Existing transferable adversarial attacks process image pixels indiscriminately, leading to the overfitting of adversarial perturbations to model-specific background features and thus limited transferability. To address this issue, we propose an Importance-Aware Pixel-Level Masking (IPM) approach. By applying pixel-level random masks to non-critical regions with minimal impact on model decisions, IPM alleviates the overfitting of adversarial perturbations and guides perturbations to focus on disrupting decision-critical regions. Specifically, we first leverage a class activation mapping technique to locate important pixel regions. Then, we apply sparse random masks to the non-critical regions of the image before each iterative attack. Finally, we compute adversarial perturbations from the mask-preprocessed images and update the adversarial perturbations on the original unmasked images. Extensive experiments on the ImageNet dataset validate the superior effectiveness of the proposed method. Integrating IPM with existing approaches can significantly further boost the performance of state-of-the-art transferable attacks.

To address the issue of underground workers entering the yellow-line restricted area (dangerous area) and meet safety standard requirements, we propose a personnel intrusion detection method based on an improved YOLOv7 object detection model. The improvements to the YOLOv7s model include: first, integrating attention mechanisms into the model's backbone layer to enhance its ability to extract features of small targets; second, improving the regression loss function to enhance the accuracy of bounding box localization. These improvements effectively address the identification rate of bounding boxes in densely populated work environments and improve the accuracy of detecting region intrusions. Additionally, we constructed an image sample library specifically for the underground mine working area, conducted sample cleaning to ensure the informational richness of the sample set. Experimental analysis shows that the improved YOLOv7 model effectively detects personnel intrusion into restricted areas, with an average precision improvement of 4.31% compared to the original YOLOv7 model, effectively reducing false positives and false negatives. Moreover, the PNPloy algorithm is used to implement the detection of dangerous area intrusion. This method has been successfully applied and practiced in engineering.

Ring Confidential Transactions (RingCT) is a classic cryptographic protocol for anonymous transactions on blockchains, currently used in the popular anonymous cryptocurrency Monero. The proof size of RingCT transactions is linearly related to the ring size, which limits the use of larger ring sizes due to the significant communication overhead it incurs. However, reducing the ring size also leads to decreased anonymity. Therefore, in recent years, many studies have focused on optimizing the proof sizes for RingCT, with the latest known solutions reducing the proof size to be logarithmic with the ring size. In this paper, we propose a new compact RingCT protocol (CRCT) for smaller proof sizes. To this end, we first design three extended schemes of the Sum Argument (CRYPTO’21), which are used to generate logarithmic-sized proofs for three distinct zero-knowledge arguments, respectively. We then introduce a new zero-knowledge proof scheme called the Difference Argument. It is used to prove that one has the knowledge of two secret values, with their difference being public. Based on these schemes, we construct our CRCT protocol, whose proof size is independent of the ring size and logarithmic with the number of source accounts. We provide concrete constructions and security proofs for the proposed cryptographic schemes. The experimental results demonstrate that CRCT exhibits significant advantages in computational efficiency and proof size over existing solutions when dealing with large ring sizes and moderate numbers of source accounts.

Ring confidential transactions (RingCT) is a classical protocol that enhances transaction privacy within blockchain technology. It is primarily utilized in anonymous cryptocurrencies such as Monero. In recent years, to counter the security threats posed by quantum computing to classical RingCT (C-RCT), several post-quantum RingCT (P-RCT) protocols were proposed. However, prior studies do not address the account transition from C-RCT to P-RCT, which is essential to ensure the quantum resistance and usability of user accounts following the upgrade from C-RCT to P-RCT. In this article, we introduce a RingCT account transition protocol (RAT), which allows users to migrate their accounts from the blockchain with C-RCT to that with P-RCT. The protocol consists of two cryptographic components: 1) a zero-knowledge proof scheme used to prove that two group generators correspond to the same witness and 2) a cross-cryptosystems commitment transition protocol. RAT is fully decentralized and does not require any trusted third parties. Additionally, the transition process does not disclose the account balance and maintains both atomicity and consistency. The security of our protocol relies on the cryptographic components within RingCT, without the need for additional security assumptions, such as a trusted setup. The experimental results demonstrate the effectiveness of our protocol in practice.

Although quantum key distribution (QKD) is proven secure in theory, the key transfer process from the QKD device to other applications introduces new vulnerability. This paper leverages the Dolev-Yao model to establish a three-entity formal model for the key transfer protocol, employing the ProVerif tool to verify its security properties—namely, authentication, confidentiality, and consistency. The analysis demonstrates that the protocol is vulnerable to a key inconsistency attack, which causes the communicating parties to receive different keys. By implementing request deduplication and enhanced HMAC-based verification, the improved protocol eliminates the vulnerability, thereby maintaining all its security properties.

