Sheng Wen

Wanlun Ma, Derui Wang, Ruoxi Sun et al.

Deep Neural Networks (DNNs) are susceptible to backdoor attacks during training. The model corrupted in this way functions normally, but when triggered by certain patterns in the input, produces a predefined target label. Existing defenses usually rely on the assumption of the universal backdoor setting in which poisoned samples share the same uniform trigger. However, recent advanced backdoor attacks show that this assumption is no longer valid in dynamic backdoors where the triggers vary from input to input, thereby defeating the existing defenses. In this work, we propose a novel technique, Beatrix (backdoor detection via Gram matrix). Beatrix utilizes Gram matrix to capture not only the feature correlations but also the appropriately high-order information of the representations. By learning class-conditional statistics from activation patterns of normal samples, Beatrix can identify poisoned samples by capturing the anomalies in activation patterns. To further improve the performance in identifying target labels, Beatrix leverages kernel-based testing without making any prior assumptions on representation distribution. We demonstrate the effectiveness of our method through extensive evaluation and comparison with state-of-the-art defensive techniques. The experimental results show that our approach achieves an F1 score of 91.1% in detecting dynamic backdoors, while the state of the art can only reach 36.9%.

Zhong Guan, Haoran Sun, Yongjian Guo et al.

In recent years, Vision-Language-Action (VLA) models have emerged as a crucial pathway towards general embodied intelligence, yet their training efficiency has become a key bottleneck. Although existing reinforcement learning (RL)-based training frameworks like RLinf can enhance model generalization, they still rely on synchronous execution, leading to severe resource underutilization and throughput limitations during environment interaction, policy generation (rollout), and model update phases (actor). To overcome this challenge, this paper, for the first time, proposes and implements a fully-asynchronous policy training framework encompassing the entire pipeline from environment interaction, rollout generation, to actor policy updates. Systematically drawing inspiration from asynchronous optimization ideas in large model RL, our framework designs a multi-level decoupled architecture. This includes asynchronous parallelization of environment interaction and trajectory collection, streaming execution for policy generation, and decoupled scheduling for training updates. We validated the effectiveness of our method across diverse VLA models and environments. On the LIBERO benchmark, the framework achieves throughput improvements of up to 59.25\% compared to existing synchronous strategies. When deeply optimizing separation strategies, throughput can be increased by as much as 126.67\%. We verified the effectiveness of each asynchronous component via ablation studies. Scaling law validation across 8 to 256 GPUs demonstrates our method's excellent scalability under most conditions.

Yuxin Cao, Yumeng Zhu, Derui Wang et al.

Face recognition pipelines have been widely deployed in various mission-critical systems in trust, equitable and responsible AI applications. However, the emergence of adversarial attacks has threatened the security of the entire recognition pipeline. Despite the sheer number of attack methods proposed for crafting adversarial examples in both digital and physical forms, it is never an easy task to assess the real threat level of different attacks and obtain useful insight into the key risks confronted by face recognition systems. Traditional attacks view imperceptibility as the most important measurement to keep perturbations stealthy, while we suspect that industry professionals may possess a different opinion. In this paper, we delve into measuring the threat brought about by adversarial attacks from the perspectives of the industry and the applications of face recognition. In contrast to widely studied sophisticated attacks in the field, we propose an effective yet easy-to-launch physical adversarial attack, named AdvColor, against black-box face recognition pipelines in the physical world. AdvColor fools models in the recognition pipeline via directly supplying printed photos of human faces to the system under adversarial illuminations. Experimental results show that physical AdvColor examples can achieve a fooling rate of more than 96% against the anti-spoofing model and an overall attack success rate of 88% against the face recognition pipeline. We also conduct a survey on the threats of prevailing adversarial attacks, including AdvColor, to understand the gap between the machine-measured and human-assessed threat levels of different forms of adversarial attacks. The survey results surprisingly indicate that, compared to deliberately launched imperceptible attacks, perceptible but accessible attacks pose more lethal threats to real-world commercial systems of face recognition.

