密码学与安全

Multi-tenant retrieval-augmented generation (RAG) services advertise per-account differential privacy as the operative leakage boundary: each account's queries are guaranteed to satisfy $(\varepsilon_{\text{acc}}, \delta_{\text{acc}})$-DP…

Do stock safety-aligned language models and their uncensored or abliterated derivatives behave differently when run as autonomous security agents? Single-turn refusal benchmarks cannot answer this question: security agents must inspect…

Bitcoin recently introduced a new protocol for the encryption of peer-to-peer (P2P) communication. The protocol, known as V2 P2P transport, represents a big step towards securing the overlay network against various previously-known attack…

Text-to-image diffusion models are increasingly developed through open-source reuse and repeated downstream fine-tuning, where reused checkpoints are difficult to verify and thus more susceptible to hidden backdoor behaviors. In such…

Critical-infrastructure operators are increasingly expected to assess and remediate vulnerabilities in deployed industrial software. However, much of this software exists as opaque industrial software (OIS), including stripped firmware,…

Knowledge Graphs (KGs) are a powerful representation of linked data, offering flexibility, semantic richness, and support for knowledge enrichment and reasoning. They help data owners organize and exploit heterogeneous data to provide…

Existing ViT backdoor attacks based on backbone-overwriting full-tuning are computationally expensive and inflict performance degradation. This has forced adversaries towards the Visual Parameter-Efficient Fine-Tuning (PEFT) paradigm,…

An Intrusion Detection System (IDS) is vital in cybersecurity, detecting unauthorized activity across networks. With attacks on network layers increasing, stronger IDSs are needed. Yet most IDSs rely on centralized detection, forcing IoT…

Pseudorandom codes (PRCs), introduced by Christ and Gunn (CRYPTO '2024), are error-correcting codes whose codewords are computationally indistinguishable from uniformly random strings, while still being decodable by someone holding the key.…

University cybersecurity protocols have intensified dramatically in response to rising threats of data breaches, ransomware, and credential theft. While necessary, these measures have created a parallel crisis of accessibility - even for…

Recent advances in Vision-Language Models (VLMs) facilitate a new class of embodied AI systems, where these models are integrated into physical platforms, e.g. robots and autonomous vehicles, to interpret visual scenes and execute natural…

Large Language Model (LLM) alignment remains vulnerable to jailbreak attacks that elicit unsafe responses, motivating pre-model and post-model guards. Pre-model guards audit the safety of prompts before invoking target models. However,…

As part of the political process, citizens may participate in signature collections to influence policy changes. In Switzerland, this even results in legally binding acts, similar to an election system. In this work, we first derive a…

Over-the-air federated learning (OTA-FL) improves communication efficiency by exploiting the superposition property of wireless channels, but this same property also creates a critical security vulnerability: the parameter server (PS)…

Unmanned aerial vehicles (UAVs) are cyber-physical systems whose attack surface spans networked avionics and on-board sensor fusion: a compromised GPS or battery module can mimic a benign mission segment and evade naive anomaly detectors.…

Fake Base Station (FBS) detection has been a critical focus of cellular security research for over two decades. However, significant financial and regulatory barriers to accessing commercial FBS (C-FBS) devices have limited direct…

Unified autoregressive models (UAMs) are transformer models that generate text as well as image tokens within a single autoregressive pass. Shared parameters and a multimodal vocabulary simplify the training pipeline and facilitate flexible…

Prompt injection attacks pose significant risks to language model safety, yet existing defenses are typically evaluated using classification performance. We show that high detection performance does not imply representational robustness.…

Large language models (LLMs) are highly susceptible to backdoor attacks (BAs), wherein training samples are poisoned using trigger-based harmful content. Furthermore, existing defenses have proven ineffective when extensively tested across…

Cryptographic primitives such as stream ciphers,Pseudorandom Number Generators (PRNGs), and block cipher modes produce sequences that are designed to be statistically indistinguishable from random data. As a result, the traditional…