Rongxing Lu

Madushi H. Pathmaperuma, Yogachandran Rahulamathavan, Safak Dogan et al.

Despite the widespread use of encryption techniques to provide confidentiality over Internet communications, mobile device users are still susceptible to privacy and security risks. In this paper, a new Deep Neural Network (DNN) based user activity detection framework is proposed to identify fine grained user activities performed on mobile applications (known as in-app activities) from a sniffed encrypted Internet traffic stream. One of the challenges is that there are countless applications, and it is practically impossible to collect and train a DNN model using all possible data from them. Therefore, in this work we exploit the probability distribution of DNN output layer to filter the data from applications that are not considered during the model training (i.e., unknown data). The proposed framework uses a time window based approach to divide the traffic flow of an activity into segments, so that in-app activities can be identified just by observing only a fraction of the activity related traffic. Our tests have shown that the DNN based framework has demonstrated an accuracy of 90% or above in identifying previously trained in-app activities and an average accuracy of 79% in identifying previously untrained in-app activity traffic as unknown data when this framework is employed.

Wenbo Jiang, Hongwei Li, Jiaming He et al.

Recently, deep learning-based Image-to-Image (I2I) networks have become the predominant choice for I2I tasks such as image super-resolution and denoising. Despite their remarkable performance, the backdoor vulnerability of I2I networks has not been explored. To fill this research gap, we conduct a comprehensive investigation on the susceptibility of I2I networks to backdoor attacks. Specifically, we propose a novel backdoor attack technique, where the compromised I2I network behaves normally on clean input images, yet outputs a predefined image of the adversary for malicious input images containing the trigger. To achieve this I2I backdoor attack, we propose a targeted universal adversarial perturbation (UAP) generation algorithm for I2I networks, where the generated UAP is used as the backdoor trigger. Additionally, in the backdoor training process that contains the main task and the backdoor task, multi-task learning (MTL) with dynamic weighting methods is employed to accelerate convergence rates. In addition to attacking I2I tasks, we extend our I2I backdoor to attack downstream tasks, including image classification and object detection. Extensive experiments demonstrate the effectiveness of the I2I backdoor on state-of-the-art I2I network architectures, as well as the robustness against different mainstream backdoor defenses.

Xue Yang, Zifeng Liu, Xiaohu Tang et al.

With the emergence of privacy leaks in federated learning, secure aggregation protocols that mainly adopt either homomorphic encryption or threshold secret sharing have been widely developed for federated learning to protect the privacy of the local training data of each client. However, these existing protocols suffer from many shortcomings, such as the dependence on a trusted third party, the vulnerability to clients being corrupted, low efficiency, the trade-off between security and fault tolerance, etc. To solve these disadvantages, we propose an efficient and multi-private key secure aggregation scheme for federated learning. Specifically, we skillfully modify the variant ElGamal encryption technique to achieve homomorphic addition operation, which has two important advantages: 1) The server and each client can freely select public and private keys without introducing a trust third party and 2) Compared to the variant ElGamal encryption, the plaintext space is relatively large, which is more suitable for the deep model. Besides, for the high dimensional deep model parameter, we introduce a super-increasing sequence to compress multi-dimensional data into 1-D, which can greatly reduce encryption and decryption times as well as communication for ciphertext transmission. Detailed security analyses show that our proposed scheme achieves the semantic security of both individual local gradients and the aggregated result while achieving optimal robustness in tolerating both client collusion and dropped clients. Extensive simulations demonstrate that the accuracy of our scheme is almost the same as the non-private approach, while the efficiency of our scheme is much better than the state-of-the-art homomorphic encryption-based secure aggregation schemes. More importantly, the efficiency advantages of our scheme will become increasingly prominent as the number of model parameters increases.

Xiangman Li, Xiaodong Wu, Qi Li et al.