Duoxun Tang, Yuxin Cao, Xi Xiao et al.

Video classification systems based on Deep Neural Networks (DNNs) have demonstrated excellent performance in accurately verifying video content. However, recent studies have shown that DNNs are highly vulnerable to adversarial examples. Therefore, a deep understanding of adversarial attacks can better respond to emergency situations. In order to improve attack performance, many style-transfer-based attacks and patch-based attacks have been proposed. However, the global perturbation of the former will bring unnatural global color, while the latter is difficult to achieve success in targeted attacks due to the limited perturbation space. Moreover, compared to a plethora of methods targeting image classifiers, video adversarial attacks are still not that popular. Therefore, to generate adversarial examples with a low budget and to provide them with a higher verisimilitude, we propose a novel black-box video attack framework, called Stylized Logo Attack (SLA). SLA is conducted through three steps. The first step involves building a style references set for logos, which can not only make the generated examples more natural, but also carry more target class features in the targeted attacks. Then, reinforcement learning (RL) is employed to determine the style reference and position parameters of the logo within the video, which ensures that the stylized logo is placed in the video with optimal attributes. Finally, perturbation optimization is designed to optimize perturbations to improve the fooling rate in a step-by-step manner. Sufficient experimental results indicate that, SLA can achieve better performance than state-of-the-art methods and still maintain good deception effects when facing various defense methods.

Jiaxuan Gao, Yongjian Guo, Zhong Guan et al.

The integration of Vision-Language-Action (VLA) models with World Models has gained increasing attention. One representative approach treats learned World Models as generative simulators, enabling policy optimization entirely within "imagination." However, when deployed as simulators for specific environments such as the LIBERO benchmark, existing World Models often suffer from poor generalization and long-horizon error accumulation. During closed-loop rollouts, these models are highly sensitive to initial-state perturbations; minor changes in color, illumination, and other visual factors can trigger cascading hallucinations, leading to severe blurriness or overexposure. Moreover, long-horizon error accumulation further degrades the quality and fidelity of predicted future states. These issues limit the reliability of World Models as simulators. To mitigate these problems, we propose Sword, a robust World Model framework. Our method introduces Structure-Guided Style Augmentation to disentangle the visual textures of interactive environments from task-relevant dynamics, thereby improving generalization. We further propose Dynamic Latent Bootstrapping, which maintains consistency between training and inference while keeping memory consumption low. Extensive experiments on the LIBERO benchmark show that our method significantly outperforms the baseline WoVR in terms of generalization, generation quality, robustness, fidelity, and the success rate of reinforcement-learning post-training for VLA models.

Yongjian Guo, Puzhuo Liu, Wanlun Ma et al.

The Model Context Protocol (MCP) has emerged as a universal standard that enables AI agents to seamlessly connect with external tools, significantly enhancing their functionality. However, while MCP brings notable benefits, it also introduces significant vulnerabilities, such as Tool Poisoning Attacks (TPA), where hidden malicious instructions exploit the sycophancy of large language models (LLMs) to manipulate agent behavior. Despite these risks, current academic research on MCP security remains limited, with most studies focusing on narrow or qualitative analyses that fail to capture the diversity of real-world threats. To address this gap, we present the MCP Attack Library (MCPLIB), which categorizes and implements 31 distinct attack methods under four key classifications: direct tool injection, indirect tool injection, malicious user attacks, and LLM inherent attack. We further conduct a quantitative analysis of the efficacy of each attack. Our experiments reveal key insights into MCP vulnerabilities, including agents' blind reliance on tool descriptions, sensitivity to file-based attacks, chain attacks exploiting shared context, and difficulty distinguishing external data from executable commands. These insights, validated through attack experiments, underscore the urgency for robust defense strategies and informed MCP design. Our contributions include 1) constructing a comprehensive MCP attack taxonomy, 2) introducing a unified attack framework MCPLIB, and 3) conducting empirical vulnerability analysis to enhance MCP security mechanisms. This work provides a foundational framework, supporting the secure evolution of MCP ecosystems.

