Biometric authentication has become increasingly popular due to its security and convenience; however, traditional biometrics are becoming less desirable in scenarios such as new mobile devices, Virtual Reality, and Smart Vehicles. For example, while face authentication is widely used, it suffers from significant privacy concerns. The collection of complete facial data makes it less desirable for privacy-sensitive applications. Lip authentication, on the other hand, has emerged as a promising biometrics method. However, existing lip-based authentication methods heavily depend on static lip shape when the mouth is closed, which can be less robust due to lip shape dynamic motion and can barely work when the user is speaking. In this paper, we revisit the nature of lip biometrics and extract shape-independent features from the lips. We study the dynamic characteristics of lip biometrics based on articulator motion. Building on the knowledge, we propose a system for shape-independent continuous authentication via lip articulator dynamics. This system enables robust, shape-independent, and continuous authentication, making it particularly suitable for scenarios with high security and privacy requirements. We conducted comprehensive experiments in different environments and attack scenarios and collected a dataset of 50 subjects. The results indicate that our system achieves an overall accuracy of 99.06% and demonstrates robustness under advanced replay attacks and AI deepfake attacks, making it a viable solution for continuous biometric authentication in various applications.

Jailbreak vulnerabilities in Large Language Models (LLMs) refer to methods that extract malicious content from the model by carefully crafting prompts or suffixes, which has garnered significant attention from the research community. However, traditional attack methods, which primarily focus on the semantic level, are easily detected by the model. These methods overlook the difference in the model’s alignment protection capabilities at different output stages. To address this issue, we propose an adaptive position pre-fill jailbreak attack approach for executing jailbreak attacks on LLMs. Our method leverages the model’s instruction-following capabilities to first output pre-filled safe content, then exploits its narrative-shifting abilities to generate harmful content. Extensive black-box experiments demonstrate our method can improve the attack success rate by 47% on the widely recognized secure model (Llama2) compared to existing approaches. Our code can be found at: https://github.com/Yummy416/AdaPPA.

Intrusion detection based on deep learning is inherently limited by the black-box nature of the models, which makes it difficult to ensure the trustworthiness of the results. Explainable Artificial Intelligence (XAI) techniques address this limitation by leveraging the powerful feature learning capabilities of black-box deep learning models and employing XAI methods to explore their decision boundaries, enabling the effective extraction of rules for intrusion detection. This approach provides a practical solution to the aforementioned challenges. In this paper, we propose an XAI-based automated rule extraction method for intrusion detection, designed to offer highly interpretable detection capabilities for encrypted traffic. The method begins by using CICFlowMeter to extract tabular traffic features and training a surrogate model to learn the representations of these features. Subsequently, an automated approach is developed to extract decision rules based on information from neural network layers, generating an initial rule set. This rule set is further refined through fine-tuning to optimize detection rules. We conducted experiments on two datasets using the generated detection rules. The experimental results demonstrate that the proposed method not only achieves excellent detection performance but also produces rules with high interpretability.

Network traffic classification stands as an essential endeavor within the realms of network security and management. The recent advances in learning-based methodologies have underscored their efficacy in deducing patterns from the side-channel features of encrypted network traffic. The unpredictability of traffic bursts can result in packet loss during retransmission, thereby generating fragmented feature patterns. Unfortunately, current approaches struggle to adapt to such fragmented features, often leading to a substantial decline in performance. To surmount this challenge, this paper introduces a pre-training-based framework, denoted as Nüwa, which imputes the side-channel features of encrypted network traffic, especially focusing on the temporal attributes of missing packets within a traffic session. Firstly, we propose a word-level Sequence2Embedding (S2E) module to transform side-channel features into tokens for model pre-training, as well as a Traffic Feature Masking strategy (TFM) to simulate the original flows changes in packet loss network. Besides, we also introduce a Traffic Feature Imputation (TFI) module to restore the missing values of original traffic flows in an efficient and context-aware manner. Experiments across four diverse real-world scenarios substantiate Nüwa’s capacity to restore the performance of prevalent temporal models, while maintaining the integrity of the imputed features. Notably, Nüwa has also demonstrated an impressive resilience, even under conditions of extensive feature loss and domain adaptation. The Nüwa prototype has been made accessible to the public for further research and development (https://github.com/Timeless-zfqi/Nuwa).

