https://guijiagi.com/tags/llm%E5%AE%89%E5%85%A8/Recent content in LLM安全 on 硅基 AGI · 智能体学习与测评Hugozh-cn本站内容采用 CC BY-NC-SA 4.0 国际许可协议授权Mon, 13 Jul 2026 05:50:00 +0800https://guijiagi.com/posts/article-78/Mon, 13 Jul 2026 05:50:00 +0800https://guijiagi.com/posts/article-78/系统性解析Prompt注入攻击的原理、分类与防御策略，附实战代码示例https://guijiagi.com/posts/adversarial-attacks-llm/Thu, 02 Jul 2026 10:32:00 +0800https://guijiagi.com/posts/adversarial-attacks-llm/深入探讨大语言模型面临的对抗攻击手法与鲁棒性提升策略https://guijiagi.com/posts/hallucination-root-cause-2026/Tue, 30 Jun 2026 16:30:00 +0800https://guijiagi.com/posts/hallucination-root-cause-2026/深入分析大模型幻觉产生的根本原因，对比2026年主流幻觉检测与缓解技术的原理、效果与适用场景https://guijiagi.com/posts/llm-security-audit-vulnerability-scanning/Tue, 30 Jun 2026 11:35:00 +0800https://guijiagi.com/posts/llm-security-audit-vulnerability-scanning/系统介绍大模型安全审计的完整方法论，包括漏洞分类、扫描工具、渗透测试流程和修复建议，提供可操作的审计指南https://guijiagi.com/posts/llm-red-team-jailbreak-20-methods/Sun, 28 Jun 2026 11:00:00 +0800https://guijiagi.com/posts/llm-red-team-jailbreak-20-methods/20种LLM越狱攻击手法实战解析及对应的防御策略，附红队测试框架https://guijiagi.com/posts/llm-safety-guardrails-design/Sun, 28 Jun 2026 10:55:00 +0800https://guijiagi.com/posts/llm-safety-guardrails-design/LLM安全护栏的完整设计：输入过滤、输出审查、实时监控的工程实现https://guijiagi.com/posts/prompt-injection-attacks-2026/Sun, 28 Jun 2026 10:20:00 +0800https://guijiagi.com/posts/prompt-injection-attacks-2026/2026年Prompt注入攻击全景：从直接注入到间接注入、多模态注入的最新攻击手法及防御方案