Weijia Yang, Tian Lan, Leyuan Liu et al.

The rapid evolution of digital currency trading, fueled by the integration of blockchain technology, has led to both innovation and the emergence of smart Ponzi schemes. A smart Ponzi scheme is a fraudulent investment operation in smart contract that uses funds from new investors to pay returns to earlier investors. Traditional Ponzi scheme detection methods based on deep learning typically rely on fully supervised models, which require large amounts of labeled data. However, such data is often scarce, hindering effective model training. To address this challenge, we propose a novel contrastive learning framework, CASPER (Contrastive Approach for Smart Ponzi detectER with more negative samples), designed to enhance smart Ponzi scheme detection in blockchain transactions. By leveraging contrastive learning techniques, CASPER can learn more effective representations of smart contract source code using unlabeled datasets, significantly reducing both operational costs and system complexity. We evaluate CASPER on the XBlock dataset, where it outperforms the baseline by 2.3% in F1 score when trained with 100% labeled data. More impressively, with only 25% labeled data, CASPER achieves an F1 score nearly 20% higher than the baseline under identical experimental conditions. These results highlight CASPER's potential for effective and cost-efficient detection of smart Ponzi schemes, paving the way for scalable fraud detection solutions in the future.

Zehang Deng, Yongjian Guo, Changzhou Han et al.

An Artificial Intelligence (AI) agent is a software entity that autonomously performs tasks or makes decisions based on pre-defined objectives and data inputs. AI agents, capable of perceiving user inputs, reasoning and planning tasks, and executing actions, have seen remarkable advancements in algorithm development and task performance. However, the security challenges they pose remain under-explored and unresolved. This survey delves into the emerging security threats faced by AI agents, categorizing them into four critical knowledge gaps: unpredictability of multi-step user inputs, complexity in internal executions, variability of operational environments, and interactions with untrusted external entities. By systematically reviewing these threats, this paper highlights both the progress made and the existing limitations in safeguarding AI agents. The insights provided aim to inspire further research into addressing the security threats associated with AI agents, thereby fostering the development of more robust and secure AI agent applications.

Yuxin Cao, Xi Xiao, Ruoxi Sun et al.

Video classification systems are vulnerable to adversarial attacks, which can create severe security problems in video verification. Current black-box attacks need a large number of queries to succeed, resulting in high computational overhead in the process of attack. On the other hand, attacks with restricted perturbations are ineffective against defenses such as denoising or adversarial training. In this paper, we focus on unrestricted perturbations and propose StyleFool, a black-box video adversarial attack via style transfer to fool the video classification system. StyleFool first utilizes color theme proximity to select the best style image, which helps avoid unnatural details in the stylized videos. Meanwhile, the target class confidence is additionally considered in targeted attacks to influence the output distribution of the classifier by moving the stylized video closer to or even across the decision boundary. A gradient-free method is then employed to further optimize the adversarial perturbations. We carry out extensive experiments to evaluate StyleFool on two standard datasets, UCF-101 and HMDB-51. The experimental results demonstrate that StyleFool outperforms the state-of-the-art adversarial attacks in terms of both the number of queries and the robustness against existing defenses. Moreover, 50% of the stylized videos in untargeted attacks do not need any query since they can already fool the video classification model. Furthermore, we evaluate the indistinguishability through a user study to show that the adversarial samples of StyleFool look imperceptible to human eyes, despite unrestricted perturbations.

Xiaoyu Sun, Xiao Chen, Kui Liu et al.

