研究人员开发了新的方法来解决大语言模型（LLMs）中的漏洞。一种名为“Persona Attack”的方法利用对话记忆绕过安全协议，在某些配置下成功率高达95%。作为回应，引入了一个名为THRD的框架，该框架通过分析时间风险累积，使用一种无需训练的方法来检测和缓解多轮越狱攻击，将攻击成功率降低至0.2%，同时对模型效用的影响最小。此外，一项研究对LLMs进行了密码分析基准测试，揭示了它们在安全环境中的潜力和局限性，并引发了对其易受某些攻击的担忧。 AI
影响 新研究突显了大语言模型不断演变的漏洞以及新型防御机制的开发，这对于维护人工智能的安全至关重要。
排序理由 多篇研究论文详细介绍了新的大语言模型漏洞和防御措施。
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
arXiv:2606.00150v1 Announce Type: cross Abstract: As Large Language Models evolve for user convenience, vulnerability to jailbreak attacks continues to be reported despite ongoing efforts in safety training. Traditional jailbreak techniques typically focus on a single prompt inje…
arXiv:2606.01738v1 Announce Type: cross Abstract: Multi-turn jailbreak attacks pose a growing threat to LLMs by exploiting conversational dynamics such as gradual escalation and cross-turn coordination. Existing defenses either rely on costly retraining -- often degrading model u…
arXiv:2505.24621v3 Announce Type: replace Abstract: Recent advancements in large language models (LLMs) have transformed natural language understanding and generation, leading to extensive benchmarking across diverse tasks. However, cryptanalysis - a critical area for data securi…