https://guijiagi.com/tags/%E5%AF%B9%E6%8A%97%E6%94%BB%E5%87%BB/Recent content in 对抗攻击 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/article-14/Sun, 12 Jul 2026 19:10:00 +0800https://guijiagi.com/posts/article-14/从对抗攻击到对齐安全，全面预测2026年下半年AI安全领域的关键趋势和新兴威胁https://guijiagi.com/posts/b2-5e1b07ab/Sun, 12 Jul 2026 19:10:00 +0800https://guijiagi.com/posts/b2-5e1b07ab/从对抗攻击到对齐安全，全面预测2026年下半年AI安全领域的关键趋势和新兴威胁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/ai-data-poisoning-2026/Thu, 02 Jul 2026 10:23:00 +0800https://guijiagi.com/posts/ai-data-poisoning-2026/深入探讨2026年AI数据投毒攻击的最新手法、检测方法与防御策略https://guijiagi.com/posts/llm-jailbreak-landscape-dan-to-multi-turn/Tue, 30 Jun 2026 10:10:00 +0800https://guijiagi.com/posts/llm-jailbreak-landscape-dan-to-multi-turn/全面梳理2026年大模型越狱技术的演进路径，从早期DAN越狱到多轮诱导、多模态攻击，分析攻防博弈的深层逻辑https://guijiagi.com/posts/data-poisoning-attacks/Sun, 28 Jun 2026 11:25:00 +0800https://guijiagi.com/posts/data-poisoning-attacks/深入解析LLM数据投毒攻击：从投毒手法到检测防御的完整方案https://guijiagi.com/posts/red-teaming-llm/Wed, 24 Jun 2026 14:00:00 +0800https://guijiagi.com/posts/red-teaming-llm/系统性介绍 LLM 红队测试方法论，涵盖攻击向量分析、自动化工具链、修复流程与真实案例