Jailbreak attacks pose a serious threat to the safety of Large Language Models (LLMs) by crafting adversarial prompts that bypass alignment mechanisms, causing the models to produce harmful, restricted, or biased content. In this paper, we propose SafeLLM, a novel unlearning-based defense framework that unlearn the harmful knowledge from LLMs while preserving linguistic fluency and general capabilities. SafeLLM employs a three-stage pipeline: (1) dynamic unsafe output detection using a hybrid approach that integrates external classifiers with model-internal evaluations; (2) token-level harmful content tracing through feedforward network (FFN) activations to localize harmful knowledge; and (3) constrained optimization to suppress unsafe behavior without degrading overall model quality. SafeLLM achieves targeted and irreversible forgetting by identifying and neutralizing FFN substructures responsible for harmful generation pathways. Extensive experiments on prominent LLMs (Vicuna, LLaMA, and GPT-J) across multiple jailbreak benchmarks show that SafeLLM substantially reduces attack success rates while maintaining high general-purpose performance. Compared to standard defense methods such as supervised fine-tuning and direct preference optimization, SafeLLM offers stronger safety guarantees, more precise control over harmful behavior, and greater robustness to unseen attacks. Moreover, SafeLLM maintains the general performance after the harmful knowledge unlearned. These results highlight unlearning as a promising direction for scalable and effective LLM safety.

Xiaodong Wu, Xiangman Li, Qi Li et al.

Text-guided image manipulation with diffusion models enables flexible and precise editing based on prompts, but raises ethical and copyright concerns due to potential unauthorized modifications. To address this, we propose SecureT2I, a secure framework designed to prevent unauthorized editing in diffusion-based generative models. SecureT2I is compatible with both general-purpose and domain-specific models and can be integrated via lightweight fine-tuning without architectural changes. We categorize images into a permit set and a forbid set based on editing permissions. For the permit set, the model learns to perform high-quality manipulations as usual. For the forbid set, we introduce training objectives that encourage vague or semantically ambiguous outputs (e.g., blurred images), thereby suppressing meaningful edits. The core challenge is to block unauthorized editing while preserving editing quality for permitted inputs. To this end, we design separate loss functions that guide selective editing behavior. Extensive experiments across multiple datasets and models show that SecureT2I effectively degrades manipulation quality on forbidden images while maintaining performance on permitted ones. We also evaluate generalization to unseen inputs and find that SecureT2I consistently outperforms baselines. Additionally, we analyze different vagueness strategies and find that resize-based degradation offers the best trade-off for secure manipulation control.

Guanlin Li, Guowen Xu, Han Qiu et al.

Generative Adversarial Networks (GANs) have been widely used in various application scenarios. Since the production of a commercial GAN requires substantial computational and human resources, the copyright protection of GANs is urgently needed. This paper presents a novel fingerprinting scheme for the Intellectual Property (IP) protection of image-to-image GANs based on a trusted third party. We break through the stealthiness and robustness bottlenecks suffered by previous fingerprinting methods for classification models being naively transferred to GANs. Specifically, we innovatively construct a composite deep learning model from the target GAN and a classifier. Then we generate fingerprint samples from this composite model, and embed them in the classifier for effective ownership verification. This scheme inspires some concrete methodologies to practically protect the modern image-to-image translation GANs. Theoretical analysis proves that these methods can satisfy different security requirements necessary for IP protection. We also conduct extensive experiments to show that our solutions outperform existing strategies.

Fan Yin, Rongxing Lu, Yandong Zheng et al.

The cloud computing technique, which was initially used to mitigate the explosive growth of data, has been required to take both data privacy and users' query functionality into consideration. Symmetric searchable encryption (SSE) is a popular solution to supporting efficient keyword queries over encrypted data in the cloud. However, most of the existing SSE schemes focus on the exact keyword query and cannot work well when the user only remembers the substring of a keyword, i.e., substring-of-keyword query. This paper aims to investigate this issue by proposing an efficient and privacy-preserving substring-of-keyword query scheme over cloud. First, we employ the position heap technique to design a novel tree-based index to match substrings with corresponding keywords. Based on the tree-based index, we introduce our substring-of-keyword query scheme, which contains two consecutive phases. The first phase queries the keywords that match a given substring, and the second phase queries the files that match a keyword in which people are really interested. In addition, detailed security analysis and experimental results demonstrate the security and efficiency of our proposed scheme.

