arXiv:2509. 16749v1 Announce Type: cross Abstract: LLMs are increasingly pervasive in the security environment, with limited measures of their effectiveness, which limits trust and usefulness to security practitioners.
arXiv:2508. 18684v2 Announce Type: replace-cross Abstract: Signature-based Intrusion Detection Systems (IDS) detect malicious activity by matching network or host events against predefined rules.
arXiv:2606. 05844v1 Announce Type: cross Abstract: Rule-based Intrusion Detection and Prevention Systems (IDPS) offer precise attack detection as well as mitigation, however their manually crafted, signature-driven rules limit adaptability to emerging and zero-day threats.
The paper introduces AUTOSIGMA, an automated system that converts unstructured cyber threat intelligence reports into Sigma detection rules. It enriches input data with a structured knowledge base, matches it against existing Sigma rule repositories, and uses a large language model as a judge to validate the generated rules. Experiments on real-world APT reports and security blogs show that AUTOSIGMA outperforms other methods in rule validity, relevancy, MITRE ATT&CK coverage, and robustness to input quality.
The paper examines how the scores of cybersecurity large language model (LLM) benchmarks vary depending on the evaluation pipeline used. By auditing eight benchmarks across ten different LLMs, the authors uncover 15 systematic failure modes and demonstrate that a single pipeline choice can shift a model’s score by over 80 percentage points, significantly altering rankings. They also show that even semantically similar tasks can produce different model rankings due to incompatible evaluation conventions, and that standardizing pipelines can move most models by at least three ranks on at least one benchmark.
arXiv:2607. 06963v1 Announce Type: cross Abstract: Large Language Models (LLMs) and generative AI (GenAI) systems, such as ChatGPT, Claude, Gemini, LLaMA, Copilot, Stable Diffusion by OpenAI, Anthropic, Google, Meta, Microsoft, Stability AI, respectively, are revolutionizing cybersecurity, enabling both automated defense and sophisticated attacks.
arXiv:2606. 30819v1 Announce Type: cross Abstract: Generative AI has emerged as a significant cybersecurity threat, with several recent attack campaigns leveraging LLMs to generate code for malicious purposes via scripting languages such as PowerShell.
arXiv:2510. 15476v3 Announce Type: replace-cross Abstract: Large Language Models (LLMs) are increasingly used as interfaces to information, code, and real-world services, making prompt-level security failures a practical concern.
arXiv:2606. 04460v1 Announce Type: cross Abstract: AI has the potential to transform cybersecurity by enabling systems that can autonomously detect, analyze, and remediate software vulnerabilities.
Large Language Models (LLMs) and generative AI (GenAI) systems, such as ChatGPT, Claude, Gemini, LLaMA, Copilot, Stable Diffusion by OpenAI, Anthropic, Google, Meta, Microsoft, Stability AI, respectively, are revolutionizing cybersecurity, enabling both automated defense and sophisticated attacks. These technologies power real-time threat detection, phishing defense, secure code generation, and vulnerability exploitation at unprecedented scales.
CS-Guard is a new benchmark that systematically evaluates guardrails for code generation security, covering 1,000 malware-generation prompts, 7 jailbreak attacks, and a novel fictional scenario attack (FSA) for text-to-code generation, as well as 331 code prompts for code-to-code generation. The study empirically tests nine guardrails across seven large language models, finding that many guardrails fail to prevent malicious code generation, with attack success rates reaching about 50% for text-to-code and up to nearly 100% for code-to-code and FSA scenarios. CS-Guard introduces a modular three-layer guardrail taxonomy and releases its benchmark and data to support future research.
arXiv:2606. 15123v1 Announce Type: cross Abstract: We study the task of CVE-conditioned exploit generation, where a model drafts proof-of-concept (PoC) exploits given software vulnerability context.
arXiv:2606. 05252v1 Announce Type: cross Abstract: Security teams routinely simulate attacks against their own systems to check whether their monitoring would catch a real intruder.