The quest for perfect quantum oblivious transfer (QOT) with information-theoretic security remains a challenge, necessitating the exploration of computationally secure QOT as a viable alternative. Unlike the unconditionally secure quantum key distribution (QKD), the computationally secure QOT relies on specific quantum-safe computational hardness assumptions, such as the post-quantum hardness of learning with errors (LWE) problem and quantum-hard one-way functions. This raises an intriguing question: Are there additional efficient quantum hardness assumptions that are suitable for QOT? In this work, leveraging the dihedral coset state derived from the dihedral coset problem (DCP), a basic variant of OT, known as the all-or-nothing OT, is studied in the semi-quantum setting. Specifically, the DCP originates from the dihedral hidden subgroup problem (DHSP), conjectured to be challenging for any quantum polynomial-time algorithms. First, a computationally secure quantum protocol is presented for all-or-nothing OT, which is then simplified into a semi-quantum OT protocol with minimal quantumness, where the interaction needs merely classical communication. To efficiently instantiate the dihedral coset state, a powerful cryptographic tool called the LWE-based noisy trapdoor claw-free functions (NTCFs) is used. The construction requires only a three-message interaction and ensures perfect statistical privacy for the receiver and computational privacy for the sender.

Ring Confidential Transactions (RingCT) is a typical privacy-preserving protocol for blockchain, which is used for the most popular anonymous cryptocurrency Monero in recent years. RingCT provides the user's identity anonymity based on the linkable ring signature. At the cost of that, the transaction size is increased linearly to the involved users. In this article, we aim to overcome this inefficient aspect of RingCT by introducing the linkable threshold ring signature (LTRS). We first propose a construction of threshold ring signatures for homomorphic cryptosystems, and present an efficient instantiation based on the intractability assumption of the discrete logarithm problem. Based on this framework, an efficient LTRS scheme and a novel construction of the RingCT protocol are presented. Our proposed RingCT protocol enables multiple payers to co-construct an anonymous transaction without revealing their secret account keys, and it is more concise under multiple input accounts. For a transaction with a ring size of 100 and the input accounts number of 64, the communication overhead is about 4% of the original RingCT protocol.

Currently, when detecting small targets in images captured from aerial drones, the mainstream detection algorithms face a series of issues, such as missed detections, false positives, and an imbalance between precision and speed. Based on the YOLOv8s model, we propose an improved algorithm for small target detection in drone-captured images. Firstly, we've enhanced the network's ability to detect small targets by adding a dedicated detection layer to the neck of the network. We've introduced an SPD (Space-to-depth) layer in the shallower layers of the network, which enables the model to effectively extract the feature information of small targets present in the images. Within the C2f module of the network, we've integrated the Triplet Attention mechanism to suppress the impact of complex backgrounds on detection and enhances the model's performance in small target detection tasks. Finally, we have added the CPCA (Channel Prior Convolutional Attention) module to the neck network, which further enhances the quality of composite features of targets and reduces sensitivity to background noise. Through the design of corresponding experiments, the final results show that our proposed detection model has improved the mAP0.5 by 6.7% on the public dataset VisDrone2019 compared to the YOLOv8s, demonstrating the advantage of our model in small target detection.

Data sharing among vehicles can effectively address the traffic congestion and accidents caused by the increasing number of vehicles, thereby enhancing traffic efficiency and the travel experience. However, it also introduces security and privacy challenges related to confidentiality, authentication, identity privacy, identity revocation, and tamper resistance. To address the above challenges, we propose a bilateral access control data sharing scheme by extending Matchmaking Encryption. Our proposal ensures data confidentiality and data source authentication by combining attribute-based encryption with identity-based encryption. Most importantly, we verify the bilateral policies within a single logical step. To achieve identity revocation, we propose a revocation scheme based on a pseudo-identity list, which can revoke all the pseudo-identities associated with a malicious user and ensure the privacy of legitimate real identity. Security analysis indicates that apart from ensuring confidentiality and authentication, our proposal resists attacks such as tampering, guessing, and collusion. We conduct theoretical complexity analysis and experimental performance evaluations to demonstrate the efficiency and practicality of our proposal for Internet of Vehicles (IoV) data sharing.

Immutability has always been a core attribute of blockchain to ensure a trusted public ledger. However, in recent years, many malicious actions on blockchain have raised controversy over strict immutability. As a practical matter, it is necessary to design redactable blockchain protocols in some cases. Previous studies have proposed methods to implement redaction features in permissionless blockchains. However, in these protocols, parties can refuse to accept redacted blocks without affecting their mining. It makes these protocols lack feasibility in practice. In this paper, we present Chainknot, a redactable blockchain protocol with forced synchronization. We use the incentives of permissionless blockchain to effectively facilitate the review of a redaction proposal, and design a novel blockchain structure that ensures the acceptance of modified blocks by parties. In addition, we present instantiations of Chainknot on the PoW-based and PoS-based blockchain systems, with further improvements for efficiency and security, respectively. The experimental results show that the additional latency our protocol imposes on the original immutable blockchain system is negligible (less than 4%).

