep neural network image classifiers. The attacker aims to have the classifier learn to classify to a target class when test images from one or more source classes contain a backdoor pattern, while maintaining high accuracy on all clean test images. Reverse-Engineering-based Defenses (REDs) against BAs do not require access to the training set but only to an independent clean dataset. Unfortunately, most existing REDs rely on an unrealistic assumption that all classes except the target class are source classes of the attack. REDs that do not rely on this assumption often require a large set of clean images and heavy computation. In this paper, we propose a Lagrangian-based RED (L-RED) that does not require knowledge of the number of source classes (or whether an attack is present). Our defense requires very few clean images to effectively detect BAs and is computationally efficient. Notably, we detect 56 out of 60 BAs using only two clean images per class in our experiments on CIFAR-10.

Vulnerability of 3D point cloud (PC) classifiers has become a grave concern due to the popularity of 3D sensors in safety-critical applications. Existing adversarial attacks against 3D PC classifiers are all test-time evasion (TTE) attacks that aim to induce test-time misclassifications using knowledge of the classifier. But since the victim classifier is usually not accessible to the attacker, the threat is largely diminished in practice, as PC TTEs typically have poor transferability. Here, we propose the first backdoor attack (BA) against PC classifiers. Originally proposed for images, BAs poison the victim classifier’s training set so that the classifier learns to decide to the attacker’s target class whenever the attacker’s backdoor pattern is present in a given input sample. Significantly, BAs do not require knowledge of the victim classifier. Different from image BAs, we propose to insert a cluster of points into a PC as a robust backdoor pattern customized for 3D PCs. Such clusters are also consistent with a physical attack (i.e., with a captured object in a scene). We optimize the cluster’s location using an independently trained surrogate classifier and choose the cluster’s local geometry to evade possible PC preprocessing and PC anomaly detectors (ADs). Experimentally, our BA achieves a uniformly high success rate (≥ 87%) and shows evasiveness against state-of-the-art PC ADs. Code is available at https://github.com/zhenxianglance/PCBA.

Backdoor data poisoning attacks add mislabeled examples to the training set, with an embedded backdoor pattern, so that the classifier learns to classify to a target class whenever the backdoor pattern is present in a test sample. Here, we address posttraining detection of scene-plausible perceptible backdoors, a type of backdoor attack that can be relatively easily fashioned, particularly against DNN image classifiers. A post-training defender does not have access to the potentially poisoned training set, only to the trained classifier, as well as some unpoisoned examples that need not be training samples. Without the poisoned training set, the only information about a backdoor pattern is encoded in the DNN's trained weights. This detection scenario is of great import considering legacy and proprietary systems, cell phone apps, as well as training outsourcing, where the user of the classifier will not have access to the entire training set. We identify two important properties of scene-plausible perceptible backdoor patterns, spatial invariance and robustness, based on which we propose a novel detector using the maximum achievable misclassification fraction (MAMF) statistic. We detect whether the trained DNN has been backdoor-attacked and infer the source and target classes. Our detector outperforms existing detectors and, coupled with an imperceptible backdoor detector, helps achieve posttraining detection of most evasive backdoors of interest.

Recently, a special type of data poisoning (DP) attack against deep neural network (DNN) classifiers, known as a backdoor, was proposed. These attacks do not seek to degrade classification accuracy, but rather to have the classifier learn to classify to a target class whenever the backdoor pattern is present in a test example. Here, we address the challenging post-training detection of backdoor attacks in DNN image classifiers, wherein the defender does not have access to the poisoned training set, but only to the trained classifier itself, as well as to clean (unpoisoned) examples from the classification domain. We propose a defense against imperceptible backdoor attacks based on perturbation optimization and novel, robust detection inference. Our method detects whether the trained DNN has been backdoor-attacked and infers the source and target classes involved in an attack. It outperforms alternative defenses for several backdoor patterns, data sets, and attack settings.

Computing systems have been facing severe technology challenges in recent years with regard to power consumption, circuit reliability, and high performance. For many years, the issues of power consumption and performance have been addressed with the use of technology scaling.However, as Dennard’s scaling tends toward an end, it has become difficult to further improve the performance under the same power constraints. In addition to power, reliability also becomes a critical issue when the feature size of the complementary metal-oxide–semiconductor (CMOS) technology is reduced below 7 nm. Thus, ensuring the complete accuracy of the signal has become increasingly challenging in recent years.

With wide deployment of machine learning (ML)-based systems for a variety of applications including medical, military, automotive, genomic, multimedia, and social networking, there is great potential for damage from adversarial learning (AL) attacks. In this article, we provide a contemporary survey of AL, focused particularly on defenses against attacks on deep neural network classifiers. After introducing relevant terminology and the goals and range of possible knowledge of both attackers and defenders, we survey recent work on test-time evasion (TTE), data poisoning (DP), backdoor DP, and reverse engineering (RE) attacks and particularly defenses against the same. In so doing, we distinguish robust classification from anomaly detection (AD), unsupervised from supervised, and statistical hypothesis-based defenses from ones that do not have an explicit null (no attack) hypothesis. We also consider several scenarios for detecting backdoors. We provide a technical assessment for reviewed works, including identifying any issues/limitations, required hyperparameters, needed computational complexity, as well as the performance measures evaluated and the obtained quality. We then delve deeper, providing novel insights that challenge conventional AL wisdom and that target unresolved issues, including: robust classification versus AD as a defense strategy; the belief that attack success increases with attack strength, which ignores susceptibility to AD; small perturbations for TTE attacks: a fallacy or a requirement; validity of the universal assumption that a TTE attacker knows the ground-truth class for the example to be attacked; black, gray, or white-box attacks as the standard for defense evaluation; and susceptibility of query-based RE to an AD defense. We also discuss attacks on the privacy of training data. We then present benchmark comparisons of several defenses against TTE, RE, and backdoor DP attacks on images. The article concludes with a discussion of continuing research directions, including the supreme challenge of detecting attacks whose goal is not to alter classification decisions, but rather simply to embed, without detection, “fake news” or other false content.