Yizhong Liu, Jianwei Liu, Marcos Antonio Vaz Salles et al.

Sharding is the prevalent approach to breaking the trilemma of simultaneously achieving decentralization, security, and scalability in traditional blockchain systems, which are implemented as replicated state machines relying on atomic broadcast for consensus on an immutable chain of valid transactions. Sharding is to be understood broadly as techniques for dynamically partitioning nodes in a blockchain system into subsets (shards) that perform storage, communication, and computation tasks without fine-grained synchronization with each other. Despite much recent research on sharding blockchains, much remains to be explored in the design space of these systems. Towards that aim, we conduct a systematic analysis of existing sharding blockchain systems and derive a conceptual decomposition of their architecture into functional components and the underlying assumptions about system models and attackers they are built on. The functional components identified are node selection, epoch randomness, node assignment, intra-shard consensus, cross-shard transaction processing, shard reconfiguration, and motivation mechanism. We describe interfaces, functionality, and properties of each component and show how they compose into a sharding blockchain system. For each component, we systematically review existing approaches, identify potential and open problems, and propose future research directions. We focus on potential security attacks and performance problems, including system throughput and latency concerns such as confirmation delays. We believe our modular architectural decomposition and in-depth analysis of each component, based on a comprehensive literature study, provides a systematic basis for conceptualizing state-of-the-art sharding blockchain systems, proving or improving security and performance properties of components, and developing new sharding blockchain system designs.

Duc-Phong Le, Rongxing Lu, Ali A. Ghorbani

The advances of the Internet of Things (IoT) have had a fundamental impact and influence in sharping our rich living experiences. However, since IoT devices are usually resource-constrained, lightweight block ciphers have played a major role in serving as a building block for secure IoT protocols. In CHES 2015, SIMECK, a family of block ciphers, was designed for resource-constrained IoT devices. Since its publication, there have been many analyses on its security. In this paper, under the one bit-flip model, we propose a new efficient fault analysis attack on SIMECK ciphers. Compared to those previously reported attacks, our attack can recover the full master key by injecting faults into only a single round of all SIMECK family members. This property is crucial, as it is infeasible for an attacker to inject faults into different rounds of a SIMECK implementation on IoT devices in the real world. Specifically, our attack is characterized by exercising a deep analysis of differential trail between the correct and faulty immediate ciphertexts. Extensive simulation evaluations are conducted, and the results demonstrate the effectiveness and correctness of our proposed attack.

Bin Hu, Zongyang Zhang, Jianwei Liu et al.

Smart contracts are regarded as one of the most promising and appealing notions in blockchain technology. Their self-enforcing and event-driven features make some online activities possible without a trusted third party. Nevertheless, problems such as miscellaneous attacks, privacy leakage, and low processing rates pre-vent them from being widely applied. Various schemes and tools have been proposed to facilitate the construction and execution of secure smart contracts. However, a comprehensive survey for these proposals is absent, hindering new researchers and developers from a quick start. This paper surveys the literature and online resources on smart contract construction and execution over the period 2008-2020. We divide the studies into three categories: (1) design paradigms that give examples and patterns on contract construction, (2) design tools that facilitate the development of secure smart contracts, and (3) extensions and alternatives that improve the privacy or efficiency of the system. We start by grouping the relevant construction schemes into the first two categories. We then review the execution mechanisms in the last category and further divide the state-of-the-art solutions into three classes: private contracts with extra tools, off-chain channels, and extensions on core functionalities. Finally, we summarize several challenges and identify future research directions toward developing secure, privacy-preserving, and efficient smart contracts.

Yogachandran Rahulamathavan, Safak Dogan, Xiyu Shi et al.

