[问与答] 大型语言模型（LLM）的安全问题，是工程问题，是算法问题，还是一个根本性的“哲学”问题？
"安全防御" 相关文章
AI武装的黑产背后，这群人正在打一场看不见的战争。
Broken-Token: Filtering Obfuscated Prompts by Counting Characters-Per-Token
双管齐下：联邦学习防投毒攻击与梯度泄露，华南理工深北莫研究成果登上TMC与IoT
SAID: Empowering Large Language Models with Self-Activating Internal Defense
ICTFICIAL OY | 提高蜜罐性能的网络欺骗技术综合调查
CrossGuard: Safeguarding MLLMs against Joint-Modal Implicit Malicious Attacks
VisuoAlign: Safety Alignment of LVLMs with Multimodal Tree Search
OpenAI、Anthropic、DeepMind联手发文：现有LLM安全防御不堪一击
GPS Spoofing Attack Detection in Autonomous Vehicles Using Adaptive DBSCAN
Backdoor Vectors: a Task Arithmetic View on Backdoor Attacks and Defenses
赠书福利 | 《域渗透实战指南》免费送
SoK: Systematic analysis of adversarial threats against deep learning approaches for autonomous anomaly detection systems in SDN-IoT networks
SafeBehavior: Simulating Human-Like Multistage Reasoning to Mitigate Jailbreak Attacks in Large Language Models
Boundary on the Table: Efficient Black-Box Decision-Based Attacks for Structured Data
Bidirectional Intention Inference Enhances LLMs' Defense Against Multi-Turn Jailbreak Attacks
Ransomware has evolved – so must our defences
无人机安全事件频发，欧盟拟建“无人机墙”
SilentStriker：无声击溃大模型
Building a Hybrid Rule-Based and Machine Learning Framework to Detect and Defend Against Jailbreak Prompts in LLM Systems