While extremely valuable to achieve advanced functions, mobile phone sensors can be abused by attackers to implement malicious activities in Android apps, as experimentally demonstrated by many state-of-the-art studies. There is hence a strong need to regulate the usage of mobile sensors so as to keep them from being exploited by malicious attackers. However, despite the fact that various efforts have been put in achieving this, i.e., detecting privacy leaks in Android apps, we have not yet found approaches to automatically detect sensor leaks in Android apps. To fill the gap, we designed and implemented a novel prototype tool, SEEKER, that extends the famous FlowDroid tool to detect sensor-based data leaks in Android apps. SEEKER conducts sensor-focused static taint analyses directly on the Android apps' bytecode and reports not only sensor-triggered privacy leaks but also the sensor types involved in the leaks. Experimental results using over 40,000 real-world Android apps show that SEEKER is effective in detecting sensor leaks in Android apps, and malicious apps are more interested in leaking sensor data than benign apps.

Kunpeng Zhang, Xi Xiao, Xiaogang Zhu et al.

Coverage-guided Greybox Fuzzing (CGF) is one of the most successful and widely-used techniques for bug hunting. Two major approaches are adopted to optimize CGF: (i) to reduce search space of inputs by inferring relationships between input bytes and path constraints; (ii) to formulate fuzzing processes (e.g., path transitions) and build up probability distributions to optimize power schedules, i.e., the number of inputs generated per seed. However, the former is subjective to the inference results which may include extra bytes for a path constraint, thereby limiting the efficiency of path constraints resolution, code coverage discovery, and bugs exposure; the latter formalization, concentrating on power schedules for seeds alone, is inattentive to the schedule for bytes in a seed. In this paper, we propose a lightweight fuzzing approach, Truzz, to optimize existing Coverage-guided Greybox Fuzzers (CGFs). To address two aforementioned challenges, Truzz identifies the bytes related to the validation checks (i.e., the checks guarding error-handling code), and protects those bytes from being frequently mutated, making most generated inputs examine the functionalities of programs, in lieu of being rejected by validation checks. The byte-wise relationship determination mitigates the problem of loading extra bytes when fuzzers infer the byte-constraint relation. Furthermore, the proposed path transition within Truzz can efficiently prioritize the seed as the new path, harvesting many new edges, and the new path likely belongs to a code region with many undiscovered code lines. The experimental results show that on average, Truzz can generate 16.14% more inputs flowing into functional code, in addition to 24.75% more new edges than the vanilla fuzzers. Finally, our approach exposes 13 bugs in 8 target programs, and 6 of them have not been identified by the vanilla fuzzers.

Chaoran Li, Xiao Chen, Ruoxi Sun et al.

The Android system manages access to sensitive APIs by permission enforcement. An application (app) must declare proper permissions before invoking specific Android APIs. However, there is no official documentation providing the complete list of permission-protected APIs and the corresponding permissions to date. Researchers have spent significant efforts extracting such API protection mapping from the Android API framework, which leverages static code analysis to determine if specific permissions are required before accessing an API. Nevertheless, none of them has attempted to analyze the protection mapping in the native library (i.e., code written in C and C++), an essential component of the Android framework that handles communication with the lower-level hardware, such as cameras and sensors. While the protection mapping can be utilized to detect various security vulnerabilities in Android apps, such as permission over-privilege and component hijacking, imprecise mapping will lead to false results in detecting such security vulnerabilities. To fill this gap, we develop a prototype system, named NatiDroid, to facilitate the cross-language static analysis to benchmark against two state-of-the-art tools, termed Axplorer and Arcade. We evaluate NatiDroid on more than 11,000 Android apps, including system apps from custom Android ROMs and third-party apps from the Google Play. Our NatiDroid can identify up to 464 new API-permission mappings, in contrast to the worst-case results derived from both Axplorer and Arcade, where approximately 71% apps have at least one false positive in permission over-privilege and up to 3.6% apps have at least one false negative in component hijacking. Additionally, we identify that 24 components with at least one Native-triggered component hijacking vulnerability are misidentified by two benchmarks.

Xiaotao Feng, Ruoxi Sun, Xiaogang Zhu et al.

