# adaptive-attacks
標記為「adaptive-attacks」的 3 篇文章
攻擊者後手問題
為何靜態 LLM 防禦在適應性對手前失敗:12 項遭繞過防禦的分析及對防禦設計的意涵。
defenseadaptive-attacksred-teamingresearchadversarial-robustness
Reasoning 模型 越獄s
How reasoning capabilities create novel jailbreak surfaces: chain-of-thought exploitation, scratchpad attacks, and why higher reasoning effort increases attack success.
reasoningjailbreakchain-of-thoughto1o3adaptive-attacksresearch
針對安全訓練的適應性攻擊
針對最新安全訓練技術的適應性攻擊研究,包括規避與相應對策。
frontier-researchadaptive-attackssafety-trainingresearch