As the popularity of Advanced Persistent Threat (APT) grows, APT malware group classification has attracted more attention recently.However, most of previous methods use simple classifiers for group classification, ignoring the bias caused by the sparse number of revealed malware and the differences in functionality distribution of most groups.In this paper, we propose a Prediction-Guided Prototype Network (PGPNet) that could quickly adapt to new classification tasks with limited supervised samples based on the metalearning architecture.Adding malware functionality classification as an auxiliary task is beneficial for feature learning, and the bias of distribution differences is eliminated by intervening the predicted results into the group classifier.Experimental results on a APT malware dataset show that PGPNet successfully exploits the contextual information and predictions of the auxiliary task and achieves state-of-the-art performance.

The recent advances in learning-based methodologies has underscored their efficacy in deducing patterns from the side-channel features of encrypted network traffic. Nonetheless, the distribution of these features has been identified as susceptible, particularly in the expansive and intricate network topologies characteristic of the modern Internet. The unpredictability of traffic bursts can result in packet loss during retransmission, thereby generating fragmented feature patterns. Unfortunately, current approaches struggle to adapt to such fragmented features, often leading to a substantial decline in performance. To surmount this challenge, this paper introduces a pre-training-based augmentation framework, denoted as Nüwa, which imputes the side-channel features of encrypted network traffic. The crux of Nüwa lies in its ability to reconstruct the side-channel features, with a particular focus on the temporal attributes of the missing packets within a traffic session. Nüwa is comprised of a word-level Sequence2Embedding module, a Traffic Noise-based Self-supervised Pre-trained Masking Strategy, and a Traffic Side-Channel Feature Imputation Module. Experiments across four diverse real-world scenarios substantiate Nüwa's capacity to restore the performance of prevalent temporal models while maintaining the integrity of the imputed features.

The detection of malicious macros has been a prominent focus of research. Previous approaches exhibit two notable shortcomings. Firstly, methods centered on document and macro code features often fall short in effectively countering targeted adversarial strategies. Secondly, detection techniques relying on deceptive information, such as visual and textual cues, although alleviating certain challenges, introduce a new vulnerability to adversarial machine learning techniques. In this paper, we present Collaborative Adaptive Feature Extraction method (CAFE), designed for robust detection based on deceptive information. The core of CAFE is a feature fusion network architecture, where modality-shared associations and modalityprivate information are modeled from feature of different modalities, resulting in independently valid and comprehensive feature representations. An adaptive feature sampling module is introduced to address partial feature absence, enhancing detection robustness. Experimental results, conducted on two datasets, demonstrate that CAFE adeptly captures shared and complementary information from two modalities, showcasing its capability for robust malicious macro detection in the presence of input noise and adversarial samples. Index Terms—Malicious Macro Detection, Multi-modal Features, Model Robustness, Security Wen Wang is corresponding author.

While various defense mechanisms have been proposed in cybersecurity, it is still unclear how these defense mechanisms should be dynamically employed to mitigate the damage of multi-stage attacks. In this work, we consider the problem of generating defense strategy in real-time to thwart multi-stage attacks. We use the Bayesian condition dependency graph (BCDG) to model the interactions between the attacker and the defender. Considering that both the attacker and the defender have uncertainty about their respective observations, we formulate the strategy selection problem as a partially observable Markov decision process (POMDP), where the attacker and the defender need to find their optimal strategies in a partially observation environment. To solve the problem of state space explosion, we develop a deep reinforcement learning (DRL) based approach to seek the optimal strategies. We conduct experiments with various settings to evaluate the effectiveness of our approach. Experiment results show that our DRL-based approach outperforms baselines, and the approach is robust to the uncertain security environment.

Cyberattacks on digital services and Internet of Things (IoT) are rising, employing complex tactics. Using intrusion detection systems (IDS) to detect and counter threats at key network points is vital for strong cybersecurity. Traditional rule-based network IDS rely on predefined rules, which may not effectively recognize the myriad complex variants of potential attacks. AI-driven methods for detecting malicious traffic offer enhanced capabilities but can fall short in terms of interpretability and performance under high-throughput network conditions. To address these challenges, we propose a LLM-based (Large Language Model) agent that utilizes multiple sources inputs to generate and generalize rules. The generated rules are designed to detect a variety of corresponding malicious threats, while the generalized rules are crafted to identify similar variant attacks. We have amassed an extensive dataset, comprising vulnerability security reports, malicious traffic, and original IDS rules from authoritative sources, which serve as input for the LLM-based agent. Subsequently, comparative experiments were conducted to assess the performance of the new rules in detecting malicious traffic. The experimental results demonstrate the superior performance of these new rules across various metrics for malicious traffic detection.

