With the development of quantum computing technology, the threat of quantum computing to classical cryptography has been significantly enhanced. Studying the impact of quantum computing on the security of classical cryptosystems is a research focus in the cryptographic community. Pseudorandom functions are fundamental building blocks in numerous cryptosystems and play a critical role in post-quantum cryptography. To the best of our knowledge, research on effective quantum attacks on pseudorandom functions remains limited in the noisy intermediate-scale quantum era. This paper proposes a distributed quantum key recovery attack on pseudorandom functions and designs the corresponding quantum circuit based on distributed quantum computing technology, Bernstein-Vazirani’s and Grover’s algorithms. The proposed attack is applied to two pseudorandom functions, pEDM and SoKAC. Compared with the existing attacks, the proposed attack has lower complexity and stronger robustness to circuit noise. The proposed attack can realize effective quantum key recovery and provide theoretical support for designing quantum secure pseudorandom functions.

Variational quantum circuits face a critical trade-off between privacy and trainability. High expressivity required for robust privacy induces exponentially large dynamical Lie algebras. This structure inevitably leads to barren plateaus. Conversely, trainable models restricted to polynomial-sized algebras remain transparent to algebraic attacks. To resolve this impasse, DyLoC is proposed. This dual-layer architecture employs an orthogonal decoupling strategy. Trainability is anchored to a polynomial-DLA ansatz while privacy is externalized to the input and output interfaces. Specifically, Truncated Chebyshev Graph Encoding (TCGE) is employed to thwart snapshot inversion. Dynamic Local Scrambling (DLS) is utilized to obfuscate gradients. Experiments demonstrate that DyLoC maintains baseline-level convergence with a final loss of 0.186. It outperforms the baseline by increasing the gradient reconstruction error by 13 orders of magnitude. Furthermore, snapshot inversion attacks are blocked when the reconstruction mean squared error exceeds 2.0. These results confirm that DyLoC effectively establishes a verifiable pathway for secure and trainable quantum machine learning.

The development of quantum computing technology poses a series of threats and challenges to the security of classical cryptosystems. In order to accurately evaluate the security of symmetric primitives in the post-quantum era, it is essential to study the quantum versions of classical cryptographic cryptanalysis on symmetric cryptosystems. However, due to the limitations of noise and depth of quantum circuits, it is challenging to implement cryptanalysis based on large-scale quantum circuits at the present stage. In this paper, we propose a full-phase distributed quantum differential cryptanalysis, which is more suitable for the current noisy intermediate-scale quantum computation model. Then, we design two variants, including a full-phase distributed quantum impossible differential cryptanalysis and a full-phase distributed quantum truncated differential cryptanalysis. We rigorously analyze the correctness and complexity of all cryptanalysis. Compared with classical cryptanalysis, the proposed cryptanalysis provides a quadratic speedup and can avoid the problem of differential paths being difficult to expand. Compared with the existing quantum cryptanalysis, the proposed cryptanalysis has lower query complexity and better robustness to circuit noise.

Quantum cloud platforms face the challenge of quantum circuit Trojan attack. Trojan attack in quantum circuits originates from dual-mode attack involving parameter implantation and structure implantation, threatening the secure execution of quantum programs in cloud environments. To address the difficulty of detecting dual-mode quantum circuit Trojan attack in quantum cloud platforms, we propose a dual-mode quantum circuit Trojan attack detection scheme based on unitary matrix features. By constructing dual-mode quantum Trojan attack and evaluating it with quantitative metrics (accuracy drop, JS divergence, etc.), a quantum circuit is represented by a unitary matrix, from which real and imaginary component features are extracted. A convolutional neural network model is deployed for detection on 1550 benchmark circuits. Experimental results indicate that our scheme achieves an accuracy of 91.6

In fault-tolerant quantum computing systems, the T gates consume more fault-tolerant resources. In this paper, we propose an automated optimization method based on the Proximal Policy Optimization (PPO) algorithm. We first translate quantum circuits into ZX-diagrams using ZX-calculus, transforming the circuit optimization problem into a graph rewriting task. Subsequently, the PPO agent, powered by graph neural networks (GNNs) to encode the complex graph structures of the ZX-diagrams, learns to predict efficient transformation paths by focusing on local node and edge features. The PPO algorithm ensures stable policy updates through its clipped objective function, which prevents excessively large and potentially detrimental changes during optimization. Experimental results confirm the method’s effectiveness, showing an average reduction in T-count of 3.12% compared to the baseline approach.

