首个衡量LLM驱动代码编辑安全漂移的基准，揭示AI改码时隐藏的安全退化风险。
arXiv:2608.15092v1 Announce Type: cross Abstract: In this work, we introduce WeSCE, a benchmark for quantifying security drift in code editing under w…
大模型生成代码安全研究新基准，揭示领域提示引发的漏洞风险变化
arXiv:2607.25225v1 Announce Type: cross Abstract: LLMs are increasingly used for code generation in critical infrastructure, yet the security effect o…
AI代理在操作中会产生幻觉并出现安全漂移，这篇论文深入剖析了风险机制与防护策略。
arXiv:2607.18366v1 Announce Type: new Abstract: Large language models (LLMs) serving as planners in tool-using autonomous agents introduce dynamic rel…