The proliferation of Internet of Things (IoT) devices has made people's lives more convenient, but it has also raised many security concerns. Due to the difficulty of obtaining and emulating IoT firmware, the black-box fuzzing of IoT devices has become a viable option. However, existing black-box fuzzers cannot form effective mutation optimization mechanisms to guide their testing processes, mainly due to the lack of feedback. It is difficult or even impossible to apply existing grammar-based fuzzing strategies. Therefore, an efficient fuzzing approach with syntax inference is required in the IoT fuzzing domain. To address these critical problems, we propose a novel automatic black-box fuzzing for IoT firmware, termed Snipuzz. Snipuzz runs as a client communicating with the devices and infers message snippets for mutation based on the responses. Each snippet refers to a block of consecutive bytes that reflect the approximate code coverage in fuzzing. This mutation strategy based on message snippets considerably narrows down the search space to change the probing messages. We compared Snipuzz with four state-of-the-art IoT fuzzing approaches, i.e., IoTFuzzer, BooFuzz, Doona, and Nemesys. Snipuzz not only inherits the advantages of app-based fuzzing (e.g., IoTFuzzer, but also utilizes communication responses to perform efficient mutation. Furthermore, Snipuzz is lightweight as its execution does not rely on any prerequisite operations, such as reverse engineering of apps. We also evaluated Snipuzz on 20 popular real-world IoT devices. Our results show that Snipuzz could identify 5 zero-day vulnerabilities, and 3 of them could be exposed only by Snipuzz. All the newly discovered vulnerabilities have been confirmed by their vendors.

Xiaogang Zhu, Shigang Liu, Xian Li et al.

Fuzzing is one of the most effective technique to identify potential software vulnerabilities. Most of the fuzzers aim to improve the code coverage, and there is lack of directedness (e.g., fuzz the specified path in a software). In this paper, we proposed a deep learning (DL) guided directed fuzzing for software vulnerability detection, named DeFuzz. DeFuzz includes two main schemes: (1) we employ a pre-trained DL prediction model to identify the potentially vulnerable functions and the locations (i.e., vulnerable addresses). Precisely, we employ Bidirectional-LSTM (BiLSTM) to identify attention words, and the vulnerabilities are associated with these attention words in functions. (2) then we employ directly fuzzing to fuzz the potential vulnerabilities by generating inputs that tend to arrive the predicted locations. To evaluate the effectiveness and practical of the proposed DeFuzz technique, we have conducted experiments on real-world data sets. Experimental results show that our DeFuzz can discover coverage more and faster than AFL. Moreover, DeFuzz exposes 43 more bugs than AFL on real-world applications.

Faezeh Farivar, Mohammad Sayad Haghighi, Alireza Jolfaei et al.

With the benefits of Internet of Vehicles (IoV) paradigm, come along unprecedented security challenges. Among many applications of inter-connected systems, vehicular networks and smart cars are examples that are already rolled out. Smart vehicles not only have networks connecting their internal components e.g. via Controller Area Network (CAN) bus, but also are connected to the outside world through road side units and other vehicles. In some cases, the internal and external network packets pass through the same hardware and are merely isolated by software defined rules. Any misconfiguration opens a window for the hackers to intrude into vehicles' internal components e.g. central lock system, Engine Control Unit (ECU), Anti-lock Braking System (ABS) or Adaptive Cruise Control (ACC) system. Compromise of any of these can lead to disastrous outcomes. In this paper, we study the security of smart vehicles' adaptive cruise control systems in the presence of covert attacks. We define two covert/stealth attacks in the context of cruise control and propose a novel intrusion detection and compensation method to disclose and respond to such attacks. More precisely, we focus on the covert cyber attacks that compromise the integrity of cruise controller and employ a neural network identifier in the IDS engine to estimate the system output dynamically and compare it against the ACC output. If any anomaly is detected, an embedded substitute controller kicks in and takes over the control. We conducted extensive experiments in MATLAB to evaluate the effectiveness of the proposed scheme in a simulated environment.

Tingmin Wu, Wanlun Ma, Sheng Wen et al.