High-quality network traffic data labeling is a critical task for network security, enabling precise, fine-grained representations of traffic flows. However, existing labeling approaches still face severe limitations in no-ground-truth network scenarios, including high manual effort, imprecise labeling, and an over-reliance on often-inaccessible auxiliary information. Consequently, high-quality labeled datasets remain scarce, significantly hindering the advancement of data-driven security techniques. Thus, we propose TSA, a Temporal Sequence Alignment algorithm designed for automated, fine-grained traffic labeling in no-ground-truth scenarios. TSA operates solely on packet sequences recorded in source-side logs and packet sequences captured from the network. By enforcing rigorous payload, delta-time, and chronological order constraints, the algorithm automatically aligns sequences at the packet level, achieving high labeling fidelity even in no-ground-truth scenarios. We evaluate TSA across 15 scenarios derived from CVE reproduction cases, public network datasets and real-world attacks. As a result, TSA achieves a labeling accuracy of up to 0.9982 and a throughput of 225 packets per second, representing an effective and practical solution for high-fidelity traffic labeling in no-ground-truth scenarios.

To break through the limitations of existing hash functions based on quantum walks with memory in diffusion efficiency and step-length control, we develop a memory-driven controlled alternate quantum walks model and propose a corresponding hash function. Firstly, the proposed model integrates lively quantum walks with quantum walks with memory and introduces a two-step memory decision mechanism. This mechanism allows historical memory to dynamically regulate the jumping amplitude in the active direction, thereby strengthening the coupling between memory and the coin operator and accelerating the diffusion of the path. Secondly, on the basis of this model, we propose a quantum hash function with variable-length output. Experimental evaluations demonstrate strong input sensitivity, robust collision resistance, and effective output confusion. Moreover, the output distribution is balanced, the perturbation response is highly sensitive, and a pronounced avalanche effect is consistently observed. Our work introduces a novel path control paradigm for the systematic design of quantum hash functions built on quantum walks.

The development of post-quantum cryptography requires cryptographic primitives that remain secure against quantum adversaries. Quantum key-recovery attacks constitute a fundamental class of quantum cryptanalytic techniques, and the resistance of a cryptosystem to such attacks serves as an important indicator of its overall security. Since the Feistel structure is a core component in the design of many block ciphers, efficient quantum key-recovery attacks on Feistel ciphers provide valuable insights into evaluating and strengthening their security. In this paper, we propose an efficient generic quantum key-recovery attack on Feistel ciphers. Our attack integrates universal hash families with Simon's and Grover's algorithms to perform key-recovery. We rigorously analyze the correctness of the proposed attack and derive its time and quantum resource complexities. A comprehensive comparison with existing keyrecovery attacks shows that the proposed approach improves both quantum resource usage and dominant query complexity, and therefore yields a more efficient quantum key-recovery attack. Our work enhances the understanding of Feistel ciphers under quantum threats and informs the design of robust cryptosystems for the post-quantum era.

Advanced Persistent Threat (APT) poses a serious security challenge because of their stealthy, multi-stage attack lifecycle. These threats can persist undetected for months, causing severe financial and reputational damage. Moreover, the growing volume and diversity of network traffic amplify the difficulty of real-time identification. Existing APT detectors-ranging from recurrent and convolutional neural networks to single-stage Transformers-often fail to jointly capture local features and global dependencies efficiently. In this paper, our model introduces a Hierarchical Local-Global Co-Attention (HLCDA) for robust and efficient APT detection. HLCDA comprises three encoders. Firstly, an xLSTM block with a Time-Scale Gate (TSG) for multi-scale temporal dependencies. Secondly, a Multi-Scale Depthwise Separable CNN (MS-DS-CNN) for short-range spatial features. Thirdly, a RoPE-enhanced Transformer (RoPET) for global contextual modeling. A key advancement is a hierarchical co-attention mechanism that enables bidirectional interaction between the TSG-xLSTM and MS-DS-CNN streams prior to fusion. The model then fuses these representations through a Dynamic Multi-Head Feature Fusion (DMFF) module. Evaluated on three real-world benchmarks (DAPT 2020, UNSW-NB15, and CIC-IDS2018), HLCDA achieved 94.60% accuracy and a 95.72% F1 score on DAPT 2020, 93.12% accuracy and 93.00% F1 on UNSW-NB15, and 92.84% accuracy and 92.54% F1 on CIC-IDS2018, while maintaining sub-millisecond inference latency, demonstrating strong capability to distinguish sophisticated APT campaigns from benign traffic. This advancement offers organizations a practical and scalable defense, strengthening resilience against evolving APT threats. (c) 2026 Elsevier Inc. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

