FPGA Acceleration of Fully Homomorphic Encryption with Adaptive Key Switching
本文针对全同态加密中的密钥切换性能瓶颈问题，提出了一种自适应选择方法，并设计了基于FPGA的加速器，显著提高了计算速度。
0 citationsRead paper
Selected work
Representative Papers
Generative AI and Federated Learning for Intrusion Detection Systems: A Survey
This study addresses the challenges confronting intrusion detection systems (IDS), including dynamically evolving attacks, data scarcity, class imbalance, and the difficulty of centralized training under strict privacy constraints. It presents a systematic review of the integration of generative artificial intelligence—encompassing generative adversarial networks (GANs), diffusion models, and large language models—with federated learning in IDS applications. The work covers key areas such as anomaly detection, synthetic data generation, data augmentation, and privacy-preserving distributed training. For the first time, it offers a structured synthesis of recent advances at the intersection of these two technological paradigms and outlines promising future directions, including domain-specific large language models and federated benchmarking frameworks, thereby establishing a novel paradigm for privacy-sensitive, distributed cybersecurity solutions.
0 citationsRead paper
NetDiffuser: Deceiving DNN-Based Network Attack Detection Systems with Diffusion-Generated Adversarial Traffic
This work proposes NetDiffuser, a novel framework for generating natural adversarial examples against deep learning-based network intrusion detection systems (NIDS), which are vulnerable to such attacks yet resistant to existing adversarial traffic generation methods. NetDiffuser innovatively integrates feature disentanglement with diffusion models: it first identifies semantically meaningful and relatively independent features within network traffic through feature decomposition, then leverages a diffusion model to inject perturbations that preserve semantic consistency while maximizing realism. The resulting adversarial samples exhibit high fidelity and effectiveness across diverse NIDS architectures. Experimental results on three benchmark datasets demonstrate that NetDiffuser achieves up to a 29.93% increase in attack success rate and reduces the AUC-ROC of adversarial sample detection by as much as 0.534, significantly outperforming current state-of-the-art baselines.
0 citationsRead paper
A data-driven discretized CS:GO simulation environment to facilitate strategic multi-agent planning research
This work addresses the challenge of simultaneously achieving strategic abstraction and environmental fidelity in multi-agent simulation. We propose DECOY, a data-driven discrete simulator that replaces low-level physical modeling (e.g., aiming, shooting) with a waypoint-based state-action discretization framework. A neural predictive model is trained directly on professional CS:GO match data to learn mappings between high-level movement decisions and tactical outcomes. Our key contribution is the first demonstration that high-fidelity combat replays—statistically indistinguishable from original gameplay at the strategic level—can be reconstructed solely from coarse-grained mobility decisions, without simulating micro-actions. Evaluation shows an average trajectory similarity exceeding 92% against ground-truth game traces. DECOY is open-sourced, providing an efficient, interpretable, and scalable simulation platform for long-horizon multi-agent planning research in 3D environments.
0 citationsRead paper
Recent publications
Latest Papers
FPGA Acceleration of Fully Homomorphic Encryption with Adaptive Key Switching
本文针对全同态加密中的密钥切换性能瓶颈问题，提出了一种自适应选择方法，并设计了基于FPGA的加速器，显著提高了计算速度。
0 citationsRead paper
Generative AI and Federated Learning for Intrusion Detection Systems: A Survey
This study addresses the challenges confronting intrusion detection systems (IDS), including dynamically evolving attacks, data scarcity, class imbalance, and the difficulty of centralized training under strict privacy constraints. It presents a systematic review of the integration of generative artificial intelligence—encompassing generative adversarial networks (GANs), diffusion models, and large language models—with federated learning in IDS applications. The work covers key areas such as anomaly detection, synthetic data generation, data augmentation, and privacy-preserving distributed training. For the first time, it offers a structured synthesis of recent advances at the intersection of these two technological paradigms and outlines promising future directions, including domain-specific large language models and federated benchmarking frameworks, thereby establishing a novel paradigm for privacy-sensitive, distributed cybersecurity solutions.
0 citationsRead paper
NetDiffuser: Deceiving DNN-Based Network Attack Detection Systems with Diffusion-Generated Adversarial Traffic
This work proposes NetDiffuser, a novel framework for generating natural adversarial examples against deep learning-based network intrusion detection systems (NIDS), which are vulnerable to such attacks yet resistant to existing adversarial traffic generation methods. NetDiffuser innovatively integrates feature disentanglement with diffusion models: it first identifies semantically meaningful and relatively independent features within network traffic through feature decomposition, then leverages a diffusion model to inject perturbations that preserve semantic consistency while maximizing realism. The resulting adversarial samples exhibit high fidelity and effectiveness across diverse NIDS architectures. Experimental results on three benchmark datasets demonstrate that NetDiffuser achieves up to a 29.93% increase in attack success rate and reduces the AUC-ROC of adversarial sample detection by as much as 0.534, significantly outperforming current state-of-the-art baselines.
0 citationsRead paper
A data-driven discretized CS:GO simulation environment to facilitate strategic multi-agent planning research
This work addresses the challenge of simultaneously achieving strategic abstraction and environmental fidelity in multi-agent simulation. We propose DECOY, a data-driven discrete simulator that replaces low-level physical modeling (e.g., aiming, shooting) with a waypoint-based state-action discretization framework. A neural predictive model is trained directly on professional CS:GO match data to learn mappings between high-level movement decisions and tactical outcomes. Our key contribution is the first demonstration that high-fidelity combat replays—statistically indistinguishable from original gameplay at the strategic level—can be reconstructed solely from coarse-grained mobility decisions, without simulating micro-actions. Evaluation shows an average trajectory similarity exceeding 92% against ground-truth game traces. DECOY is open-sourced, providing an efficient, interpretable, and scalable simulation platform for long-horizon multi-agent planning research in 3D environments.
0 citationsRead paper