Computer users are generally faced with difficulties in making correct security decisions. While an increasingly fewer number of people are trying or willing to take formal security training, online sources including news, security blogs, and websites are continuously making security knowledge more accessible. Analysis of cybersecurity texts can provide insights into the trending topics and identify current security issues as well as how cyber attacks evolve over time. These in turn can support researchers and practitioners in predicting and preparing for these attacks. Comparing different sources may facilitate the learning process for normal users by persisting the security knowledge gained from different cybersecurity context. Prior studies neither systematically analysed the wide-range of digital sources nor provided any standardisation in analysing the trending topics from recent security texts. Although LDA has been widely adopted in topic generation, its generated topics cannot cover the cybersecurity concepts completely and considerably overlap. To address this issue, we propose a semi-automated classification method to generate comprehensive security categories instead of LDA-generated topics. We further compare the identified 16 security categories across different sources based on their popularity and impact. We have revealed several surprising findings. (1) The impact reflected from cyber-security texts strongly correlates with the monetary loss caused by cybercrimes. (2) For most categories, security blogs share the largest popularity and largest absolute/relative impact over time. (3) Websites deliver security information without caring about timeliness much, where one third of the articles do not specify the date and the rest have a time lag in posting emerging security issues.

Derui, Wang, Chaoran Li et al.

Deep Neural Networks (DNNs) are vulnerable to deliberately crafted adversarial examples. In the past few years, many efforts have been spent on exploring query-optimisation attacks to find adversarial examples of either black-box or white-box DNN models, as well as the defending countermeasures against those attacks. In this work, we explore vulnerabilities of DNN models under the umbrella of Man-in-the-Middle (MitM) attacks, which has not been investigated before. From the perspective of an MitM adversary, the aforementioned adversarial example attacks are not viable anymore. First, such attacks must acquire the outputs from the models by multiple times before actually launching attacks, which is difficult for the MitM adversary in practice. Second, such attacks are one-off and cannot be directly generalised onto new data examples, which decreases the rate of return for the attacker. In contrast, using generative models to craft adversarial examples on the fly can mitigate the drawbacks. However, the adversarial capability of the generative models, such as Variational Auto-Encoder (VAE), has not been extensively studied. Therefore, given a classifier, we investigate using a VAE decoder to either transform benign inputs to their adversarial counterparts or decode outputs from benign VAE encoders to be adversarial examples. The proposed method can endue more capability to MitM attackers. Based on our evaluation, the proposed attack can achieve above 95% success rate on both MNIST and CIFAR10 datasets, which is better or comparable with state-of-the-art query-optimisation attacks. At the meantime, the attack is 104 times faster than the query-optimisation attacks.

Mahdi Dibaei, Xi Zheng, Kun Jiang et al.

Cyber security is one of the most significant challenges in connected vehicular systems and connected vehicles are prone to different cybersecurity attacks that endanger passengers' safety. Cyber security in intelligent connected vehicles is composed of in-vehicle security and security of inter-vehicle communications. Security of Electronic Control Units (ECUs) and the Control Area Network (CAN) bus are the most significant parts of in-vehicle security. Besides, with the development of 4G LTE and 5G remote communication technologies for vehicle-toeverything (V2X) communications, the security of inter-vehicle communications is another potential problem. After giving a short introduction to the architecture of next-generation vehicles including driverless and intelligent vehicles, this review paper identifies a few major security attacks on the intelligent connected vehicles. Based on these attacks, we provide a comprehensive survey of available defences against these attacks and classify them into four categories, i.e. cryptography, network security, software vulnerability detection, and malware detection. We also explore the future directions for preventing attacks on intelligent vehicle systems.

Xiaogang Zhu, Xiaotao Feng, Tengyun Jiao et al.

