作者: 團結 時間: 2025-3-21 20:17 作者: 失敗主義者 時間: 2025-3-22 03:26
Reinforcement Learning Approach to?Generate Zero-Dynamics Attacks on?Control Systems Without State Se space. We develop several attackers and detectors iteratively until the attacker and detectors no longer improve. In addition, we also show that the reinforcement learning based attacker successfully executes an attack in the same manner as the theoretical attacker described in previous literature.作者: 親密 時間: 2025-3-22 07:51
DScope: To Reliably and?Securely Acquire Live Data from?Kernel-Compromised ARM Devicesdevelopment board and have also tested . ’s reliability against various forms of denial of service attacks. Our experiments show that a user can dynamically import data acquisition routines to the device to extract kernel objects and runtime stacks from an attack scene or a kernel crashing site.作者: 尾巴 時間: 2025-3-22 11:44 作者: Frenetic 時間: 2025-3-22 15:23 作者: Frenetic 時間: 2025-3-22 20:25 作者: mighty 時間: 2025-3-23 00:18 作者: 我要威脅 時間: 2025-3-23 04:06
PassGPT: Password Modeling and (Guided) Generation with Large Language Modelsre we leverage PassGPT sampling procedure to generate passwords matching arbitrary constraints, a feat lacking in current GAN-based strategies. Lastly, we conduct an in-depth analysis of the entropy and probability distribution that PassGPT defines over passwords and discuss their use in enhancing existing password strength estimators.作者: GLEAN 時間: 2025-3-23 06:58 作者: angina-pectoris 時間: 2025-3-23 09:47
Conference proceedings 202423, which took place in The Hague, The Netherlands, during September 25-29, 2023...The 93 full papers presented in these proceedings were carefully reviewed and selected from?478?submissions. They were organized in topical sections as follows:.Part I: Crypto...Part II:?Network, web and internet; pri作者: amnesia 時間: 2025-3-23 16:41 作者: 放棄 時間: 2025-3-23 19:38 作者: COWER 時間: 2025-3-24 01:10 作者: BALK 時間: 2025-3-24 04:06 作者: 雕鏤 時間: 2025-3-24 09:34 作者: 額外的事 時間: 2025-3-24 11:05
The Power of?MEME: Adversarial Malware Creation with?Model-Based Reinforcement Learning However, machine learning models are susceptible to adversarial attacks, requiring the testing of model and product robustness. Meanwhile, attackers also seek to automate malware generation and evasion of antivirus systems, and defenders try to gain insight into their methods. This work proposes a 作者: 移動 時間: 2025-3-24 17:15 作者: 神圣在玷污 時間: 2025-3-24 20:50
Machine Learning for?SAST: A Lightweight and?Adaptable Approachr study funded by Germany’s Federal Office for Information Security (BSI). SAST describes the practice of applying static analysis techniques to program code on the premise of detecting security-critical software defects early during the development process. In the past, this was done by using rule-作者: 暫時別動 時間: 2025-3-25 00:04 作者: 并排上下 時間: 2025-3-25 03:45 作者: intrude 時間: 2025-3-25 10:28 作者: 連鎖 時間: 2025-3-25 11:41 作者: 下級 時間: 2025-3-25 18:26 作者: Facet-Joints 時間: 2025-3-25 21:23
Efficient Pruning for?Machine Learning Under Homomorphic Encryptioners confidentiality of the model and the data, but at the cost of large latency and memory requirements. Pruning neural network (NN) parameters improves latency and memory in plaintext ML but has little impact if directly applied to HE-based PPML..We introduce a framework called . that comprises new作者: 情愛 時間: 2025-3-26 03:54 作者: foreign 時間: 2025-3-26 04:31
On the?(In)Security of?Manufacturer-Provided Remote Attestation Frameworks in?Androidroduced Manufacturer-provided Android Remote Attestation (MARA) frameworks. The MARA framework helps an app conduct a series of integrity checks, signs the check results, and sends them to remote servers for a remote attestation. Nonetheless, we observe that real-world MARA frameworks often adopt tw作者: Alveoli 時間: 2025-3-26 09:22 作者: FEMUR 時間: 2025-3-26 15:17
: Split Input-to-State Mapping for?Effective Firmware Fuzzingmated testing methods rehost firmware in emulators and attempt to facilitate inputs from a diversity of methods (interrupt driven, status polling) and a plethora of devices (such as modems and GPS units). Despite recent progress to tackle peripheral input generation challenges in rehosting, a firmwa作者: 偏狂癥 時間: 2025-3-26 17:28
IPS: Software-Based Intrusion Prevention for?Bare-Metal Embedded Systemsk many security primitives, including the well-known Address Space Layout Randomization (ASLR) and Data Execution Prevention (DEP), and their integrity can be compromised using a single vulnerability. Proposed defenses have not yet been deployed due to their requirements for firmware source code ava作者: Spartan 時間: 2025-3-26 23:18