With the rapid advancement of quantum computing, quantum compilation has become a crucial layer connecting high-level algorithms with physical hardware. In quantum cloud computing, compilation is performed on the cloud platforms, which expose user circuits to potential risks, such as structural leakage and output predictability. To address these issues, we propose the encrypted-state quantum compilation scheme based on quantum circuit obfuscation (ECQCO), the first secure compilation scheme tailored for the co-location of compilers and quantum hardware for quantum cloud platforms. It applies quantum homomorphic encryption to conceal output states and instantiates a structure obfuscation mechanism based on quantum indistinguishability obfuscation, effectively protecting both functionality and topology of the circuit. In addition, an adaptive decoupling obfuscation algorithm is designed to suppress potential idle errors while inserting pulse operations. The proposed scheme achieves information-theoretic security and guarantees computational indistinguishability under the quantum random oracle. Experimental results on benchmark datasets demonstrate that ECQCO achieves a total variation distance of up to 0.7 and a normalized graph edit distance of 0.88, enhancing compilation-stage security. Moreover, it introduces only a slight increase in circuit depth, while keeping the average fidelity change within 1.1%, thus achieving a practical balance between security and efficiency.

In fault-tolerant quantum computing systems, T gates consume more fault-tolerant resources. In this paper, we propose a T-count reduction method based on the Proximal Policy Optimization (PPO) algorithm, minimizing the number of T gates in quantum computations. Initially, within the framework of ZX-calculus graphical language, quantum circuits are transformed into ZX-diagrams. Subsequently, the PPO algorithm is employed to learn a policy that predicts optimal transformation trajectories. To effectively leverage the topological structure of ZX-diagrams, we employ graph neural networks (GNNs) to encode the policy trained via PPO algorithm, while identifying possible transformations through the local structural properties of individual nodes or edges. The proposed method achieves an average 10.17

The pulse‐level quantum control presents a large range of external parameter dependencies, including control field noise, frequency detuning, nonlinearities, and uncertainty of Hamiltonian parameters, which can lead to significant deviation from the target quantum gate. These terms are not usually considered directly in standard optimization scenarios for robustness, but are often found in analytical solutions. The latter are often difficult to emerge and generalize to different settings. This paper proposes a diffusion‐based reinforcement learning method for pulse‐level quantum robust control (PQC‐DBRL) to enhance the robustness of pulse‐level quantum gate control. PQC‐DBRL does not require an accurate Hamiltonian model of the underlying system, effectively mitigating deviations from target quantum gates caused by control field noise and parameter uncertainties. The quantum pulse control problem is formulated as a conditional generative modeling task, leveraging diffusion reinforcement learning to capture unobserved system information. Furthermore, the results show that PQC‐DBRL pulses maintain a fidelity greater than 0.95 for 100% of the cases and greater than 0.999 for 32.16% of the cases, outperforming GRAPE, which achieves 0.999 fidelity for only 12.48% of the cases under the same noise conditions. In large‐scale experiments with repeated gate operations, PQC‐DBRL demonstrates significantly higher resilience to cumulative errors, maintaining fidelity advantages even after 200 gate repetitions. Additionally, when evaluated across different Hamiltonian variations, PQC‐DBRL shows smaller fidelity variance compared to GRAPE, indicating higher robustness against system parameter fluctuations. This paper offers a promising solution to scalable, noise‐resilient quantum control in practical quantum computing applications.

Gene Multiple Sequence Alignment is crucial for genomic data analysis, forming the basis for studying its biological significance. The digitization of genomic data allows collaborative analysis on cloud platforms, improving the efficiency and precision of genomic research. However, gene sequences contain sensitive information, posing a risk of privacy leakage with unauthorized access. Balancing privacy, accuracy, and efficiency in multiple sequence alignment for long gene sequences remains a challenge. In this paper, we propose a distributed privacy-preserving multiple sequence alignment scheme for long sequences based on secure multi-party computation. Our scheme includes a method for segmenting long sequences to achieve partially distributed computing and a privacy-preserving method for calculating edit distance among subsequences using secret sharing. The scheme consists of a distributed computing phase and an aggregate computing phase, optimizing efficiency by dropping repeated subsequences alignment. Our proposed scheme achieves accurate and efficient privacy-preserving alignment for long gene sequences.