Fuzzing is a promising technique for detecting security vulnerabilities. Newly developed fuzzers are typically evaluated in terms of the number of bugs found on vulnerable programs/binaries. However,existing corpora usually do not capture the features that prevent fuzzers from finding bugs, leading to ambiguous conclusions on the pros and cons of the fuzzers evaluated. A typical example is that Driller detects more bugs than AFL, but its evaluation cannot establish if the advancement of Driller stems from the concolic execution or not, since, for example, its ability in resolving a dataset`s magic values is unclear. In this paper, we propose to address the above problem by generating corpora based on search-hampering features. As a proof-of-concept, we have designed FEData, a prototype corpus that currently focuses on four search-hampering features to generate vulnerable programs for fuzz testing. Unlike existing corpora that can only answer "how", FEData can also further answer "why" by exposing (or understanding) the reasons for the identified weaknesses in a fuzzer. The "why" information serves as the key to the improvement of fuzzers.

Xiaotao Feng, Qin Wang, Xiaogang Zhu et al.

With the frantic development of smart contracts on the Ethereum platform, its market value has also climbed. In 2016, people were shocked by the loss of nearly $50 million in cryptocurrencies from the DAO reentrancy attack. Due to the tremendous amount of money flowing in smart contracts, its security has attracted much attention of researchers. In this paper, we investigated several common smart contract vulnerabilities and analyzed their possible scenarios and how they may be exploited. Furthermore, we survey the smart contract vulnerability detection tools for the Ethereum platform in recent years. We found that these tools have similar prototypes in software vulnerability detection technology. Moreover, for the features of public distribution systems such as Ethereum, we present the new challenges that these software vulnerability detection technologies face.

Derui Wang, Chaoran Li, Sheng Wen et al.

This paper demonstrates that Non-Maximum Suppression (NMS), which is commonly used in Object Detection (OD) tasks to filter redundant detection results, is no longer secure. Considering that NMS has been an integral part of OD systems, thwarting the functionality of NMS can result in unexpected or even lethal consequences for such systems. In this paper, an adversarial example attack which triggers malfunctioning of NMS in end-to-end OD models is proposed. The attack, namely \texttt{Daedalus}, compresses the dimensions of detection boxes to evade NMS. As a result, the final detection output contains extremely dense false positives. This can be fatal for many OD applications such as autonomous vehicles and surveillance systems. The attack can be generalised to different end-to-end OD models, such that the attack cripples various OD applications. Furthermore, a way to craft robust adversarial examples is developed by using an ensemble of popular detection models as the substitutes. Considering the pervasive nature of model reusing in real-world OD scenarios, Daedalus examples crafted based on an \textit{ensemble of substitutes} can launch attacks without knowing the parameters of the victim models. Experimental results demonstrate that the attack effectively stops NMS from filtering redundant bounding boxes. As the evaluation results suggest, Daedalus increases the false positive rate in detection results to $99.9\%$ and reduces the mean average precision scores to $0$, while maintaining a low cost of distortion on the original inputs. It is also demonstrated that the attack can be practically launched against real-world OD systems via printed posters.

Xiao Chen, Chaoran Li, Derui Wang et al.

Machine learning based solutions have been successfully employed for automatic detection of malware on Android. However, machine learning models lack robustness to adversarial examples, which are crafted by adding carefully chosen perturbations to the normal inputs. So far, the adversarial examples can only deceive detectors that rely on syntactic features (e.g., requested permissions, API calls, etc), and the perturbations can only be implemented by simply modifying application's manifest. While recent Android malware detectors rely more on semantic features from Dalvik bytecode rather than manifest, existing attacking/defending methods are no longer effective. In this paper, we introduce a new attacking method that generates adversarial examples of Android malware and evades being detected by the current models. To this end, we propose a method of applying optimal perturbations onto Android APK that can successfully deceive the machine learning detectors. We develop an automated tool to generate the adversarial examples without human intervention. In contrast to existing works, the adversarial examples crafted by our method can also deceive recent machine learning based detectors that rely on semantic features such as control-flow-graph. The perturbations can also be implemented directly onto APK's Dalvik bytecode rather than Android manifest to evade from recent detectors. We demonstrate our attack on two state-of-the-art Android malware detection schemes, MaMaDroid and Drebin. Our results show that the malware detection rates decreased from 96% to 0% in MaMaDroid, and from 97% to 0% in Drebin, with just a small number of codes to be inserted into the APK.

