密码学与安全

DNS integrations leverage the discovery, trust, and uniqueness of the global Domain Name System with a linkage to another naming ecosystem, so the DNS name can help identify resources such as a cryptocurrency wallet or software component.…

The transition to post-quantum cryptography in blockchain systems such as Bitcoin and Ethereum is often framed as a purely cryptographic problem. In practice, it also presents significant economic and infrastructural challenges: in globally…

Misbehavior detection in Vehicle-to-Everything (V2X) networks is a second line of defense against insider falsification attacks that cryptographic mechanisms alone cannot address. Existing learning-based Misbehavior Detection Schemes (MDSs)…

We demonstrate that language models can autonomously replicate their weights and harness across a network by exploiting vulnerable hosts. The agent independently finds and exploits a web-application vulnerability, extracts credentials, and…

Attackers willing to compromise computing systems can use malicious peripherals as an attack vector, threatening users that cannot verify the hardware's authenticity. To address this problem, our work uses the Security Protocol and Data…

Autonomous AI agents now transact at production scale -- 69,000 bots executing 165 million transactions across 50 million USDC in cumulative volume on a single marketplace -- without any shared trust layer between participants. Regulatory…

Personalized LLM agents maintain persistent cross-session state to support long-horizon collaboration. Yet, this persistence introduces a subtle but critical security vulnerability: routine user-agent interactions can gradually reshape an…

Malware and malware-based attacks are becoming more prevalent and complex. Attackers regularly come up with new techniques that have the ability to evade conventional and signature-based malware defense. In order to address such threats,…

Agentic AI systems can plan, call tools, inspect code, interact with web applications, and coordinate multi-step workflows. These same capabilities change the economics of cyber offense. The central near-term risk is not that every…

The safety alignment of Large Language Models (LLMs) remains vulnerable to Harmful Fine-tuning (HFT). While existing defenses impose constraints on parameters, gradients, or internal representations, we observe that they can be effectively…

As Vision-Language Models (VLMs) are increasingly deployed as autonomous cognitive cores for embodied assistants, evaluating their privacy awareness in physical environments becomes critical. Unlike digital chatbots, these agents operate in…

Preprints are essential for the timely and open dissemination of research. arXiv, the most widely used preprint service, takes the idea of open science one step further by not only publishing the actual preprints but also LaTeX sources and…

The ability to simulate human privacy decisions has significant implications for aligning autonomous agents with individual intent and conducting cost-effective, large-scale privacy-centric user studies. Prior approaches prompt Large…

Homomorphic encryption (HE) is a prominent framework for privacy-preserving machine learning, enabling inference directly on encrypted data. However, evaluating softmax, a core component of transformer architectures, remains particularly…

As large language models (LLMs) become integral to applications such as question answering and content creation, reliable content attribution has become increasingly important. Watermarking is a promising approach, but most existing methods…

We present SWaRL, a robust and fidelity-preserving watermarking framework designed to protect the intellectual property of code LLMs by embedding unique and verifiable signatures in the generated program. Existing watermarking approaches…

In-context learning (ICL) allows LLMs to adapt to new tasks via a few demonstrations, but those demonstrations may contain sensitive data. Differentially private (DP) ICL mechanisms mitigate this risk by injecting noise into the aggregation…

Large language models (LLMs) have become indispensable for automated code generation, yet the quality and security of their outputs remain a critical concern. Existing studies predominantly concentrate on adversarial attacks or inherent…

Smart grids are exposed to passive eavesdropping, where attackers listen silently to communication links. Although no data is actively altered, such reconnaissance can reveal grid topology, consumption patterns, and operational behavior,…

Standard techniques for differentially private estimation, such as Laplace or Gaussian noise addition, require guaranteed bounds on the sensitivity of the estimator in question. But such sensitivity bounds are often large or simply unknown.…