Immutability is widely recognized as one of the blockchain’s key security attributes. However, in recent years, incidents involving the use of blockchain for disseminating illegal or malicious information have raised concerns over its strict immutability. To address these issues, redactable blockchains are proposed as a novel solution, permitting authorized content redactions without compromising the structural integrity of the blockchain. Unfortunately, current solutions are unable to restrict the abuse of redaction privilege, except for relying on a trusted authority or committee, which contradicts the trustlessness principle of blockchain. In this paper, we propose a controlled redactable blockchain protocol that allows for a limited number of redactions and supports a transparent setup. The cryptographic tools enabling this functionality are our proposed t -times chameleon hash ( t -CH) and signature ( t -CS) schemes, where generating more than t collisions will expose the trapdoor. We present security models, discrete logarithmbased instantiations, and formal security proofs for both t -CH and t -CS. Subsequently, we present the construction of our redaction protocol in both permissioned and permissionless settings. Finally, we experimentally demonstrate the effectiveness of the proposed protocol in practice.

\emph{Noisy trapdoor claw-free function} (NTCF) as a powerful post-quantum cryptographic tool can efficiently constrain actions of untrusted quantum devices. However, the original NTCF is essentially \emph{2-to-1} one-way function (NTCF$^1_2$). In this work, we attempt to further extend the NTCF$^1_2$ to achieve \emph{many-to-one} trapdoor claw-free functions with polynomial bounded preimage size. Specifically, we focus on a significant extrapolation of NTCF$^1_2$ by drawing on extrapolated dihedral cosets, thereby giving a model of NTCF$^1_{\kappa}$ where $\kappa$ is a polynomial integer. Then, we present an efficient construction of NTCF$^1_{\kappa}$ assuming \emph{quantum hardness of the learning with errors (LWE)} problem. We point out that NTCF can be used to bridge the LWE and the dihedral coset problem (DCP). By leveraging NTCF$^1_2$ (resp. NTCF$^1_{\kappa}$), our work reveals a new quantum reduction path from the LWE problem to the DCP (resp. extrapolated DCP). Finally, we demonstrate the NTCF$^1_{\kappa}$ can naturally be reduced to the NTCF$^1_2$, thereby achieving the same application for proving the quantumness.

In recent years, Federated Learning has attracted much attention because it solves the problem of data silos in machine learning to a certain extent. However, many studies have shown that attacks based on Generative Adversarial Networks pose a great threat to Federated Learning. This paper proposes Defense-GAN, a defense method against Generative Adversarial Network attacks under Federated Learning. Under this method, the attacker cannot learn the real image data distribution. Each Federated Learning participant uses SHAP to explain the model and masks the pixel features that have a greater impact on classification and recognition in their respective image data. The experimental results show that while attacking the federated training model using masked images, the attacker cannot always obtain the ground truth of the images. At the same time, this paper also uses CutMix to improve the generalization ability of the model, and the obtained model accuracy is only 1% different from that of the model trained with the original data. The results show that the defense method proposed in this paper can not only resist Generative Adversarial Network attacks in Federated Learning and protect client privacy, but also ensure that the model accuracy of the Federated model will not be greatly affected.

Anonymous cryptocurrencies are distributed applications based on blockchain technology that enhance the anonymity of user transactions. However, they also provide cover for many criminal activities. Existing traceable modifications of anonymous transaction protocols do not support public verification of the traceability of transactions, meaning dishonest users can still evade tracing by forging partial proofs. In this paper, we first propose EPoK, which is a partially extractable zero-knowledge proof scheme designed to address this issue. We provide the instantiation of EPoK, as well as its formal security proofs. Combined with EPoK and the classic anonymous transaction protocol RingCT, we propose a traceable anonymous transaction protocol TRCT for tracing participants' addresses and transaction amounts in anonymous transactions. Compared to previous works, TRCT is the first to implement public verification of the transaction's traceability while still maintaining anonymity. This ensures that users cannot forge relevant proofs to evade tracing. We present and prove the security properties of TRCT, and experimentally compare its efficiency with the original RingCT protocol. The result shows that the extra overhead generated by TRCT is small.