Aion: Secure Transaction Ordering Using TEEsroadcast their commands to all processes. This is impractical due to the impact on scalability, and thus it discourages the adoption of a fair ordering of commands. Alternative approaches to order-fairness allow clients do send their commands to only one process, but provide a weaker notion of order作者: Palpable 時間: 2025-3-27 03:05 作者: hidebound 時間: 2025-3-27 05:21
https://doi.org/10.1007/978-3-642-11273-7t grids, and vehicular networks. This paper investigates a subset of stealthy attacks known as zero-dynamics-based stealthy attacks. While previous works on zero-dynamics attacks have highlighted the necessity of highly accurate knowledge of the system’s state space for generating attack signals, ou作者: 擴張 時間: 2025-3-27 13:15 作者: Additive 時間: 2025-3-27 17:18
https://doi.org/10.1007/978-3-642-11273-7 However, machine learning models are susceptible to adversarial attacks, requiring the testing of model and product robustness. Meanwhile, attackers also seek to automate malware generation and evasion of antivirus systems, and defenders try to gain insight into their methods. This work proposes a 作者: optic-nerve 時間: 2025-3-27 20:13
https://doi.org/10.1007/978-3-642-11273-7ir private training datasets. Therefore, without revealing the private dataset, the clients can obtain a deep learning (DL) model with high performance. However, recent research proposed poisoning attacks that cause a catastrophic loss in the accuracy of the global model when adversaries, posed as b作者: 野蠻 時間: 2025-3-27 23:00
https://doi.org/10.1007/978-3-642-11273-7r study funded by Germany’s Federal Office for Information Security (BSI). SAST describes the practice of applying static analysis techniques to program code on the premise of detecting security-critical software defects early during the development process. In the past, this was done by using rule-作者: Saline 時間: 2025-3-28 05:12
The Disintegration of the American Empirese attacks against the next word prediction model used in Google’s GBoard app, a widely used mobile keyboard app that has been an early adopter of federated learning for production use. We demonstrate that the words a user types on their mobile handset, e.g. when sending text messages, can be recover作者: Aphorism 時間: 2025-3-28 07:09 作者: SSRIS 時間: 2025-3-28 10:25 作者: BOAST 時間: 2025-3-28 17:45 作者: Gudgeon 時間: 2025-3-28 20:53
The Colonial Experience in French Fictionlls are non-trivial without knowing the types of objects at compile time. Addressing this challenge, . is increasingly added to dynamically-typed languages, a prominent example being TypeScript that introduces static typing to JavaScript. Gradual typing improves the developer’s ability to verify pro作者: 芳香一點 時間: 2025-3-29 00:31 作者: antiandrogen 時間: 2025-3-29 03:22
The Colonial Experience in French Fictionately, measuring and comparing the effectiveness of various debloating methods is challenging due to the absence of a universal benchmarking platform that can accommodate diverse approaches. In this paper, we first present . (.ing .mark for .pplications), an extensible and sustainable benchmarking p作者: 波動 時間: 2025-3-29 11:00
https://doi.org/10.1057/9780333982907roduced Manufacturer-provided Android Remote Attestation (MARA) frameworks. The MARA framework helps an app conduct a series of integrity checks, signs the check results, and sends them to remote servers for a remote attestation. Nonetheless, we observe that real-world MARA frameworks often adopt tw作者: Manifest 時間: 2025-3-29 13:36 作者: 滲透 時間: 2025-3-29 17:18 作者: PSA-velocity 時間: 2025-3-29 22:55
https://doi.org/10.1007/978-3-030-65836-6k many security primitives, including the well-known Address Space Layout Randomization (ASLR) and Data Execution Prevention (DEP), and their integrity can be compromised using a single vulnerability. Proposed defenses have not yet been deployed due to their requirements for firmware source code ava作者: 偽書 時間: 2025-3-30 02:20
Race-Thinking and the Parsi Social Dramaroadcast their commands to all processes. This is impractical due to the impact on scalability, and thus it discourages the adoption of a fair ordering of commands. Alternative approaches to order-fairness allow clients do send their commands to only one process, but provide a weaker notion of order作者: BROW 時間: 2025-3-30 05:42 作者: reserve 時間: 2025-3-30 10:56
978-3-031-51481-4The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerl作者: 沒有貧窮 時間: 2025-3-30 14:21 作者: Peristalsis 時間: 2025-3-30 19:41 作者: EWER 時間: 2025-3-30 23:16 作者: Offset 時間: 2025-3-31 03:16
https://doi.org/10.1007/978-3-642-11273-7y distributed (non-IID). In this work, we propose FLGuard, a novel byzantine-robust FL method that detects malicious clients and discards malicious local updates by utilizing the contrastive learning technique, which showed a tremendous improvement as a self-supervised learning method. With contrast