Privacy-preserving applications allow users to perform on-line daily actions without leaking sensitive information. Privacy-preserving scalar product is one of the critical algorithms in many private applications. The state-of-the-art privacy-preserving scalar product schemes use either computationally intensive homomorphic (public-key) encryption techniques such as Paillier encryption to achieve strong security (i.e., 128-bit) or random masking technique to achieve high efficiency for low security. In this paper, lattice structures have been exploited to develop an efficient privacy-preserving system. The proposed scheme is not only efficient in computation as compared to the state-of-the-art but also provides high degree of security against quantum attacks. Rigorous security and privacy analyses of the proposed scheme have been provided along with a concrete set of parameters to achieve 128-bit and 256-bit security. Performance analysis shows that the scheme is at least five orders faster than the Paillier schemes and at least twice as faster than the existing randomisation technique at 128-bit security.

Xue Yang, Yan Feng, Weijun Fang et al.

Although federated learning improves privacy of training data by exchanging local gradients or parameters rather than raw data, the adversary still can leverage local gradients and parameters to obtain local training data by launching reconstruction and membership inference attacks. To defend such privacy attacks, many noises perturbation methods (like differential privacy or CountSketch matrix) have been widely designed. However, the strong defence ability and high learning accuracy of these schemes cannot be ensured at the same time, which will impede the wide application of FL in practice (especially for medical or financial institutions that require both high accuracy and strong privacy guarantee). To overcome this issue, in this paper, we propose \emph{an efficient model perturbation method for federated learning} to defend reconstruction and membership inference attacks launched by curious clients. On the one hand, similar to the differential privacy, our method also selects random numbers as perturbed noises added to the global model parameters, and thus it is very efficient and easy to be integrated in practice. Meanwhile, the random selected noises are positive real numbers and the corresponding value can be arbitrarily large, and thus the strong defence ability can be ensured. On the other hand, unlike differential privacy or other perturbation methods that cannot eliminate the added noises, our method allows the server to recover the true gradients by eliminating the added noises. Therefore, our method does not hinder learning accuracy at all.

Shuo Chen, Rongxing Lu, Jie Zhang

The Internet of Thing (IoT) has been a hot topic in both research community and industry. It is anticipated that in future IoT, an enormous number of sensors will collect the physical information every moment to enable the control center making better decisions to improve the quality of service (QoS). However, the sensors maybe faulty and thus generate inaccurate data which would compromise the decision making. To guarantee the QoS, the system should be able to detect faulty sensors so as to eliminate the damages of inaccurate data. Various faulty sensor detection mechanisms have been developed in the context of wireless sensor network (WSN). Some of them are only fit for WSN while the others would bring a communication burden to control center. To detect the faulty sensors for general IoT applications and save the communication resource at the same time, an efficient faulty sensor detection scheme is proposed in this paper. The proposed scheme takes advantage of fog computing to save the computation and communication resource of control center. To preserve the privacy of sensor data, the Paillier Cryptosystem is adopted in the fog computing. The batch verification technique is applied to achieve efficient authentication. The performance analyses are presented to demonstrate that the proposed detection scheme is able to conserve the communication resource of control center and achieve a high true positive ratio while maintaining an acceptable false positive ratio. The scheme could also withstand various security attacks and preserve data privacy.

Shuo Chen, Rongxing Lu, Jie Zhang

The singular value decomposition (SVD) is a widely used matrix factorization tool which underlies plenty of useful applications, e.g. recommendation system, abnormal detection and data compression. Under the environment of emerging Internet of Things (IoT), there would be an increasing demand for data analysis to better human's lives and create new economic growth points. Moreover, due to the large scope of IoT, most of the data analysis work should be done in the network edge, i.e. handled by fog computing. However, the devices which provide fog computing may not be trustable while the data privacy is often the significant concern of the IoT application users. Thus, when performing SVD for data analysis purpose, the privacy of user data should be preserved. Based on the above reasons, in this paper, we propose a privacy-preserving fog computing framework for SVD computation. The security and performance analysis shows the practicability of the proposed framework. Furthermore, since different applications may utilize the result of SVD operation in different ways, three applications with different objectives are introduced to show how the framework could flexibly achieve the purposes of different applications, which indicates the flexibility of the design.