Advanced Persistent Threat (APT) malware group classification has attracted more attention recently. Previous methods have two downsides. First, most use conventional classifiers ignoring the bias caused by the sparse number of revealed malware. Second, they conducted on closed-set without considering the constant stream of novel APT groups. In this paper, we propose a framework for open-set APT malware classification under a few-shot setting. First, the pre-trained encoder extracts the dynamic behavioral features of APT malware. Then the prototypes of known APT groups are calculated. Based on these prototypes the classification probability of the test sample is calculated. Finally, we devise plug-and-play open-set loss and dynamic triplet threshold modules to construct clear boundaries of known categories to achieve open-set recognition. Experimental results conducted on two datasets show that our approach achieves state-of-the-art performance, enabling the detection of known APT malware and recognition of unknown malware with few known APT-labelled malware.

Backward-compatible stereoscopic display, a novel display technique that can simultaneously present satisfying 3D effects to viewers with stereo glasses and clear 2D contents to viewers without, aims at helping the people who are unsuitable for watching 3D movies for a long time. In this article, we introduce two versions of backward-compatible stereoscopic display: the simpler version is far simpler than Hidden Stereo (the state-of-the-art method) while preserving competitive 2D–3D effects; the advanced version, which we call 2D-Ghost-Free Stereoscopic Display, overcomes the limitation that Hidden Stereo and the simpler version are both confined to small absolute disparity. 2D-Ghost-Free Stereoscopic Display improves tolerable disparity range by adding depth-of-field in the regions with large disparity, so that it can be applied to more scenes of 3D movies. User experiments and theoretical analysis both demonstrate the superiority of the 2D-Ghost-Free Stereoscopic Display over the state-of-the-art method and our simpler version. In addition, to make the user experiments double-blind and automatic, we developed a user study system that can automatically present 3D images and videos in NVIDIA 3D Vision 2 and collect corresponding votes of subjects on stimuli, whereas the previous researchers did not state that their user experiments were double-blind.

Stereo matching is an essential basis for various applications, but most stereo matching methods have poor generalization performance and require a fixed disparity search range. Moreover, current stereo matching methods focus on the scenes that only have positive disparities, but ignore the scenes that contain both positive and negative disparities, such as 3D movies. In this paper, we present a new stereo matching pipeline that first computes semi-dense disparity maps based on binocular disparity, and then completes the rest depending on monocular cues. The new stereo matching pipeline have the following advantages: It 1) has better generalization performance than most of the current stereo matching methods; 2) relaxes the limitation of a fixed disparity search range; 3) can handle the scenes that involve both positive and negative disparities, which has more potential applications, such as view synthesis in 3D multimedia and VR/AR. Experimental results demonstrate the effectiveness of our new stereo matching pipeline.

Stereo matching is an essential basis for various applications in computer vision, but currently, most stereo matching methods have poor generalization performance and require a fixed disparity search range. In this work, we propose to adopt the operation "sub-pixel three-around-maximum", which supports an unfixed disparity search range, to replace the currently popular operation soft argmax. We also propose to directly supervise the feature extractor by three loss functions. Just depending on the feature extractor, we can obtain an accurate semi-dense disparity map before cost aggregation, which can help to remove adaptively the redundancy part of the predefined disparity search range. An adaptive disparity search range for each stereo pair can save much time and memory. The proposed architecture achieves state-of-the-art cross-domain generalization performance in the datasets KITTI2012&2015, and experimental results demonstrate that our method supports unfixed and adaptive disparity search range.

Moving target defense (MTD) suggests a game-changing way of enhancing web security by increasing uncertainty and complexity for attackers. A good number of web MTD techniques have been investigated to counter various types of web attacks. However, in most MTD techniques, only fixed attributes of the attack surface are shifted, leaving the rest exploitable by the attackers. Currently, there are few mechanisms to support the whole attack surface movement and solve the partial coverage problem, where only a fraction of the possible attributes shift in the whole attack surface. To address this issue, this paper proposes a Web Gene Tree (WGT) based MTD mechanism. The key point is to extract all potential exploitable key attributes related to vulnerabilities as web genes, and mutate them using various MTD techniques to withstand various attacks. Experimental results indicate that, by randomly shifting web genes and diversely inserting deceptive ones, the proposed WGT mechanism outperforms other existing schemes and can significantly improve the security of web applications.

