arXiv subjects

Ali Arastehfard

Publications and source records attributed to Ali Arastehfard.

3 recordsLinked to original sources

arXiv2026

SoK: Systematizing LLM Prompt Security: Taxonomies, Datasets, and Unified Evaluation of Attacks and Defenses

Large Language Models (LLMs) are increasingly used as interfaces to information, code, and real-world services, making prompt-level security failures a practical concern. Although jailbreak attacks, defenses, datasets, and automated judgers have advanced rapidly, evaluation remains fragmented across threat models, access assumptions, cost budgets, datasets, and success criteria. This makes reported attack success rates and defense gains hard to compare. This SoK systematizes LLM prompt security across concepts, data, tooling, and measurement. We propose linked taxonomies for jailbreak attacks, defenses, and model vulnerabilities, while separating technical mechanisms from attacker and defender capabilities. We also formalize threat, access, and cost assumptions as explicit evaluation metadata. To support reproducible evaluation, we release JailbreakDB, PromptSecurity-Eval, and PromptSecurity, a modular platform that represents each experiment as a tuple of model, attack, defense, dataset, and judger. Using matched evaluations across models, attacks, defenses, and judgers, we show that access regime, native harmful-query behavior, attack cost, defense backfire, taxonomy subcategory, and judger choice all materially affect security conclusions. Together, these artifacts support reproducible, cost-aware, and taxonomy-grounded evaluation of LLM prompt security. Leaderboard: https://datasec-lab.github.io/PromptSecurityLeaderboard/. Dataset: https://huggingface.co/datasets/youbin2014/JailbreakDB. GitHub: https://github.com/datasec-lab/PromptSecurity.

cs.CR↗

arXiv2025

Secure and Efficient $L^p$-Norm Computation for Two-Party Learning Applications

Secure norm computation is becoming increasingly important in many real-world learning applications. However, existing cryptographic systems often lack a general framework for securely computing the $L^p$-norm over private inputs held by different parties. These systems often treat secure norm computation as a black-box process, neglecting to design tailored cryptographic protocols that optimize performance. Moreover, they predominantly focus on the $L^2$-norm, paying little attention to other popular $L^p$-norms, such as $L^1$ and $L^\infty$, which are commonly used in practice, such as machine learning tasks and location-based services. To our best knowledge, we propose the first comprehensive framework for secure two-party $L^p$-norm computations ($L^1$, $L^2$, and $L^\infty$), denoted as \mbox{Crypto-$L^p$}, designed to be versatile across various applications. We have designed, implemented, and thoroughly evaluated our framework across a wide range of benchmarking applications, state-of-the-art (SOTA) cryptographic protocols, and real-world datasets to validate its effectiveness and practical applicability. In summary, \mbox{Crypto-$L^p$} outperforms prior works on secure $L^p$-norm computation, achieving $82\times$, $271\times$, and $42\times$ improvements in runtime while reducing communication overhead by $36\times$, $4\times$, and $21\times$ for $p=1$, $2$, and $\infty$, respectively. Furthermore, we take the first step in adapting our Crypto-$L^p$ framework for secure machine learning inference, reducing communication costs by $3\times$ compared to SOTA systems while maintaining comparable runtime and accuracy.

cs.CR↗

arXiv2025

SecureV2X: An Efficient and Privacy-Preserving System for Vehicle-to-Everything (V2X) Applications

Autonomous driving and V2X technologies have developed rapidly in the past decade, leading to improved safety and efficiency in modern transportation. These systems interact with extensive networks of vehicles, roadside infrastructure, and cloud resources to support their machine learning capabilities. However, the widespread use of machine learning in V2X systems raises issues over the privacy of the data involved. This is particularly concerning for smart-transit and driver safety applications which can implicitly reveal user locations or explicitly disclose medical data such as EEG signals. To resolve these issues, we propose SecureV2X, a scalable, multi-agent system for secure neural network inferences deployed between the server and each vehicle. Under this setting, we study two multi-agent V2X applications: secure drowsiness detection, and secure red-light violation detection. Our system achieves strong performance relative to baselines, and scales efficiently to support a large number of secure computation interactions simultaneously. For instance, SecureV2X is $9.4 \times$ faster, requires $143\times$ fewer computational rounds, and involves $16.6\times$ less communication on drowsiness detection compared to other secure systems. Moreover, it achieves a runtime nearly $100\times$ faster than state-of-the-art benchmarks in object detection tasks for red light violation detection.

cs.CR↗