最近一年的 AI 安全论文
本站共收录 20 篇- NetInspector：测量并提升 LLM 可靠生成意图网络策略的能力
NetInspector: Measuring and Improving LLM Capabilities for Reliable Intent-Based Networking Policy Generation
- 打破规划器完整性边界：针对 LLM 驱动具身智能体的环境状态文本注入攻击
Breaking Planner Integrity Boundary: Enviroment State-Text Injection Attack on LLM-Driven Embodied Agents
- MMAligner：通过表示校准保护多模态大语言模型
MMAligner: Safeguarding Multimodal Large Language Models through Representation Calibration
- LLM 智能体安全双重性：自我安全与赋能网络安全综合综述
LLM agents security duality: a comprehensive survey of self-security and empowered cybersecurity
- SoK：AI 安全代码生成的进展、陷阱与前进路径
SoK: AI Secure Code Generation: Progress, Pitfalls, and Paths Forward
- AgentSentry：利用时序因果诊断与上下文净化缓解 LLM 智能体的间接提示注入
AgentSentry: Mitigating Indirect Prompt Injection in LLM Agents via Temporal Causal Diagnostics and Context Purification
- DualGauge：联合评测 LLM 与编码智能体仅依据规格生成代码的安全性与功能性
DualGauge: Automated Joint Security-Functionality Benchmarking of Specification-Only Code Generation by LLMs and Coding Agents
署名以 dblp 和本人的 Google Scholar 为准；只列本站收录的 AI 安全论文。