Quantum network coding provides a solution for information transmission at bottleneck channels. However, the single-path scheduling approach is hampered in the face of massive data transmission demands, which is particularly challenging in large-scale network communication scenarios. Here we propose a continuous-variable quantum network coding (CVQNC) scheme for parallel transmission of multiple quantum states based on optical frequency combs. The quantum communication channel is naturally constructed due to the nonclassical correlations exhibited by the symmetric sideband modes at around half the pump frequency on the squeezed field. Source nodes with preshared multiplex entanglement enable the transmission of multiple quantum states in parallel and simultaneously, which increases the channel capacity of the network and relieves the single-path transmission pressure. In particular, fidelity is calculated as a critical measure of transmission performance, and we provide a method to increase fidelity using photon catalysis, along with squeezing parameters and beam splitter transmittance for non-Gaussian source states at optimal fidelity. Furthermore, the channel capacity and fidelity of the CVQNC system are predicted on the basis of the k-nearest neighbor algorithm, and the predicted data match well with the actual derivation.

Quantum network coding aims to address the communication congestion problem of bottleneck channels. Different from existing methods, we propose a counterfactual quantum network coding protocol. This protocol requires not only no physical particle traveling through the bottleneck network but also no pre-shared entanglement between the senders, and thus, it provides a new perspective for the development of quantum network coding theory.

The continuous development of quantum computing technology has brought potential threats to the traditional cryptographic system, which has attracted the attention of the cryptographic community. As quantum computing enters the noisy intermediate-scale quantum era, quantum computing models are constrained by quantum resources and circuit noise. It is necessary to evaluate the security of cryptographic primitives accurately, combined with the development status of quantum computing. In this paper, we propose a full-phase distributed quantum impossible differential cryptanalysis by combining the Bernstein-Vazirani algorithm, quantum phase estimation algorithm, and quantum counting algorithm with the miss-in-the-middle technique. We rigorously analyze the correctness and complexity of the proposed cryptanalysis and design the corresponding distributed quantum circuits. Compared with the classical impossible cryptanalysis, our cryptanalysis avoids the influence of the number of encryption rounds on the cryptanalysis results and has lower complexity. Compared with the existing quantum differential cryptanalysis, the proposed cryptanalysis has lower complexity, shallower circuit depth, and stronger robustness to circuit noise.

Quantum network coding enables multi-party quantum interactions over bottleneck channels. The communication distance is usually constrained by the instability of quantum states. Twin-field quantum key distribution is designed to improve the communication distance, but it can only be applied in point-to-point scenarios. In this paper, we propose a feasible twin-field quantum network coding scheme, which enables multi-party key distribution over long distance in networks with bottleneck channels. Firstly, a butterfly network model is designed for long-distance communication. Then, the coding method at each intermediate node is designed to overcome the limitations of bottleneck channels. Finally, the key rate and decoy-state transmission features are derived to demonstrate the security and efficiency advantages. Compared with the original point-to-point twin-field quantum key distribution protocol, the proposed scheme can distribute a group key among the four end nodes of the butterfly network with only one transmission round. The key rate R and the channel transmittance eta still satisfy R proportional to eta 1/2. Such results will extend the communication distance and provide a foundation for the construction of large-scale quantum networks.

A smart grid is a large-scale cyber-physical fusion system where the physical layer and the cyber layer are interdependent. Attackers only need to invade the cyber layer, relying on its control function and cascading failures, to paralyze the smart grid. To analyze the real interaction between adversaries in the smart grid, we propose a new multistage adversarial game model that protects smart grid against cyber-physical attacks. Attackers can destroy the regional control center of the cyber layer, whereas defenders can strengthen the physical layer nodes. The solution of such game is based on the Minimax-Q algorithm. Both attackers and defenders can actively participate in the game and maximize their minimum rewards in this multistage game. Cascading failures and optimal load shedding are utilized to design the reward function. We apply this model to IEEE 39-bus system and compare it with a nonadversarial game. When defenders are passive players, the load loss is 52.48 % after two actions by the attacker. When defenders are active players, it takes at least four pairs of confrontation actions to inflict a 50% load loss. The results show that attacks on the cyber layer are more destructive and also give rise to various defense strategies.