The homogeneity and uniformity of static data storage and access make data leakage one of the most severe security threats. Dynamic data techniques such as data randomization and diversification, are effective approaches to mitigate data theft and illegal data modification. By increasing data diversity and dynamics, the data attack surface shifting space can be expanded to confuse attackers and influence their further actions. However, there are only a few dynamic data techniques developed because of the difficulty in encoding multiple data formats and the loss of compatibility in data formats. In this paper, we propose a new dynamic data approach that integrates the data deception techniques based on Moving Target Defense (MTD). By changing the data size, data authenticity, and users' data access privilege, the approach significantly expands the data attack surface shifting space. Moreover, the approach provides dynamic data access based upon both users' attributes and users' operations. Through dynamic analysis and experiments, the paper shows that the proposed dynamic data technique can expand the attack surface shifting space at a lower cost, protect the sensitive data, and impose no significant burden on the system.

An advanced persistent threat (APT) attack is a powerful cyber-weapon aimed at the specific targets in cyberspace. The sophisticated attack techniques, long dwell time and specific objectives make the traditional defense mechanism ineffective. However, most existing studies fail to consider the theoretical modeling of the whole APT attack. In this paper, we mainly establish a theoretical framework to characterize an information-based APT attack on the internal network. In particular, our mathematical framework includes the initial entry model for selecting the entry points and the targeted attack model for studying the intelligence gathering, strategy decision-making, weaponization and lateral movement. Through a series of simulations, we find the optimal candidate nodes in the initial entry model, observe the dynamic change of the targeted attack model and verify the characteristics of the APT attack.

When facing the on‐going cyber epidemic threats, individuals usually set up cyber defences to protect their own devices. In general, the individual‐level cyber defence is considered to mitigate the cyber threat to some extent. However, few previous studies focus on the interaction between individual‐level defence and cyber epidemic attack from the perspective of dynamics. In this study, the authors propose a two‐way dynamical framework by coupling the individual defence model with the cyber epidemic model to study the interaction between the network security situation and individual‐level defence decision. A new individual‐based heterogeneous model for cyber epidemic attacks is established to emphasise the individual heterogeneity in defence strategy. In the meanwhile, a Markov decision process is used to characterise the defence decision in the individual defence decision model. The theoretical and numerical results illustrate that the individual‐level defence can dampen the cyber epidemic attack, but the current network security situation, in turn, influences the individual defence decision. Moreover, they obtain a glimpse of the network security situation and the individual defence with respect to different cyber epidemic scenarios.

In the past decade, the researchers paid more attention to the cheating problem in visual cryptography (VC) so that many cheating prevention visual cryptography schemes (CPVCS) have been proposed. In this paper, the authors propose a novel method, which first makes use of Latin square to prevent cheating in VC. Latin squares are utilised to guide the choosing of authentication regions in different rows and columns of each divided block of the shares, which ensures that the choosing of authentication regions is both random and uniform. Without pixel expansion, the new method provides random regions authentication in each divided block of all shares. What is important is that the proposed method is applicable to both ( k , n )‐deterministic visual cryptography scheme (( k , n )‐DVCS) and ( k , n )‐probabilistic visual cryptography scheme (( k , n )‐PVCS). Experimental results and properties analysis are given to show the effectiveness of the proposed method.

Visual Cryptography (VC) has been studied as a significant way of information security. In VC, original secret is divided into two images called shares. VC shares show no clue for secret perceptually, whereas participants are able to obtain the secret by simply superimposing the shares. Despite the obvious advantages of VC in crucial secret protection, one of its issues appears to be the authentication method for VC shares. It is likely to seek assistance from other areas of digital image processing. As an international standard reading guidance for the visually impaired people, Braille has been widely used as an effective communication channel. In this paper, we will explain Braille encoding and explain how it is applied to handle the authentication problem in VC. Our contribution is to use Braille for VC. To the best of our knowledge, this is the first time the Braille has been employed to the authentication of VC.

Watermarking is a technique to protect the copyright of digital media such as image, text, music and movie. In this study, a robust watermarking scheme for multiple cover images and multiple owners is proposed. The proposed scheme makes use of the visual cryptography (VC) technique, transform domain technique, chaos technique, noise reduction technique and error correcting code technique where the VC technique provides the capability to protect the copyright of multiple cover images for multiple owners, and the rest of the techniques are applied to enhance the robustness of the scheme.