Tingmin Wu, Lihong Tang, Rongjunchen Zhang et al.

Android users are increasingly concerned with the privacy of their data and security of their devices. To improve the security awareness of users, recent automatic techniques produce security-centric descriptions by performing program analysis. However, the generated text does not always address users' concerns as they are generally too technical to be understood by ordinary users. Moreover, different users have varied linguistic preferences, which do not match the text. Motivated by this challenge, we develop an innovative scheme to help users avoid malware and privacy-breaching apps by generating security descriptions that explain the privacy and security related aspects of an Android app in clear and understandable terms. We implement a prototype system, PERSCRIPTION, to generate personalised security-centric descriptions that automatically learn users' security concerns and linguistic preferences to produce user-oriented descriptions. We evaluate our scheme through experiments and user studies. The results clearly demonstrate the improvement on readability and users' security awareness of PERSCRIPTION's descriptions compared to existing description generators.

Rongjunchen Zhang, Xiao Chen, Jianchao Lu et al.

Intelligent Personal Assistant (IA), also known as Voice Assistant (VA), has become increasingly popular as a human-computer interaction mechanism. Most smartphones have built-in voice assistants that are granted high privilege, which is able to access system resources and private information. Thus, once the voice assistants are exploited by attackers, they become the stepping stones for the attackers to hack into the smartphones. Prior work shows that the voice assistant can be activated by inter-component communication mechanism, through an official Android API. However, this attack method is only effective on Google Assistant, which is the official voice assistant developed by Google. Voice assistants in other operating systems, even custom Android systems, cannot be activated by this mechanism. Prior work also shows that the attacking voice commands can be inaudible, but it requires additional instruments to launch the attack, making it unrealistic for real-world attack. We propose an attacking framework, which records the activation voice of the user, and launch the attack by playing the activation voice and attack commands via the built-in speaker. An intelligent stealthy module is designed to decide on the suitable occasion to launch the attack, preventing the attack being noticed by the user. We demonstrate proof-of-concept attacks on Google Assistant, showing the feasibility and stealthiness of the proposed attack scheme. We suggest to revise the activation logic of voice assistant to be resilient to the speaker based attack.

Derek Wang, Chaoran Li, Sheng Wen et al.

Deep neural networks (DNNs) are known to be vulnerable to adversarial examples which contain human-imperceptible perturbations. A series of defending methods, either proactive defence or reactive defence, have been proposed in the recent years. However, most of the methods can only handle specific attacks. For example, proactive defending methods are invalid against grey-box or white-box attacks, while reactive defending methods are challenged by low-distortion adversarial examples or transferring adversarial examples. This becomes a critical problem since a defender usually does not have the type of the attack as a priori knowledge. Moreover, existing two-pronged defences (e.g., MagNet), which take advantages of both proactive and reactive methods, have been reported as broken under transferring attacks. To address this problem, this paper proposed a novel defensive framework based on collaborative multi-task training, aiming at providing defence for different types of attacks. The proposed defence first encodes training labels into label pairs and counters black-box attacks leveraging adversarial training supervised by the encoded label pairs. The defence further constructs a detector to identify and reject high-confidence adversarial examples that bypass the black-box defence. In addition, the proposed collaborative architecture can prevent adversaries from finding valid adversarial examples when the defence strategy is exposed. In the experiments, we evaluated our defence against four state-of-the-art attacks on $MNIST$ and $CIFAR10$ datasets. The results showed that our defending method achieved up to $96.3\%$ classification accuracy on black-box adversarial examples, and detected up to $98.7\%$ of the high confidence adversarial examples. It only decreased the model accuracy on benign example classification by $2.1\%$ for the $CIFAR10$ dataset.