Background As one of the fundamental problems in bioinformatics, the double digest problem (DDP) focuses on reordering genetic fragments in a proper sequence. Although many algorithms for dealing with the DDP problem were proposed during the past decades, it is believed that solving DDP is still very time-consuming work due to the strongly NP-completeness of DDP. However, none of these algorithms consider the privacy issue of the DDP data that contains critical business interests and is collected with days or even months of gel-electrophoresis experiments. Thus, the DDP data owners are reluctant to deploy the task of solving DDP over cloud. Results Our main motivation in this paper is to design a secure outsourcing computation framework for solving the DDP problem. We at first propose a privacy-preserving outsourcing framework for handling the DDP problem by using a cloud server; Then, to enable the cloud server to solve the DDP instances over ciphertexts, an order-preserving homomorphic index scheme (OPHI) is tailored from an order-preserving encryption scheme published at CCS 2012; And finally, our previous work on solving DDP problem, a quantum inspired genetic algorithm (QIGA), is merged into our outsourcing framework, with the supporting of the proposed OPHI scheme. Moreover, after the execution of QIGA at the cloud server side, the optimal solution, i.e. two mapping sequences, would be transferred publicly to the data owner. Security analysis shows that from these sequences, none can learn any information about the original DDP data. Performance analysis shows that the communication cost and the computational workload for both the client side and the server side are reasonable. In particular, our experiments show that PP-DDP can find optional solutions with a high success rate towards typical test DDP instances and random DDP instances, and PP-DDP takes less running time than DDmap, SK05 and GM12, while keeping the privacy of the original DDP data. Conclusion The proposed outsourcing framework, PP-DDP, is secure and effective for solving the DDP problem.

Vision Transformer(ViT)，一种与卷积神经网络截然不同的架构网络，具有多种优势，包括设计简单性，健壮性，并且已经在许多视觉任务上取得sota。然而与卷积神经网络比，ViT缺少归纳偏置，因此需要大量的数据集预训练从中学习归纳偏置，使得在小型数据集上从头开始训练效果并不好。本文目的是设计一个鲁棒的训练小规模数据集的方案。采用两阶段的方式。第一阶段，设计一种自监督学习方案，从小数据集上进行训练，从中学习归纳偏置，作为初始化权重。第二阶段，对ViT图片分割阶段进行优化，并使用初始化权重在优化的ViT模型上，使用小数据集进行微调。通过在多种公开小数据集上进行广泛的实验证明，与现有算法相比，本文提出的方法有更好的表现。

Unsupervised Domain Adaptation (UDA) has been subject to comprehensive investigation and has achieved significant success in real-world scenarios by transferring information from labeled source domains to non-labeled target domains. Nonetheless, the vulnerability of UDA models to adversarial attacks remains a formidable challenge. While Adversarial Training (AT) is acknowledged as one of the most potent defense mechanisms, it cannot be directly applied to UDA settings. Furthermore, there have been few studies that explore the application of AT in UDA setting. In this paper, we strive to leverage Generative Adversarial Network (GAN) to generate adversarial examples for target data, subsequently incorporating them into AT. To generate high-quality adversarial examples and achieve better adversarial robustness of UDA models, we propose the AAT algorithm. We apply AAT to two different UDA algorithms and evaluate on three datasets. The results demonstrate that our model achieves improved adversarial robustness and a balance between accuracy on clean data and accuracy under adversarial conditions.

Dynamic searchable encryption methods allow a client to perform searches and updates over encrypted data stored in the cloud. Recent schemes enable secure searching over an encrypted database stored in a server, but these schemes only focus on hiding the access pattern using ORAM. Although the traditional Oblivious Random Access Machine (ORAM) can hide the access pattern, which refers to the set of documents that match the client’s queries, it also incurs significant communication overhead and cannot hide the search pattern. Existing researches show that the general dynamic searchable symmetric encryption (DSSE) scheme is vulnerable to statistical attacks due to the leakage of both search patterns and access patterns. Therefore hiding the access pattern is not enough, it is essential to hide both access patterns and search patterns with high efficiency. To overcome this limitation, a DSSE scheme called obliviously shuffled incidence matrix DSSE (OSM-DSSE) is proposed in this paper to access the encrypted data obliviously without using ORAM. The OSM-DSSE scheme realizes efficient search and update operations based on an incidence matrix. In particular, a shuffling algorithm using Paillier encryption combines the 1-out-of-n obliviously transfer (OT) protocol to hide access pattern and an optimized padding scheme to obfuscate the search pattern with low storage overhead. Besides, Simulation results and security analysis confirm that OSM-DSSE scheme achieves high security and efficient searches. Also, this scheme provides adaptive security against malicious attacks by adversaries. Furthermore, OSM-DSSE is capable of searching for a keyword out of 9 × 10^10 keyword-file pairs within 2.19 s which is 3–4 × better execution efficiency than the state-of-art solutions.