Most cheating immune visual cryptography schemes (CIVCS) are based on a traditional visual cryptography scheme (VCS) and are designed to avoid cheating when the secret image of the original VCS is to be recovered. However, all the known CIVCS have some drawbacks. Most usual drawbacks include the following: the scheme needs an online trusted authority, or it requires additional shares for the purpose of verification, or it has to sacrifice the properties by means of pixel expansion and contrast reduction of the original VCS or it can only be based on such VCS with specific access structures. In this study, the authors propose a new CIVCS that can be based on any VCS, including those with a general access structure, and show that their CIVCS can avoid all the above drawbacks. Moreover, their CIVCS does not care about whether the underlying operation is OR or XOR.

A visual cryptography scheme (VCS) is a kind of secret sharing scheme which allows the encoding of a secret image intonshares distributed tonparticipants. The beauty of such a scheme is that a set of qualified participants is able to recover the secret image without any cryptographic knowledge and computation devices. An extended visual cryptography scheme (EVCS) is a kind of VCS which consists of meaningful shares (compared to the random shares of traditional VCS). In this paper, we propose a construction of EVCS which is realized by embedding random shares into meaningful covering shares, and we call it the embedded EVCS. Experimental results compare some of the well-known EVCSs proposed in recent years systematically, and show that the proposed embedded EVCS has competitive visual quality compared with many of the well-known EVCSs in the literature. In addition, it has many specific advantages against these well-known EVCSs, respectively.

Droste [CRYPTO’96] proposed a construction of threshold visual cryptography scheme (TVCS) under the visual cryptography model of Naor and Shamir, i.e. the visual cryptography model with the underlying operation OR. In this article, we give three extensions of TVCS. First, we prove that the TVCS proposed by Droste which was based on the OR operation is still a valid TVCS under the XOR operation, and then we propose a method to further reduce its pixel expansion. We then propose an interesting construction of TVCS with all shares being concolorous. Finally, we give a construction of threshold extended visual cryptography scheme (TEVCS) with the underlying operation OR or XOR. All of our schemes can be applied to the visual cryptography model introduced by Tuyls et al. (First Int. Conf. Security in Pervasive Computing 2004, International Patent with Application No.: PCT/IB2003/000261).

Two common drawbacks of the visual cryptography scheme (VCS) are the large pixel expansion of each share image and the small contrast of the recovered secret image. In this paper, we propose astep constructionto construct$\hbox{VCS}_{\rm OR}$and$\hbox{VCS}_{\rm XOR}$for general access structure by applying (2,2)-VCS recursively, where a participant may receive multiple share images. The proposed step construction generates$\hbox{VCS}_{\rm OR}$and$\hbox{VCS}_{\rm XOR}$which have optimal pixel expansion and contrast for each qualified set in the general access structure in most cases. Our scheme applies a technique to simplify the access structure, which can reduce the average pixel expansion (APE) in most cases compared with many of the results in the literature. Finally, we give some experimental results and comparisons to show the effectiveness of the proposed scheme.

In IEE Proceedings Communications, 2005, 152, (6), an efficient authenticated identity-based multicast scheme from bilinear pairing was proposed. Wang proved that the scheme is secure against the adversary who can forge a ciphertext, but it is shown that their scheme is not secure against an inside forger.

Visual cryptography scheme (VCS) is a kind of secret-sharing scheme which allows the encryption of a secret image into n shares that are distributed to n participants. The beauty of such a scheme is that, the decryption of the secret image requires neither the knowledge of cryptography nor complex computation. Colour visual cryptography becomes an interesting research topic after the formal introduction of visual cryptography by Naor and Shamir in 1995. The authors propose a colour (k, n)-VCS under the visual cryptography model of Naor and Shamir with no pixel expansion, and a colour (k, n)-extended visual cryptography scheme ((k, n)-EVCS) under the visual cryptography model of Naor and Shamir with pixel expansion the same as that of its corresponding black and white (k, n)-EVCS. Furthermore, the authors propose a black and white (k, n)-VCS and a black and white (k, n)-EVCS under the visual cryptography model of Tuyls. Based on the black and white schemes, the authors propose a colour (k, n)-VCS and a colour (k, n)-EVCS under the same visual cryptography model, of which the pixel expansions are the same as that of their corresponding black and white (k, n)-VCS and (k, n)-EVCS, respectively. The authors also give the experimental results of the proposed schemes, and compare the proposed scheme with known schemes in the literature.